Commodore: A Company On The Edge t1c1o

VARIANT PRESS www.variantpress.com Copyright © 2011 by Brian Bagnall All rights reserved, including the right of reproduction in whole or in part in any form. Photographs courtesy of Raymond J. Feagans, Neil Harris, Yash Terakura and Dave Haynie. Library and Archives Canada Cataloguing in Publication Bagnall, Brian, 1972Commodore : a company on the edge / Brian Bagnall. -- 2nd ed. Previously published as: On the edge. Includes bibliographical references and index. ISBN 978-0-9738649-6-0 (hardcover) 1. Commodore International--History. 2. Computer industry--Canada--History. 3. Computer industry—United States--History. I. Title. HD9696.2.A2B33 2010 338.7’610040971 C2010-904107-0

Introduction

“Do you ? I do.” - Bouncing Souls, ’87 Commodore Business Machines ended operations on April 29, 1994. Enough time has ed to examine what Commodore meant to the world. Amid the chaos, infighting and excitement, Commodore was able to achieve some remarkable industry firsts. It was the first major company to show a personal computer, before even Apple and Radio Shack. It sold a million computers before anyone else. It released the first true multimedia computer. Yet with all these firsts, Commodore receives almost no credit as a pioneer. The history of early computers has tended to focus on Microsoft, IBM, and Apple, snubbing contributions made by Commodore. “There is a lot of revisionism going on and I don’t think it’s fair,” says Commodore 64 designer Robert Yannes. “People wanted to ignore Commodore.” An early-popularized story of the microcomputer revolution was Accidental Empires, by Robert X. Cringely (born Mark Stephens). The former Apple employee perpetuated a select view of the microcomputer revolution—a view that not everyone accepts as accurate. In Infinite Loop, Michael Malone writes, “The pseudonymous Cringely is notorious for his sloppy way with facts.” In his book, Cringely said, “Commodore wasn’t changing the world; it was just trying to escape from the falling profit margins of the calculator market while running a stock scam along the way.” In reality, Commodore employees worked tirelessly to deliver state-of-the-art technology to its customers at prices far lower than Apple’s. PBS adapted Cringley’s book as a popular TV series, Triumph of the Nerds (1996). The adaptation ignored Commodore completely.

Turner Network Television produced a movie called Pirates of Silicon Valley (1999), based on a more credible book, Fire in the Valley, by Paul Freiberger & Michael Swaine. Regrettably, the producers ignored much of the book and focused on Steve Jobs, Bill Gates, and IBM. “The PC came out, we changed players, and the whole early history just got lost,” says PET designer Chuck Peddle. Peddle deplores the emphasis on IBM, Apple, and Microsoft at the expense of earlier developers. “None of that is true. It’s not fair that the stuff that happened earlier has been so badly ignored.” “I’m not sure it’s intentional, it’s just the west coast mindset that everything happens on the west coast, so we don’t even need to pay attention to what happened everywhere else,” says Yannes. “Most of the revisionist stuff I read was coming out of California and Commodore was mostly successful after it left California.” When writers are not ignoring Commodore, they often get their facts wrong. In The Silicon Boys and their Valley of Dreams, David Kaplan describes the Apple IPO in 1980 and then adds, “But Apple soon bred competition. Radio Shack and Commodore and even Atari, among others, started selling their own personal computers.” In truth, Commodore and Radio Shack began selling personal computers in 1977, and Atari followed in 1979. This rosy picture of Apple starting the microcomputer industry crumbles under inspection. Apple had a very slow start and eventually climbed to first place sometime in the early 1980s, only to lose its lead to Commodore once again. In the very earliest days, Commodore was pioneering the consumer microcomputer industry. While IBM pushed business computers and Apple pushed style, Commodore put computers into the hands of ordinary consumers. Throughout the eighties, Commodore consistently had the best prices, often with the best technology. The Commodore 64 is the Model T of computers, selling more units than any other single computer model, according to the Guinness Book of World Records. In the summer of 2004, I began interviewing Commodore insiders. We traveled back to the seventies, eighties, and nineties and relived the Commodore experience. I am thankful to each of the participants for taking me on that journey, something I will never forget.

The journey has ended for me, but for you it is about to begin. I hope you enjoy reading this book as much as I enjoyed writing it. Brian Bagnall May 22, 2005 (updated October 16, 2010)

Prologue:

The Rise of Commodore

Hailing a taxi in New York City in the early 1950s might have put you in the company of future business titan Jack Tramiel. As you sat in the back seat, two large, bulging eyes would appraise you through the rearview mirror, determining if you were worth anything to him. At the time, Tramiel was positioning himself for riches and glory. It was a humble beginning, but driving a taxi was a blissful step up from the work camps in Poland during World War II. In July 1947, a 19-year-old Idek Tramielski proposed to and married a fellow concentration camp survivor named Helen Goldgrub in . While there, the Hebrew Immigrant Aid Society ed Idek and helped him and his wife emigrate from Europe by paying for their ocean liner tickets to New York City. Idek changed his name to Jack Tramiel. In 1948, Tramiel enlisted in the US Army and served as a cook at Fort Dix. Later, he ed the First Army Office Equipment Repair Department, which was responsible for maintaining and repairing almost 25-thousand pieces of office equipment. Tramiel served two tours of duty in Korea in 1950 and in 1952. After his service ended, Tramiel worked for a typewriter repair company called Ace Typewriter. There, he met a fellow technician named Manfred Kapp, and the two started a repair company called Singer Typewriter. In 1958, Tramiel, Kapp, and their families moved to Toronto, Canada and formed a typewriter manufacturing company called Commodore. “People sometimes got a little bit frightened of him because he was moving and talking a little bit too fast,” says Kapp. The company quickly grew until a scandal rocked the Canadian financial scene, with Commodore at the center. After an embarrassing public inquiry, Commodore was finished—or so it seemed.

In 1966, a Canadian Investor named Irving Gould purchased Commodore and redirected Tramiel into the burgeoning calculator business. By 1973, savage competition from Texas Instruments and Japanese calculator manufacturers began hurting Commodore’s profits. Tramiel began to look elsewhere for cheaper calculator parts.

CHAPTER 1

Microprocessors 1973-1974

High-tech companies need three players in order to succeed: a financier, a technology god, and a juggernaut with a Type A personality. Commodore would require these three ingredients to take them to a new level. They had Irving Gould, with his financial expertise and deep pockets. They had Jack Tramiel, so aggressive people sometimes referred to him as the scariest man alive. All Commodore needed was a visionary engineer to take Commodore into a new field of technology. In the 1970s, the image of a computer genius was not in the mold of the young hacker. Teenaged tycoons like Bill Gates had not yet filtered into the public consciousness. Instead, the accepted image of a technological genius was a middle-aged man with graying hair and glasses, preferably wearing a long white lab coat and working at Hewlett-Packard or 3M. Chuck Peddle was the image of a technology wizard, with his wire-framed glasses, receding hairline, and slightly crooked teeth. At two hundred and fifty pounds, the five foot eleven inch engineer always struggled with his weight. He describes himself at that time as “totally out of shape,” but he was characteristically optimistic and never without a joke or story to tell. Peddle possessed the ability to see further into the future than most of his contemporaries and he obsessively searched for the next big innovation. His mind was always active, sometimes to the point of causing sleep deprivation. “I don’t sleep much,” he says. “Never did.” In fact, the pattern of sleeplessness went back to his earliest days.

* * * Peddle’s last name had a peculiar origin. “My grandfather was an immigrant into Newfoundland from down near Poole in Bournemouth, England,” he says. “The history of the name is very funny because there is a river just west of Bournemouth, the River Piddle. My family is apparently from there. My grandfather’s family had taken the name from the river, so the family name was Piddle.” Unfortunately, the name had an embarrassing meaning. “He came to Newfoundland, Canada and discovered Piddle meant ‘a little pee.’ The name Piddle was an easy transition to Peddle. All the Peddles in the United States started from my grandfather.” Peddle’s grandfather was a clipper ship captain, operating mainly out of Newfoundland. “He came to the United States some of the time because he jumped back and forth to his clipper ship. My father was born in the United States.” His father was one of 21 kids. A US citizen, he lived in Newfoundland for his early life before returning to the country of his origin. “My father came to the United States to live with his brother,” says Peddle. “His brother threw him out when he was 17.” Peddle’s father worked as a butcher, then as a farm machinery salesman, and finally a real estate broker. The poor region made it difficult to a family. “The whole area was very depressed,” he says. “My father was very poor.” Charles Ingerham Peddle was born in Bangor, Maine in 1937, one of eight children. “There were two families,” he says. “The first three children were by another mother. I was the oldest of the second family, and my mother’s first child. The other ones were quite a bit older. They were old enough to take care of us.” Peddle was a restless child. “My mother said that when I was young I used to lie awake in my crib. I would cry and fuss and didn’t sleep as much as the other kids.” His parents raised him in Augusta, the state capital of Maine, which had a population of just over 20,000. In high school, he worked as a theatre usher. In his senior year of high school, Peddle thought he found his calling. “In high

school I worked at a radio station. I wasn’t getting paid. I was doing it as a way of getting trained.” Radio announcing seemed to offer a glamorous life for Peddle. “I really wanted to be a radio announcer. Now, that really doesn’t mean very much, but back then was pre-TV and radio announcers were big.” Nearing the end of high school, he traveled to Boston to try out for a scholarship in broadcasting. While there, he realized he did not have enough natural talent. Returning to Augusta, he talked things over with the radio station owner, who told him, “I’ll employ you as a radio announcer, but you will always be stuck in Maine because you are not good enough.” Peddle walked away. He spent some time in the military as he contemplated his future. “I went into the Marine Corps just before I got out of high school in 1955,” he says. “The Marine Corps sent me to California.” Peddle fell in love with “the Golden State.” During this time, his former science teacher, who had recognized a gift in his student, encouraged him to enter engineering. Peddle listened to his advice, but was unsure he wanted to enter the sciences. In the meantime, he worked on a road crew sculpting the hillsides around roads. “I didn’t want a pick and shovel job. I wasn’t sure what I was going to do and I was dirt poor. Luckily, in Maine you can be dirt poor and still get by.” Unable to earn enough to pay for tuition fees, he applied for student loans. At the end of summer, Peddle entered the University of Maine and enrolled in engineering and business courses. Partway through the first year, the university required students to choose a discipline. Unfortunately, he still had no idea what he wanted to do. “I really loved physics, so I took engineering physics with an electrical minor.” At the time, computing was in a dismal state at Maine. “There wasn’t a computer on campus, nor was there anyone on the campus who was computer literate.” In his junior year, things began to change. “On the entire campus, there was one analogue computer, which had been bought in the last four months,” he recalls. “The analogue computer was so primitive and they didn’t know how to use it.” During his first academic year, he received standard engineering training, devoid

of computers. Over 200 miles away, at the Massachusetts Institute of Technology (MIT), a revolution was occurring which would soon change his situation. * * * Chuck Peddle’s main influence was the legendary inventor and mathematician, Claude Elwood Shannon. Though virtually unknown to the world, Shannon was the founding father of the modern electronic communications age. He was an eccentric, who terrified students by riding his unicycle at night through the hallways while juggling. He also built a reputation for inventions that were of little practical value to anyone. Over the years, he filled his beachside house with juggling robots, maze-solving robot mice, chess-playing programs, mind-reading machines, and an electric chair to transport his children down to the lake. In 1948, while working at Bell Labs, Shannon produced a groundbreaking paper, A Mathematical Theory of Communication. In it, he rigorously analyzed the concept of Information Theory and how pictures, words, sounds and other media are transmitted using a stream of ones and zeros. He even coined the word “bit.” Peddle was enchanted with his theories. “Today, you take this for granted, but you have to that someone had to dream all this up,” he says. “Everyone else’s work stands on his shoulders and most people don’t even know it.” In 1956, Shannon returned to MIT at Lincoln Labs as a lecturer and Artificial Intelligence researcher. While there, he spread his concepts on Information Theory. “He changed the world,” says Peddle. “Shannon was not only a pioneer, but a prophet. He effectively developed a following, almost like a cult.” One of Shannon’s cultists would soon spread the word to Peddle at the University of Maine. During Peddle’s senior year, the University of Maine accepted a lecturer from MIT who studied under Claude Shannon. According to Peddle, “He had a nervous breakdown, so he left MIT. The University of Maine was so happy to get him because he was so superior to the type of instructor they could normally get. They gave him the opportunity to teach only four classes per week between the hours of eleven o’clock and noon. The guy was being totally babied and should have been since he was a great instructor. He decided to put together a

class to teach people about Information Theory.”

Chuck Peddle, father of the 6502 (Das neue P.M. Computerheft).

At the time, Peddle was enrolling for his junior year. The new Information Theory class happened to fit into his schedule. As Peddle recalls, “It changed my life.” The class began with the instructor discussing the eyes and ears as the primary sensors for receiving information. “He started teaching us about Boolean algebra and binary logic, and the concept of Information Theory,” recalls Peddle. “I just fell in love. This was where I was going to spend my life.” However, the topic that interested Peddle the most was computers. “Information Theory was interesting, and I’ve used it from time to time, but the computer stuff this guy taught me was life changing.” Peddle immersed himself in computer theory. “I got an A on my senior paper in physics class by giving a discussion on binary and Boolean arithmetic. I was trying to build an AND gate in my senior class [using early transistors] and the top electrical engineers on campus couldn’t help me figure out the structures and why my AND gate didn’t work.” Peddle and a friend even tried growing a transistor crystal but soon gave up. As graduation approached, Peddle began searching for permanent employment. He had married while in college and already had a family. “I came out of college and I had three kids; two and a half, actually. I had the third one right after [graduation].” The new responsibilities motivated him to find a better life. Peddle knew he wanted to live in California and he wanted to work in computers. “At all of the companies of any size, like GE (General Electric) and RCA (Radio Corporation of America), you went to work on a training program for a year or two. You really were just interviewing to their training program.” GE made the best impression on Peddle. “I kind of fell in love with GE,” he says. “When I got my offer, I thought I would take it, because they had such a

good training program.” * * * Peddle and his young family moved to California to start a new life with GE. Before long, he was working at GE’s computing facility in Phoenix, Arizona. He worked with massive mainframe computers, similar to those seen in the 1965 film Alphaville. The first computer he used was an NCR 304, which he describes as a “very old, very slow machine with small capacity.” At GE, his group was deg computers, using computers. He entered programs into the NCR 304 computer by feeding a stack of punch cards into a card reader. “I would set up long six- or seven-hour runs, drive across the city and go to bed, with the instructions, ‘If this breaks, call me.’ People would wake me up in the middle of the night, I would find a solution in ten minutes and go back to sleep.” In 1961, Peddle and two of his coworkers developed the concept for variable sector disk formatting. They even filed a patent for their idea. Years later, Peddle would use this idea to give Commodore disk drives more data storage than the competition. In 1963, John G. Kemeny developed the BASIC computer language at Dartmouth College in New Hampshire, along with Tom Kurtz. They developed BASIC (Beginner’s All-purpose Symbolic Instruction Code) for the GE-235 mainframe computer, and as a result, Peddle was almost immediately aware of it. “I taught BASIC the day after it was invented,” he claims. “I got one of the original BASIC manuals from a guy in Dartmouth and taught my people in Phoenix.” A year later, Kemeny and Kurtz created the revolutionary Dartmouth TimeSharing System (DTSS) for the GE-235. With the time-sharing system, multiple s could interact with the mainframe computer simultaneously using remote terminals. General Electric immediately recognized the value of this new system and used it to form the basis of a new multi-million dollar business. “Two years later, GE goes into the time-sharing business,” recalls Peddle. “They’re selling time-sharing to everybody. It was a big goddamned deal.”

NCR 304 Mainframe Computer System.

With the time-sharing business suddenly ballooning, GE sent Peddle to its largest computing center in Evendale, Ohio to set up time-sharing systems for General Electric Aircraft Engines (GEAE), which designed and manufactured jet engines. The massive computer facility contained ten IBM-7094 mainframe systems, five GE-600s, and twenty-five GE-225s. “We were running timesharing for about 4,000 engineers and programmers.” The refrigerated computing facility seemed futuristic in the mid-sixties, with white tiled walls, raised floors, and row upon row of mainframe computers. Setting up the time-sharing systems was laborious, and Peddle often stayed at the computer facility around the clock. During this time, he picked up a habit originated by GE founder Thomas Edison. “I stole the idea of cots from him. Everyone understood that if I was tired, I would go to my office and take a halfhour nap.” The experience gave him valuable knowledge that he would later use to develop his own computers. “I got a really good understanding of what worked on timesharing and what didn’t work, and what people wanted.” While working at GE, Chuck met John Paivinen, who would later become involved with him at Commodore. “John Paivinen was my manager at GE. He’s the guy who put GE in the computer business.” * * * The time-sharing business Peddle helped develop at GE was phenomenally successful, but in the late ’60s, it started failing due to increased competition. By this time, Peddle had risen to a high-level management position. Suddenly, “Time-sharing crashed. Out of business. Goodbye,” he says. “Companies started figuring out how much money they were spending on these time-sharing services and it was millions. GE was just cleaning up, but it just wasn’t cost effective the way it was being done, so companies kept cutting it off and they moved the computers internally.” Peddle’s interest shifted to improving cash s. At the time, shared

computing kept the brains of the computer at one central location and people could only interact with the computer system using terminals, which consisted of a keyboard and monitor. Peddle envisioned distributed intelligence, where he would transform the dumbterminal into an intelligent-terminal that could have a printer connected to it, or other peripherals and data entry devices. “I sat down and derived the principles of distributed intelligence during a four-month period,” he says. “There was a focus on five or six stations around a minicomputer in a centralized architecture. My concept was you moved the intelligence to the place where you used it.” “Then I started trying to teach GE about it.” Unfortunately, in 1970 GE decided they were no longer interested in computers. “I was getting nowhere with GE because they were getting ready to sell the computer business. Two months later, they sold the company to Honeywell.” He had the option to receive a severance package or move elsewhere in GE. The decision was easy for Peddle and two colleagues, who each took a severance package. “We said, ‘This is found money, so we’re going to start our own business.’ We had already started on the cash business, and I had a deal with Exxon,” he says, referring to the oil company. Peddle left for Phoenix. The three partners immersed themselves in their intelligent-terminals. During this time, Peddle devised several inventions that would have made him wealthy had he chosen to patent them. “We invented the credit-card-driven-gasoline pump, the first credit verification terminal [credit card reader] and the first point-of-sale terminal [electronic cash ].” He now laments, “It’s too bad we didn’t patent the shit out of it because we could have been very wealthy as a result of that.” During development, he realized the distributed-intelligence terminal needed a fundamentally new component to make their ideas work. “We needed our own microprocessor.” This realization would lead Peddle on an extraordinary journey that would change millions of lives. At first, he tried to develop the technology within his fledgling company but it was hopeless without funding. “We had everything going for us, but we didn’t know how to raise money.” It was time for Peddle and his team to move on. * * *

Chuck Peddle and his wife now had four children, but the stresses of Peddle leaving his secure job at GE caused the marriage to disintegrate. They divorced in 1971. “I put a bag of clothes in my [Austin-Healey] Sprite and drove away,” he says. Within weeks, in what Peddle a “planned transition,” Peddle married his ex-partner’s ex-wife, Shirley, a voluptuous blonde with two children. “I took some time out, because there was a change in life—going through the divorce and all that.” In 1972, Peddle tried to start a word processing company using Digital Equipment Corporation (DEC) time-sharing systems. “We actually did the first on-line text processing system, setting type for newspapers,” he says. Once again, Peddle’s idea was too far ahead of the times, given the primitive capabilities of computers. “That company didn’t succeed either.” The experience gave Peddle valuable knowledge he would need to develop the next generation of microprocessors. “I had done all the microelectronics and knew why a microprocessor needed to happen, and how to make a microprocessor, and how to make things that used microprocessors,” he says. “But I didn’t have a microprocessor because they weren’t around yet.” In 1973, a recruiter approached Peddle about collaborating on Motorola’s new microprocessor program in Mesa, Arizona. “I went down and talked to the guy who was running the program, who was a calculator guy,” says Peddle, referring to Tom Bennett, the father of the 6800 microprocessor. Peddle’s experience with cash s won him the job. “He asked me to design the system I needed for my cash . He basically hired me to finish the program.” The microprocessor program at Motorola was already a year and a half in development when Peddle began. He started work at Motorola in 1973, around the time when Large Scale Integration (LSI) of semiconductor technology allowed the circuitry of a calculator or computer to fit onto a single chip. As the Intel 4004 and 8008 processors were gaining popularity, Motorola decided to enter the microprocessor market. Tom Bennett created the original architecture for the 6800. “They kind of muddled their way through the architecture for the 6800, which had some flaws in it,” recalls Peddle. Bennett essentially duplicated the instruction set of the DEC PDP-8

minicomputer on the chip, effectively creating a microprocessor rather than a microcontroller. “I was able to fix some of those flaws but it was too late for others,” says Peddle. The final 8-bit microprocessor had 40 pins, 4,000 transistors and an instruction set of 107 operations. Peddle also made a major contribution to the project by instigating several chips for the 6800. The 6800 microprocessor had to interact with hardware, so the engineers designed specialized chips for this purpose. One chip to emerge was the 6820 Peripheral Interface Adapter, designed by a rookie engineer named Bill Mensch. The engineers called it the PIA chip. The 6820 became a major reason for the eventual popularity of the 6800. Although Motorola engineers grasped the importance of what they had created, the management and salespeople knew very little of microprocessors. According to Peddle, some managers at Motorola even tried to kill the project. “So I built a demo of the chip using some of the hardware for my cash to show everybody that microprocessors really did work.” The salespeople at Motorola required an education on microprocessors but there were no courses. “They didn’t know how to sell it, so I put together a training class for their applications engineers,” says Peddle. Peddle was also instrumental in making some of the first deals for Motorola, including Hewlett-Packard, Tektronics, NCR (National Cash company), Ford Motor Company, Unisys, and Burroughs (makers of calculators). “I wound up going into the field presenting the architecture because I was the only one in the company who could intelligently talk to customers and have architectural discussions.” The presentations usually ended the same way. “The guys would sit down, we would explain the 6800, and they would just fucking fall in love,” says Peddle. However, the $300 price tag for a single 6800 processor prevented engineers from adopting the 6800 microprocessor in low-cost products. According to Peddle, someone would invariably say, “You’re charging too much for it. What I want to use it for is not to replace a minicomputer. I want to use it to replace a controller, but at $300 per device it’s not cost effective.” Armed with this knowledge, Peddle had an epiphany. He recognized the vast market for limited-function, cost-reduced microprocessors—known as

microcontrollers. Both Intel and Motorola were overlooking an important niche. In between business trips, he pushed Motorola for a microcontroller, but each time he met with resistance. “That’s one of the reasons Chuck left,” says Peddle’s future employee, Bill Seiler. “He was frustrated with their hardheadedness, and not looking down the road to what they had to do and the evolution of their processor.” Peddle began to look for ways to make the microcontroller happen without Motorola. “I returned to Motorola after one of these trips, and I had a letter there, formally instructing me Motorola was not going to follow a cost-reduced product. I was ordered to stop working on it.” Undeterred, Peddle wrote a letter saying, “This is product abandonment, therefore I am going to pursue this idea on my own. You don’t have any rights to it because this letter says you don’t want it.” From that moment on, Peddle stopped working on microprocessors for Motorola. He continued teaching classes, but his true focus was finding a way to make his low-cost microprocessor. * * * While still employed at Motorola, Peddle tried raising money to fund his microprocessor. He visited semiconductor manufacturer Mostek (not to be confused with MOS Technology) and talked to the company chairman, L.J. Sevin[1], but he was not interested. Peddle continued talking to people in the semiconductor business. One day, Peddle ran into a former boss from GE who now worked at Ford Motor Company. He told Peddle about John Paivinen, another ex-GE employee who was now running a semiconductor company in Norristown, Pennsylvania. “When I started looking around for partners, I knew Paivinen was a killer computer guy,” he recalls. “I called him up. He said, ‘Come on down. Let’s talk about it.’” Peddle flew to Pennsylvania to examine the semiconductor manufacturer, MOS Technology. The facility was located at 950 Rittenhouse Road, a 14-acre site in an industrial park, called the Valley Forge Corporate Center. “MOS Technology designed just calculator chips,” says Mensch. “They were the leading independent supplier of calculator chips.” Peddle was impressed with the small

firm. It had good credentials and many customers, among them a calculator company named Commodore. According to Peddle, “Paivinen immediately loved the idea of doing the product.” The two discussed the specifications for the microprocessor, but MOS Technology was only capable of manufacturing chips using the P-channel process. The two engineers agreed they would need the more advanced Nchannel process to achieve the low cost and high speed they required. Paivinen felt he could deliver the N-channel process. “He had taught himself process development when he was working at General Instrument, and was really good at it,” says Peddle. “He considered himself to be a competitor to [Andrew] Grove [of Intel]. He was convinced he could do a five-volt N-channel process in the same amount of time it would take me to develop the microprocessor.” The new partnership seemed to hold promise. Paivinen told Peddle, “Move your people and we’ll set up a second group within the company. You run your own show.” Back in Arizona, Motorola was about to move its 6800 development team. “They were going to move us from Phoenix to Austin, Texas. They just started a new facility down there,” recalls Mensch. “That’s were Motorola’s microprocessor teams were reassembled.” Peddle’s new offer came at an opportune time for the 6800 developers. “We didn’t want to go to Austin, Texas,” explains Mensch. Bill Mensch, who was working on the 6840 Programmable Timer Unit at the time, found it a hard decision to leave. “Motorola was awesome. I can’t say anything negative about them,” he says. With the end of the 6800 project, Motorola wanted to move the engineers to a new project. “We had to pull up stakes anyway, so why not go to Pennsylvania where you can do something new, have some exciting times and who knows?” For Mensch, who grew up in Pennsylvania, it was like going home. Peddle negotiated the contracts for the on his team. “The deal was, if the microprocessor took off, we would have a piece of the company,” recalls Peddle.

Motorola had shown samples of the 6800 to companies since the spring of 1974, but in August 1974, Motorola publicly introduced the 6800 chip for $300. The 6800 would eventually become successful for Motorola, in no small part to the efforts of Peddle. In the same month, Peddle and seven coworkers from the engineering and marketing department left Motorola to pursue their own vision. The team included Will Mathys, Bill Mensch, Rod Orgill, Ray Hirt, Harry Bawcom, and Terry Holdt.[2] The departure of several of Motorola’s top engineers seriously drained the company of much needed expertise on the eve of the 6800 debut. On August 19, 1974, the team started work on their new processor at MOS Technology. With Chuck Peddle and his band of engineers, MOS Technology would radically change the market for computers. [1] L. J. Sevin became a prominent venture capitalist responsible for funding startups like Compaq, Lotus, Cyprus, and Mostek. [2] Terry Holdt later became president of S3, a semiconductor company that supplied a popular all-in-one chipset for IBM PC compatible computers.

CHAPTER 2

MOS Technology 1974-1975

In 1969, a large industrial manufacturing company called Allen-Bradley wanted to enter the new semiconductor business. It financed the creation of MOS Technology. The three men who founded the new startup were Mort Jaffe (Chief Financial Officer), Don McLaughlin (VP of engineering), and John Paivinen (company president). For the first five years, MOS Technology supplied calculator chips and other semiconductor parts to the electronics industry. Then, on August 19, 1974, Chuck Peddle and his team of former Motorola employees began working on a revolution in microprocessor technology. This revolution would occur on the east coast at Valley Forge, Pennsylvania, 20 miles from Philadelphia. It was an appropriate place for a revolution. Almost 200 years earlier, Valley Forge was the turning point in the American Revolution when General George Washington’s tired and bloodied troops retreated to Valley Forge for the winter, only to fight back with an unwavering offensive. Peddle and his band of engineers would also retreat for the winter, emerging in the summer to unleash a powerful new weapon. In the 1970s Valley Forge was a small, dispersed town with a population of about 400 people. MOS Technology headquarters resided in the peaceful setting, along a lone country road surrounded by wildlife. Street names like Adams Avenue, Monroe Boulevard, Madison Avenue, and Jefferson Avenue celebrated the revolutionary past. Directly across the road from MOS was an alluring golf course, General Washington Country Club, tempting the MOS executives to squeeze in a round after work. Less than a mile away was the Audubon Wildlife

Sanctuary, a park filled with serene trails where Canada geese gathered in the fall while migrating south. Horse trails wove through the surrounding countryside. Riders would often emerge from the bushes and stare at this out-ofplace, high-tech firm. They could scarcely understand what was happening inside. The headquarters originated from the 1950s. It was a box-shaped, two-story building with glass windows along the front and sides. Stray golf balls frequently bounced off the windows, occasionally leaving small, bullet-sized holes that no one ever repaired. To the side and rear of the building were two huge parking lots, largely deserted since most people preferred to use the circular driveway out front.

MOS Technology deg and fabricating semiconductor chips.

The engineering lab on the second floor was the fountainhead of ideas for the company. This was where engineers designed their semiconductor chips. Don McLaughlin headed the engineering lab. He subdivided the lab into a maze of smaller rooms, each with a specific task. It was in this environment that Peddle would plan the centerpiece of his revolution. In a delicious irony, he did not design the microprocessor for computers. “It was never intended to be a computer device. Never in a million years,” he reveals. Instead, he envisioned a microcontroller for cash s, home electronics, home appliances, automobiles, and industrial machines—just about everywhere except personal computers. “If we were going to do a computer, we would have done something else.” “The 6502 was never, and I mean never, talked about as being a U (Central Processing Unit) for a computing system,” says Bill Mensch. “The instruction set was picked for controllers. In other words, the controller instruction set was designed specifically to compete with the Intel 4040, which was a two- or threechip-system solution.” The ex-Motorola employees split into three groups, each with its own areas of expertise. “We came in and effectively took over two or three rooms, and operated totally independent of the rest of the company for a long time,” says Peddle. On his first day on the job, Mensch had an unsettling experience. “When I first arrived there I got flat tires on my car. I think they were pissed at me because I was part of the team coming from Motorola. I’ve never had flat tires before or after, and it’s like, ‘What’s going on? There are nails in my tires.’” Mensch was confused about the incident at first, but later came to believe that some of the calculator engineers were initially resentful of the new team of interlopers from Arizona. * * *

Chuck Peddle, Will Mathys, and Rod Orgill would collaborate to design the initial architecture for the new microprocessor. “It was just the perfect product, the perfect time, the perfect team,” says Peddle. The architects’ task was similar to deg a small city, except the streets in this city would be paved with metal. Electrons would inhabit their city, traveling the streets until they reached a set of traffic lights at a transistor. Timing within this little city would be critical, otherwise traffic would halt, causing the chip to lock. Peddle and his group intentionally numbered their chips starting with 6500, so it would sound similar to the Motorola 6800. “It was a cheaper version of the 6800 and we intended to have a whole string of them,” he explains. “In hindsight, with many years and lawsuits behind us now, it was designed to sound like the 6800.” The centerpiece of their project was the 6502 microprocessor. “The 6502 was what we were driving for,” says Peddle. Instead of using the instruction set from the 6800, which was similar to the PDP8, Peddle used a more refined instruction set based on his discussions with infield engineers. The instruction set turned out to be similar to the PDP-11. “The architecture of the 6502 has a wonderful core. It’s a wonderful instruction set,” says Mensch. To create the architecture, the three engineers created a simple diagram to represent the structure of the chip. “We would start with a basic block diagram,” says Peddle. Normally chip designers perform simulations on the instruction set, to ensure the steps produce the correct logic. Unfortunately, MOS Technology lacked the resources. “We couldn’t get the simulation tools we needed, so we had to do it all by hand,” says Mensch. “It was a Herculean effort that we put in.” Some of the most important design work occurred outside the walls of the MOS Technology building. “We put some of the more significant stuff in while drinking booze at Orgill’s house one night,” says Peddle. “The way to do really creative work is to work on it and then sometimes you’ve got to let it alone. If somebody gets a bright idea at a party, you take time out and you go argue about it.”

Al Charpentier was one of the calculator chip designers at MOS Technology. He witnessed Peddle driving his team to build the new processor. “Chuck was an interesting character,” he recalls. “He could be a bit pompous, but he had a vision and he was pushing that vision. Chuck was the visionary.” Peddle created a concept called pipelining, which handled data in a conveyor belt fashion. Instead of stopping while the microprocessor performed the arithmetic, the chip was ready to accept the next piece of data right away, while internally it continued processing data. This feature would make the chip faster than anything produced by Intel or Motorola at the time. A one-megahertz 6502 was equivalent to a four-megahertz Intel 8080. Peddle decided to create a version of the chip called the 6501, which could drop into a Motorola 6800 socket. “It was definitely not a clone,” he says. “Architecturally it’s a 6502. The only difference is it plugs into a Motorola socket.” The engineers hoped the 6501 would steal attention from Motorola. “Chuck sold the concept that Motorola would be out there building all these boards that will use the 6800 and we’re going to come along with a cost-reduced microprocessor to replace it pin for pin,” explains Mensch. “We just pull out the 40-pin package, plug in the 6501, and reprogram it.” Unfortunately, socket compatibility would later provoke a retaliation from Motorola. With the basic design complete, the layout of the transistors could begin. During this phase, the architects worked with the layout team to make everything fit within an allocated space. * * * The layout team consisted of two main engineers: Bill Mensch and Ray Hirt. A third engineer, Harry Bawcom, aided the layout artists. Soon, two more ed: Mike Jaynes and Sydney-Anne Holt. It was their task to turn an abstract block diagram into a large-scale representation of the surface of the microprocessor. Orgill was primarily responsible for the 6501 chip and Mensch for the 6502. Mensch grew up in a small farming community in Pennsylvania. “I lived on a dairy farm until I was 19,” he recalls. “It was a wonderful setting.” The future chip designer had plenty of company on the farm. “My mother had

ten children,” he says. The family had a history of suicides. Both his great grandmother and great-great grandmother had committed suicide, then when Bill was three his grandfather committed suicide. The troubling history hung over the whole family. During college, his father ran into medical difficulties. “He had a heart attack and he had glaucoma,” says Mensch. “He was uncomfortable and he had a back problem, and he was physically not able to work. So he just felt like he was a burden.” While in college, Mensch’s father committed suicide. After Mensch graduated from the University of Arizona, Tom Bennett hired him to work at Motorola on the 6800. “Mensch was literally right out of school,” says Peddle. At Motorola, Peddle was impressed with Mensch’s natural talent. “He was just spectacular doing N-channel design and layout. He was the world’s best layout guy.” Mensch was dependable, which made him a favorite with MOS engineers. “Bill was a good guy,” says Charpentier. “He was very knowledgeable and knew what he was doing.” Rod Orgill, the youngest member of the team, worked at Motorola on the fabrication process of the 6800. Out of everyone on the team, Orgill had the most diverse set of abilities. Peddle relates, “Rod was a combination of chip designer and architect.” Peddle claims the 6501 was a marketing game, with no serious plans to produce them in large quantities, but Rod Orgill believed the 6501 would be more successful than the 6502. According to Mensch, “We made a bet and said who’s going to have the highest volume. Rod says, ‘There’s no question: following Motorola’s marketing, the 6501 will sur your [6502] design and yours won’t even have a chance.’” The bet was on. * * * The young engineers worked in a small room on the second floor containing several large art tables. Here, Mensch and Orgill brooded over thick sheets of vellum paper. The layout consisted of thousands of polygons, each a specific size and shape. Thin lines called traces connected the polygons, creating a complex

circuit. Incredibly, the engineers created the layout in pencil, one component at a time. The task was formidable, with a completed diagram containing approximately 4,300 transistors.[1] Near the end of the design process, the engineers realized their architecture would not fit within the allotted area of the microchip. “When we sat down to optimize the system, we discovered we were 10 mills too wide,” says Peddle. “The design was almost done. Mathys and I put a big piece of paper down on a table and sat there and optimized every line until we got rid of 10 mills.”

MOS 6502 team (L to R) Wilbur Mathys, William Mensch, Rodney Orgill, Ray Hirt, Terry Holdt (seated), Chuck Peddle, Harry Bawcom, Michael Jaynes, Sydney-Anne Holt.

The engineers were on a tight deadline to have the product ready for the Wescon show in September. They obsessively searched for ways to recycle lines in the schematic, thus reducing the area. Peddle grimly recalls, “Mathys and I had to keep redoing the architecture to make sure they stayed within that area.” To print the microchips, the engineers used a process called Metal Oxide Semiconductor, or simply MOS. This process used six layers of different materials, printed one on top of the other, to build the tiny components on the surface of a silicon wafer. This meant the layout artists had to create six different diagrams, one on top of the other. The process required incredible precision because the layers had to line up exactly. The surface of the chip was necessarily dense in order to fit everything into a small area, so the artists squeezed transistors and pathways close to each other. If a single layer deviated by more than a few microns, it could touch another pathway and create a short circuit. After the layout was completed, the engineers faced the soul-draining task of rechecking their design. The most sophisticated tool in this process was a small metal ruler, or more accurately, a scale. Commodore engineer Bil Herd recalls, “They would take their scales out of their pocket—don’t call them rulers—and they would measure for months! They would measure each transistor and make sure it was two millimeters by point seven.” Mensch, Orgill, Jaynes and Bawcom sat bleary-eyed over their drawings, sometimes for 12 hours a day, painstakingly measuring every point on the layout. They measured the size of components, the distance between components, the distance between traces, and the distance between traces and components. If they found a component out of place, they had to erase and redraw the component. With a touch of sympathy in his voice, Herd explains, “You could be a really talented designer, but if you couldn’t check your design with the mind-numbing repetitiveness, your stuff didn’t work and you would get

a bad reputation.” During this time, Rod Orgill suffered a tragedy that placed additional pressure on Mensch. “His father died during the project and he left to be with his family in Idaho,” recalls Mensch. “He was gone for a couple of weeks, so that’s when I had to take over the project.” The engineers kept small cots in the room so they could work for long uninterrupted periods followed by a few hours of rest. “With the semiconductor guys, that tends to be something you do when you are doing that at a certain level of design,” recalls Peddle. “You tend to just keep going.” Even today, Peddle is still in awe of Mensch’s ability as a layout engineer. “Bill has this unique ability to look at the requirements for a circuit, and he can see how it is going to lay out in his head,” he says. “He’s just totally unique. Nobody matches Mensch.” The semiconductor team not only developed a microprocessor, they also developed the ing chips. The first was the 6520 PIA chip, which was a clone of the Motorola 6820 PIA. Mensch immediately began deg a chip called the 6530, which contained 1 kilobyte of ROM (Read Only Memory), 64 bytes of RAM (Random Access Memory), a timer, and two I/O (Input/Output) ports for moving data. The chip, which Mensch called the RRIOT chip (RAM, ROM, I/O and Timer), allowed engineers to assemble a simple microcomputer using only two chips. * * * In June 1975, the chip design was ready. It was now up to the process engineers to imprint the design onto tiny silicon wafers. Months earlier, Paivinen promised Peddle he would have the N-channel process ready. He was true to his word. “He gave me everything I wanted,” says Peddle. The procedure to shrink a large, dense design onto something smaller than a thumbtack is both mysterious and under-appreciated. In many ways, it is also the most important step and, if intelligently planned, it can reduce the cost of a microchip dramatically. Engineers simply call this step “the process.” When Paivinen and his two partners founded MOS Technology, it was their

explicit goal to be the best process company in the business. “MOS Technology’s business premise when they started was that they knew how to process better than other people,” says Peddle. Engineers at the time documented very little of what they did, and most process engineers stored the process in their heads. In order to print the transistors and other components to a silicon chip, the engineers had to create a mask. The mask blocks out everything except for the parts of the chip they want, much like a stencil blocks spray-paint to produce letters. The mask relied on the principles of photography and light. To transform the circuit diagram into a mask, the engineers used a material borrowed from the graphics industry called Rubylith. Rubylith is a sheet of acetate film with a red base covering the surface. Since the semiconductor industry was in an early stage of development, the tools to transfer the diagram were outrageously primitive. According to Bil Herd, “They were doing chips by cutting Rubylith with razor blades. They would push some tables together, kick off their shoes, and jump up on the tables.” It was up to engineers Mike Jaynes and Harry Bawcom to perform the tedious task of cutting out pieces from the Rubylith to form the mask. Bob Yannes, who arrived at MOS just after the Rubylith years, says, “I can’t imagine using that stuff. You’re looking at this huge red plastic thing in front of you and you’re supposed to peel off the parts that are supposed to stay and leave the parts that are supposed to go away. Unless you were very careful, you got the two confused and you ended up peeling off the stuff that is supposed to go away. Then you start taping it back down again.” With engineers crawling over the huge sheets of acetate film, it was vital sharp toenails were not exposed, which could drag over the surface and slice into the acetate. Engineers were not known for their attention to appearance and it became vital to keep fresh pairs of socks available. “Everyone would wear fresh socks with no holes in the toes for getting on the table,” explains Herd with some amusement. Jaynes, Hirt and Bawcom created six Rubylith masks for the 6502 chip, one for each layer. Once completed, the engineers photographically reduced each of the large sheets of Rubylith to create a smaller negative. Engineers chemically etched a tiny metal mask using this negative. The technicians would eventually

use this mask, almost like a rubber stamp, to create thousands of microprocessors. Precise robotic machines used the tiny metal mask to duplicate the pattern over the entire surface of the silicon wafer. In the early 1970s, the metal mask made with the surface of the silicon so the electrons could flow through the mask, imprinting the design to the surface. “People used to have what they call masks, which were pretty destructive on the mask,” recalls Peddle. “They actually put the mask on the chip and it got worn out very quickly.” Every time a mask wore out, the designers had to go through the laborious process of creating a new mask. At MOS Technology, John Paivinen implemented a new way to fabricate chips. “They were one of the first companies to use non- mask liners,” says Peddle. “At that time everybody was using masks.” With non- masks, the metal die did not touch the wafer. Once the engineers worked out all the flaws in the mask, it would last indefinitely. Paivinen and Holdt handed off the completed mask to the MOS technicians, who began fabricating the first run of chips. Bil Herd summarizes the situation. “No chip worked the first time,” he states emphatically. “No chip. It took seven or nine revs [revisions], or if someone was real good, they would get it in five or six.” Normally, a large number of flaws originate from the layout design. After all, there are six layers (and six masks) that have to align with each other perfectly. Imagine deg a town with every conceivable layer of infrastructure placed one on top of another. Plumbing is the lowest layer, followed by the subway system, underground walkways, buildings, overhead walkways, and finally telephone wires. These different layers have to connect to each other perfectly, otherwise the town will not function. The massive complexity of such a system makes it likely that human errors would creep into the design. After fabricating a run of chips and probing them, the layout engineers usually have to make changes to their original design and the process repeats from the Rubylith down. “Each run is a couple of hundred thousand [dollars],” says Herd. Implausibly, the engineers detected no errors in Mensch’s layout. “He built seven different chips without ever having an error,” says Peddle with disbelief in his

voice. “Almost all done by hand. When I tell people that, they don’t believe me, but it’s true. This guy is a unique person. He is the best layout guy in the world.” * * * With the mask complete, mass fabrication of the microchips could begin. Fabrication occurred in an alien-like environment on the second floor of the MOS Technology building called the clean rooms. These hermetically sealed rooms produced a nearly dust-free environment. The precautions were extreme, since a single grain of dust during the etching process could cause a miniature short-circuit. To enter the clean rooms, lab technicians were required to don hairnets, beard nets, moustache guards, gloves, paper booties, and white jumpsuits. “It makes you look like a bunny,” says Peddle. “We used a lot of them.” As a final measure, the technicians walked over a sticky-mat to remove the last traces of dust before stepping into the airlock. Within a crimson-tinted darkroom, technicians replicated print after print of the 6502 circuit. They coated the round silicon wafers with thin layers of metallic substances. After each layer, technicians placed the wafer into a special machine that copied the circuit from the metal mask to the surface of the silicon wafer. Electrons flowed through the mask, causing a thin layer to harden in the shape of the circuit. Each wafer had the chip pattern imprinted approximately fifty times. In another room, bathed in yellow light, technicians developed the microchips. This process was almost exactly like developing a photograph. A studious technician carefully washed each wafer with chemical solutions that removed all but the hardened circuitry. The industrial strength solvents went by names like trichloroethene, trichloroethane, dichloroethene, dichloroethane, and vinyl chloride. The technicians repeated the process six times for the six layers, each time using a different set of chemicals and metallic substances. “You put this mask on the device and do whatever step you are going to do, and then you take it off, put another mask on, and do another step,” says Peddle. The top layer was aluminum, which was the best conductor. Beneath the aluminum were various semiconductors such as germanium. Each layer went by a different name, such as diffusion layer, buried , depletion layer,

polysilicon, poly-metal , and metal. With all six layers applied, the wafers entered an oven to bond the circuitry. Technicians then added a ivation layer[2] to protect the fragile metallic circuitry from oxidation. After applying the ivation layer, a machine sliced the wafers into individual chips, each smaller than a fingernail. The chemical etching process used dangerous industrial solvents. Inevitably, the solvents evaporated into the air, which worried some of the staff. Robert Russell, an early Commodore employee, chuckles about the general indifference regarding this threat. “They had a chemical release in the production line that turned all your blueprints that were hanging on the walls different colors,” he recalls. “You would come in and they would all be yellow or green. You kind of hoped that wasn’t happening when you were breathing it.” “The production of semiconductors produces all kinds of nasty byproducts,” says engineer Bob Yannes. Inevitably, accidents occurred. “I things happening like occasionally we’d have a gas leak in the front end and you’d have people walking through the building saying, ‘Hurry! Get out of the building!’” Most people were ignorant of the dangers posed by the semiconductor industry. “This is a time in history when everybody looked at the clean rooms and the guys all wearing their bunny suits, and how sterile it was, and everybody wanted a semiconductor company in their hometown,” says Peddle. “It was high-tech, big money, and clean as opposed to a foundry or something like that. What they didn’t realize was these guys were dealing with the most poisonous, noxious shit in the world, and they had to put it somewhere.” The semiconductor industry was still new in 1970 when John Paivinen and his partners created MOS Technology. “Nobody in the semiconductor industry had a clue about how to deal with the stuff they were making for years,” says Peddle. “John did the best he could and he actually did pretty well.” The industrial solvents drained from the chip fabrication line into a 250-gallon concrete holding tank. “They built these double tanks and they stored it underground. But you know, we just didn’t have the technology,” says Peddle. “Let me just make a point—John Paivinen was a very meticulous guy, and he absolutely designed his tanks the best he could given the environment at the

time.” In early 1974, a serious disaster occurred. Technicians monitoring the tank realized it was emptier than it should have been. During the cold Pennsylvania winters, the concrete tank developed a small crack. “Some of their storage tanks leaked and it leached into the ground,” recalls Yannes. Paivinen kept the spill quiet, even from Peddle. “We didn’t him until the summer of 1974, and they wouldn’t have told us about it anyhow,” he says. “With all due respect, they keep that stuff a lot quieter in Silicon Valley. There’s been a whole bunch of stories about breast cancer being much higher in Silicon Valley, and there’s a bunch of other anomalies.” As the Environmental Protection Agency (EPA) later determined, the leak was the source of groundwater contamination in the area.[3] The Valley Forge Corporate Center bordered a residential development that relied on well water, so there was cause for concern. Fortunately, water tests at the time indicated the solvent had not yet entered the water table. Paivinen replaced the faulty tank with an unlined steel tank. * * * After the chemical solvents etched the chips, the technicians inserted the flecks of silicon and metal into an easy-to-handle package. Modern semiconductor companies typically place their chips in black plastic shells with silver pins. Back in 1975, MOS Technology placed its microprocessors in distinctive white ceramic shells with forty gold-plated pins. As if a price drop from $300 to $25 was not radical enough, Peddle and his team planned to release an ultra-low-cost microprocessor called the 6507. “Our goal was to do a $5 processor,” Peddle states flatly. “The 6507, which was a subset of [the 6502], could be made at a cheaper price. It was designed to be a really small package.” The packaging determined how cheaply Peddle could sell his chips. “Packaging costs money and pin outs cost money,” explains Peddle. “Back in those days, those big 40-pin packages were very expensive.” The 6507 package was plastic instead of ceramic and contained only 28 pins. In a perfect world, every single chip would work. If they fabricated 10,000

chips, they would ideally have 10,000 working chips. However, imperfections snuck in from every imaginable source. Inconsistencies in the etching process caused flaws. Small particles of dust getting in the way of the mask caused flaws. Even impurities in the silicon wafer produced flaws. The number of flaws the engineers could defeat determined the chip yield. Technicians methodically tested every single chip to determine if it worked. In 1975, most chipmakers considered a 30% yield to be quite successful. The industry simply discarded the remaining 70% of non-working chips. The process was inherently inefficient and resulted in monumental chip prices. If Paivinen wanted to achieve low-cost microprocessors, he would have to use every trick available to raise the yield. In the seventies, most semiconductor houses tested their chips with a Fairchild Century system. The huge machine occupied an entire room and cost almost a million dollars. As Bill Mensch explains, “We couldn’t afford them at MOS Technology.” Instead, Mensch constructed a small handheld chip tester that resembled a computer motherboard covered in IC chips. Every single chip from MOS Technology was hand tested by the homebrew device for the first year and a half of 6502 production. Through careful planning and innovation, MOS Technology achieved a chip yield of 70% or better. Peddle attributes this success to Paivinen and his non mask process. “Because they could afford to spend a lot more money making a perfect mask, they got much better yields,” he says. The low production costs meant Peddle’s vision would come true. Mensch designed the processor for speeds of up to two megahertz, but surprisingly it was capable of more. “Motorola and Intel processors ran at one megahertz at that time,” says Mensch. “The 6502 actually ran comfortably at four megahertz.” * * * In order to code a program to run on the 6502 processor, Peddle needed a development system. He turned to a former GE employee, Larry Hittle, to create the system. Peddle recalls Hittle’s contribution to GE. “He put GE in the communications business,” says Peddle. “He was the guy that put together the communications

system for the original Dartmouth machine.” Like many ex-GE employees, Hittle was an entrepreneur. “He got the idea of starting his own CRT [Cathode Ray Tube] company, a company called Courier Systems. So he spun a company out of GE. He started Courier and did all the things wrong that all the other entrepreneurs were doing. Venture capitalists stole him blind, and he finally dropped out of that company to start his own little assembly company.” Hittle formed another company called Monolithic Systems, and hired engineers Michael Corder, Bob Smith and Don Cook. MOS Technology contracted Monolith for the development system. “He was heavily involved with MOS Technology for a while,” says Peddle. The team set out to create development tools for the 6502, including a full compiler and assembler so s could program the microprocessor. They would call the development system the MDT650, which stood for Microcomputer Development Terminal. To write code for the 6502, developers would use one machine, such as a PDP11, to compile the code and then it to the device under development. This is called a cross compiler/assembler. The engineers needed a compiler for the 6502. The original compiler came from an Iowa-based company. “Com did all the developer software for MOS Technology,” says John Feagans, who would become a Commodore employee. “I also wrote the assembler when I worked for Com in Ames, Iowa. The original cross assembler was written in FORTRAN and ran on the mainframes of the day. I literally hand compiled each line into PDP-11 assembler first.” The Monolith Systems engineers took the compiler and modified it to work with the 6502. “Michael Corder wrote the resident 6502 version for the MDT650, an in-circuit emulator and development system done by Larry Hittel and Don Cook in Phoenix, Arizona,” says Feagans. Once the code is compiled, it needs to run on a 6502. Modern computers are powerful enough to have software emulation of other machines, but back in the seventies computers were not powerful enough. Instead, Larry Hittle and Don Cook created an emulator out of hardware, called an In-Circuit Emulator, or

ICE. Larry Hittle and his team succeeded in their task of creating development tools for the 6502. The MDT650 system included 64 kilobytes of memory and a chassis with 14 board slots. They planned to sell the system for $3,950 to potential 6502 developers. For now, they used the system for their own projects. * * * As part of MOS Technology’s marketing plan to encourage people to experiment with the 6502, the engineers developed two small computer systems. “They worked on them while we were finishing up the processor and getting ready to do the marketing,” explains Peddle. Engineers and hobbyists, the idea went, would use them to evaluate the 6502 instruction set and develop their own systems. The first development system offered by MOS was in kit form, which reduced the selling price. Since the unit was designed primarily to instruct the on the workings of computer systems in general, and the 6502 in particular, MOS Technology contracted Microcomputer Associates of Santa Clara, California to write the unit’s internal program. The two founders, Ray Holt and Manny Lemas, taught engineers how to use microprocessors. Peddle relates, “You have to understand how little the world knew of microprocessors in 1974, ’75 and ’76. There were guys making big money selling classes on microprocessors during that time.” Manny Lemas had worked for Peddle during his GE days, while Ray Holt had an impressive background working on the F-14 Tomcat project for the Navy.[4] The hardware engineers developed the kit at MOS Technology. Inside, engineers stared intently at oscilloscopes. Others pressed hot soldering irons against components to weld them to circuit boards. Small pieces of circuitry were scattered chaotically across the room. Since the 6502 microprocessor and ing chipset contained almost everything necessary for a computer, the design was minimal. When assembled, the could connect it to a teletype machine or a computer terminal. The biggest job was programming the built-in ROM code for the computer. This consisted of a debugger and monitor program, appropriately called the “demon.” According to Peddle, Manny Lemas and Michael Corder (who previously

developed the MDS650 compiler) programmed the demon. The programmers used a time-sharing system to develop the code, which they burned into the chip developed by Bill Mensch, a 6530-004 RRIOT chip. The engineers gave the kit a simple name. Peddle and his team liked acronyms, thus they christened it the Terminal Interface Monitor, or TIM. TIM would begin a predilection at MOS Technology and Commodore for asg friendly threeletter names to their products. Those ordering the $30 development kit received the grey-ceramic 6530-004 chip and a manual consisting of 14 photocopied sheets, folded and stapled in the middle. Included in the manual were a suggested schematic, the TIM monitor commands, a few sample programs and a listing of the monitor code. It was up to the to provide the resistors, transistors, capacitors, wire, and even the 6502 microprocessor. Though receiving a computer in the form of a kit does not seem particularly -friendly, hobbyists at the time clamored to build their own computer. Nonetheless, a good portion of the kits failed to operate upon completion. Rather than using a prepared circuit board, many buyers simply wire-wrapped the chips together, either on a breadboard or a piece of generic perfboard. After placing the required components on the board, builders hand wired the chips one pin at a time, resulting in a snarl of fine multicolored wires. Once the chips were in place, the then had to construct or purchase a separate power supply for the TIM. Finally, the TIM was (as the name suggests) able to interface with a standard ASCII terminal or teletype machine. Ultimately, TIM was just a stepping-stone to developing and marketing a fully assembled computer. * * * MOS Technology developed a second system concurrently with the TIM. This computer was slightly more -friendly—at least by 1975 standards. Rather than a chip and some instructions, this system arrived fully assembled, minus a power supply. It was a true development system, enabling s to code programs for the 6502. The inspiration for the new computer came from Don McLaughlin, one of MOS Technology’s founders. Peddle recalls, “McLaughlin said, ‘Listen, I think this is

a product that will help sell the [6502]’. They thought it was a good idea because they were calculator guys.” Peddle and a programming manager named Bob Winterhalt agreed with the idea and the three men began their design. According to MOS Technology employee Al Charpentier, his friend and fellow engineer performed the actual hands-on design work of the system. “That was done by a guy by the name of John May,” he recalls. “He was sort of the primary mover on that project.” At this early stage in microcomputer development, -friendly personal computers were barely on the horizon. Niceties like a video monitor, keyboard, software, power supply, or an enclosure were not part of most designs. The recently released Altair relied on switches for input and blinking lights for output. Any other interfaces had to be added by the . By today’s standards, it was comically impossible for most people to contemplate using these machines. This new sibling of TIM would share similarities, but differ in a few areas. As with the TIM, this unit would contain a 6502 processor running at one megahertz. However, McLaughlin advanced TIM’s basic design slightly, branching out in a unique direction. Instead of reading data from a row of flashing lights, the new computer would contain a six-digit display. Each digit in the display had seven segments, which could display numbers and letters. The primitive display was a step up from tiny lights representing binary digits used on most other systems. McLaughlin also improved on the basic input method for personal computers at the time. Rather than a row of switches for binary input, McLaughlin specified a keypad. John May eventually selected a black keypad with 23-buttons. This was a remarkable improvement over other microcomputers of the time, allowing s to enter code more easily. Both the keypad and the LED display reside directly on the surface of the printed circuit board (PCB), along with over a hundred precariously exposed components. The lack of a case or a power supply for the new computer clearly indicated MOS Technology was not targeting the machine for the mass market. A careless could easily damage the machine. Little TIM provided a paltry 256 bytes of memory, hardly enough to store three lines of characters on an 80-character computer display. TIM’s bigger brother would contain a full kilobyte of memory, comprised of eight MOS Technology 6102 memory chips. At the time, 1024 bytes was a generous amount. There was

even room for expansion. Two 44-pin edge connectors made data and control signals available to the builder for additional functionality. The computer also contained a 6530 RRIOT chip, designed by Bill Mensch. “After completing the design of the 6502, I immediately began design of the 6530 device,” he says. “It was for a two-chip system.” To a teletype machine, John May used the code from the TIM system. The 2-kilobyte program, also named TIM, contained the code to operate a cassette tape unit for storage, drive the alphanumeric display, and accept input from the 23 keys of the keypad. It also contained a monitor program, which allowed s to view memory contents and change code. A tiny bootstrap program would automatically start the monitor on reset. This was the pinnacle of -friendliness in 1975. The name for this new computer followed the tradition set by TIM. The TIM allowed input from a terminal, hence Terminal Input Monitor. The new system allowed input from a tiny black keyboard, so McLaughlin dubbed it the Keyboard Input Monitor, or KIM. They also added a number after the computer name, a practice later continued by Commodore. It contained one kilobyte of memory, hence KIM-1.[5] McLaughlin, a calculator engineer, had designed his first computer. “If you look at a KIM single board computer, it was a calculator engineer’s view of a single board computer,” says Mensch. “You had a Keyboard Input Monitor, so you used the keypad to code in instructions on the single board computer with a little LED character display.” The finished computer looked like a giant calculator. * * * The team now had dozens of working microprocessor chips, a development system, and demonstration computers, but their battle was just beginning. “We brought it out on schedule, on cost, and on target,” says Peddle. It was time to sell their microprocessors. [1] In contrast, an Intel Xeon introduced in 2008 has 1.9 billion transistors. [2] This final layer, called the ivation layer, was difficult for Paivinen to perfect. For mysterious reasons, small pinholes appeared in the ivation layer. After a year or so, the areas on the chip around these tiny holes would begin to

oxidize and the chip ceased functioning. [3] In December 1986, the EPA collected soil samples, surface water, and water from nearby residential wells. Tests revealed low levels of trichloroethene and other volatile organic compounds in the soil and shallow bedrock underneath 950 Rittenhouse. MOS Technology began a soil-cleaning program to extract the dangerous solvents, and in 1996, the residents received public water lines from an outside water source. [4] Holt claims he invented the world’s first microprocessor for the Navy in November 1969, approximately a year before Intel. Security restrictions by the Navy prevented him from disclosing this until 1999—by which time most people accepted that Intel was the first. [5] Former MOS Technology engineer Robert Yannes owns the first KIM-1. “I have a very rare thing that I scavenged out of scrap heap at MOS Technology: the original prototype KIM-1, and it still works. It’s a little bit different than what went into production,” he says.

CHAPTER 3

Selling the Revolution 1975-1976

Chuck Peddle envisioned a series of microprocessors of varying size and complexity. Price was the key to achieving widespread use. The full-featured version would sell for between $20 and $25. This meant the actual production cost could not exceed $12, otherwise it would be unprofitable.[1] “As I recall, the 4040 sold for $29 at the time in low quantities,” says Mensch. “One of the reasons we picked the $20 price tag for the 6501 and $25 for the 6502 was so that we could show a clear intention to compete with the 4040 system. If we were looking at competing with the 6800 or the 8080, we would have priced it at probably $150, which was half their price. But we weren’t intending to go after the 6800 or the 8080 market at all. We were going after the 4040 market.” With microprocessor economics, MOS desperately needed to sell high volumes of chips to overcome their design costs. According to Al Charpentier, the burgeoning microprocessor industry was having problems establishing itself. “You’ve got a new technology that everybody is interested in, but it’s not taking off,” he explains. “The numbers back then were tiny. They were scientific curiosities because they were so expensive. So [MOS] wanted to drive the interest level way up, and that’s how the $20 price tag got hammered in.” The price seemed unreasonably low compared to Motorola. “We wanted to own the market,” says Peddle. “If you want to own a market, you take a price point that you make good money at, and you make sure nobody else can play with you. You build big, fast companies that way.”

When asked why he did not choose a slightly higher price, say $50, he replies, “Because then I don’t get the design win.” A design win occurs when a customer’s company incorporates a component into one of their products. “At twelve bucks and fifteen bucks and twenty bucks, I get design wins everywhere.” Peddle was after widespread success. “We wanted people to put microprocessors everywhere. We were trying to change the world.” With a limited budget for promotion, it would be up to Peddle and his team to publicize the chip. “We wanted to launch the product in a spectacular way because we were a crummy ass little company in Pennsylvania,” explains Peddle. At first, he attempted to garner free publicity from newspapers. “Some people liked the story and put us on the cover of their newspaper, which hyped us up,” he recalls. Prior to launching the 6502, MOS Technology hired Petr Sehnal, a friend of Chuck’s from his days at GE. “Petr was a Czechoslovakian intellectual who came over to this country,” recalls Peddle. “He was kind of acting as a program manager and getting everything ready for the show, and he was the west coast sales manager.” To reach their target audience, Peddle and Paivinen planned to take the 6502 to the masses. The annual Western Electronic Show and Convention (Wescon) was being held in San Francisco on September 16. Sehnal knew the show would be the best place to launch Peddle’s revolutionary new product. The microprocessor would be useless to engineers without documentation. Peddle recalls, “We were coming down to launching, and my buddy [Petr Sehnal] kept telling me, ‘Chuck, you’ve got to go write these manuals.’ I kept saying, ‘Yeah, I’ll get around to it.’” Peddle did not get around to it. With Wescon rapidly approaching, and no manual in sight, Sehnal approached John Paivinen and told him, “He’s not doing it.” “John Paivinen walked into my office with a security guard, and he walked me out of the building,” recalls Peddle. According to Peddle, Paivinen gave him explicit instructions. “The only people you’re allowed to talk to in our company is your secretary, who you can dictate stuff to, and Petr. You can’t come back to work until you finish the two manuals.”

Peddle accepted the situation with humility. “I wrote them under duress.” A week later, with Petr Sehnal editing, he emerged from his exile with his task complete. “There were two great books they had that got everybody like Jobs and Wozniak into using the 6502,” says Bill Seiler, an early of the manuals. “There was a hardware manual and there was a software manual, and he had written the software manual. It was really good because it gave you code and lots of ideas for programming.” The 6502 would have manuals for Wescon. The team planned to sell samples of the 6501 and 6502 microprocessors at Wescon, along with the ing chipset. “We then took out a full-page ad that said, ‘Come by our booth at Wescon and we’ll sell you a microprocessor for twenty-five dollars.’ We ran that ad in a bunch of places,” recalls Peddle. The most prominent ment appeared in the September 8, 1975 issue of Electronic Engineering Times (EE Times).

ment for WESCON75. Note the emphasis on the 6502 over the 6501 (EE Times, Sept. 8, 1975, pp. 38-39).

On September 16, 1975, the day Wescon began, the engineers took out a patent on their chip, which lists the inventors as Charles Ingerham Peddle, Wilbur L. Mathys, William D. Mensch, Jr., and Rodney H. Orgill (in that order). Things were going well until his team arrived for the show. Peddle recalls, “We went to the show and they told us, ‘No fucking way you’re going to sell anything on the floor. It’s not part of our program. If we had seen these ads, we would have killed you.’” Having come so far and worked so hard, Peddle and his team were not ready to give up. “They told us this just enough in advance that we took a big suite, the MacArthur Suite, at the St. Francis Hotel,” says Peddle. MOS Technology would sell their contraband microchips from booth 1010 by redirecting buyers to a pickup location, much like drug dealers. The show ran from September 16th to the 19th. Mensch re the excitement around the new processor. “There is a second level you can look out over the convention floor and there were hundreds of people, all trying to get to the MOS Technology booth. I would man it for a couple of hours at a time with others.” Motorola also attended the same show. “What that did for Motorola is really piss them off,” recalls Mensch. “Here are ten people leaving Motorola that just completed the 6800. Motorola had no traction yet in the marketplace, and now they’re delivering a processor that’s caught the eye of the industry. They are really pissed.” The MOS Technology engineers redirected visitors wanting to purchase chip samples. “People would come by the booth and we’d say, ‘No you can’t do it here. Go to the MacArthur Suite and we can sell you the processors,’” recalls Peddle. “We became so popular people would get on the bus at the convention center and ask, ‘Is this the bus to the MacArthur suite?’”

The promise of low-cost microprocessors caused a sensation. Many people thought the $25 chip was a fraud or assumed it performed poorly. Peddle was confident these questions would resolve themselves once people started using his chips. The purpose of selling the chips at Wescon was not to raise money. It was to cultivate developer interest. Chuck’s wife Shirley greeted the engineers. “His wife Shirley was at the door and they had a big glass vase full of 6502s,” says Seiler. “If you paid $25, they’d give you one and they’d give you the manuals for an extra $10.” The large jars full of microchips seemed to indicate MOS Technology was capable of fabricating large volumes of the 6502 chip. This was subterfuge. “Only half of the jar worked,” reveals Peddle. “The chips at the top of the jar were tested and we knew the ones on the bottom didn’t work, but that didn’t matter. We had to help make the jar look full.” To dispel any notion that the 6502 was in some way inferior to other chips of the time, Peddle provided handouts with direct comparisons to their competitors. “I provided benchmarks that proved we were better,” he says. The manuals gently introduced readers to the concepts of microprocessor systems, explaining how to design a microprocessor system using the 6500 family of chips. It was a bible for microcomputer design. “Everyone told us how good they were to use,” Peddle recalls. “We were very proud of that.” Inside the suite, folding tables displayed an array of products, such as the 6530 RRIOT chip. “You could buy this little RAM/ROM I/O device for another $30,” says Peddle, who demonstrated the 6501 and 6502 chips, along with tiny development systems such as the TIM and KIM-1 microcomputers. Larry Hittle and Bob Smith demonstrated the ICE development system they designed for MOS Technology. “They would go around the suite and they would see the development systems, and they would find out how to log onto the timesharing systems so they could develop code,” says Peddle. “Then they would wander away.” If all went well, the engineers and hobbyists would go out into the world and design products with the 6502. Waiting in line outside the MacArthur Suite was Steve Wozniak, who thought he might be able to use the chip for a homebrew

computer project. “Woz and Steve Jobs bought their first 6502 and they used it in the Apple I,” says Bill Seiler. Peddle’s documentation undoubtedly influenced Wozniak’s design. The 6502 did not immediately improve MOS Technology’s finances, but it had a major impact on the computer industry. “It spawned a whole class of s, called hackers back then,” says Charpentier. “It changed the world,” says Peddle.[2],[3] Peddle looks back at the 6502 project as one of the best experiences of his life. “It was a unique time in history,” he recalls. “You only get to do one of those in your lifetime, I think.” * * * After Wescon, Chuck Peddle and Will Mathys began traveling across North America. “We started getting all kinds of phone calls from people wanting to know what was going on,” recalls Peddle. Many of them were calling because they were unable to attend Wescon. The two engineers agreed to handle the most promising twenty percent, while marketing sent literature to the other eighty percent. A MOS Technology secretary dutifully wrote down every call. “I was just flipping through them and I spotted this Grass Valley, California address.” The address was for Cyan Engineering. Peddle had previously met with two of its engineers at Wescon, Steve Mayer and Ron Milner. They had visited the MacArthur suite and spoke at length with Peddle in one of the back rooms about developing a video game console with the 6502. “I picked up the phone and the guy that answers is Steve Mayer. He was on the research team doing the Atari programmable game system. We did the Wescon show, got in a car, and drove up to Grass Valley.” The engineers gathered and began thinking of ways to design the game console using the cost-reduced 6507 chip. “In two days, we put together the architecture using my microprocessor and my cheap I/O device, and [the Jolt Computer] developed by [Microcomputer Associates], all under Mayer’s direction.” During two days of intense work, the engineers discussed price. “We literally put

together the plan and agreement. They came back and said, ‘This is the price I need for the chipset.’” Atari wanted to pay an incredibly low twelve dollars–not just for the 6507 microprocessor, but also for the ing chipset. Peddle and Paivinen mulled over the proposition. Atari already had some early success with a standalone Pong arcade machine and it was about to release the home version of Pong to the mass market. According to Peddle, he and Paivinen concluded that, “If we can make some money at it, let’s do it, because these guys have a chance for some volume.” Two days after the chip was unveiled at Wescon, Mayer had a deal. “He was truly an early adopter,” says Peddle. “We sold a microprocessor, RAM, ROM, and I/O chip all for twelve bucks, at a time when nobody could do a [similar chipset] for under $150.” Peddle learned Atari was also considering other processors, including the Motorola 6800 microprocessor and the Intel 8080 microprocessor. “Atari was tickled pink they had gotten MOS technology to sign up and sell them processors for $20 because at the time the 8080 was going for over $80, $90 or $100 bucks (depending on purchase quantities),” says Bill Seiler. “They thought they got a tremendous deal. They didn’t know that MOS Technology was yielding way higher than anybody else in the business. They were making processors for four or five bucks.” The other companies were unaware of their competition. “Motorola told their management they had Atari locked up.” It was a premature conclusion. Atari required a semiconductor company to fabricate its custom chips for the game console. Chuck Peddle and John Paivinen suggested Synertek. “The guy who was the president at Synertek was the project engineer in a computer lab in GE several years before, by the name of Bob Schreiner,” recalls Peddle. In a move that seems strange, Peddle and Paivinen wanted other companies to produce the 6502 microprocessor. They did not want MOS Technology to be the only semiconductor company producing the 6502. “If you want to be a big player in the microprocessor business, you have to have a second source for your chips,” explains Peddle. Peddle was more than happy to sell manufacturing rights to Synertek. “Schreiner

didn’t have a microprocessor, wanted one, and wanted the Atari business,” explains Peddle. Now Atari would purchase chips from both MOS Technology and Synertek. When asked why MOS Technology did not hold on to a 6502 monopoly, Peddle quickly replies, “People won’t give you a monopoly. They want a second source. Atari would never have signed with us if we didn’t have a second source.” From Atari’s point of view, a monopoly on the supply of chips would prevent price competition and almost guarantee MOS Technology could charge unreasonable prices. A second supplier gave Atari confidence that price competition would keep prices reasonable. “Schreiner was actually with us when we were doing the development,” reveals Peddle. “It turned out to be good for both of us.” The Grass Valley engineering team continued development on its game console. A former Synertek engineer named Jay Miner (who would later work for Commodore) developed the custom chips responsible for graphics and sound. “That Atari team was full of red-hot people,” reflects Peddle. Bob Shreiner even benefitted by getting a contract for Synertek to produce the custom video chip developed by Miner. “Atari gave him the contract for doing the custom layout for the game chip,” says Peddle. The Grass Valley team, including Steve Mayer, also used the 6502 to produce some early coin-operated arcade games. These were all black-and-white raster games. Sprint 2 (1976) was the first arcade game to use a microprocessor. This was followed by Night Driver (1976) which was the first first-person arcade game ever developed. Finally, Starship 1 (1977) was a Star Trek rip-off. Atari also experimented with the Motorola 6800 in the arcade game Drag Race (1977). In October 1975, Atari informed the other companies that it had decided to use the MOS Technology chipset. Motorola was furious. “They sued us a week after Atari announced they were using our product,” recalls Peddle. On November 3, 1975, Motorola filed a lawsuit against MOS Technology alleging patent infringement. “They claimed we had stole all their intellectual property. We didn’t steal anything,” contends Peddle. “It didn’t look like the 6800, it didn’t smell like the 6800, we didn’t want it to be like the 6800.”

Mensch recalls the personal nature of the lawsuit. “They were pissed so they sued MOS Technology and each of us individually.” Motorola knew it could have had the 6502 when Peddle offered it to the company a year earlier. Now, in the face of losing a valuable contract with Atari, the magnitude of its error became apparent. “They sued us because their pride was hurt,” claims Peddle. In the meantime, Peddle continued traveling the country, putting his microprocessor into everything from printers to entertainment devices. “He was out on the road after writing the manuals and getting the processor going, trying to sell them into all sorts of designs,” says Bill Seiler. “There was a company in California that he designed the printer for. It was a little dot matrix printer that used the 6502.” “If I wanted to sell my microprocessor to a printer company, I had to go help design the product for them,” says Peddle. “I would pack my development system up and we would go somewhere. In a week, I would design a microprocessor system. That’s how I built my sales up.” * * * Following Wescon, news about the 6502 began to appear in magazines and periodicals. EE Times was skeptical, and in an article entitled “Does the Country Need a Good $20 Microprocessor?” questioned if MOS Technology would be able to provide good documentation and for such a cheap product. It also asked if the 6502 was superior to other processors. Despite the lower price, Bill Seiler maintains the 6502 was better than the more expensive 6800, 8080 and Z80 processors of the time. “They did a better job of the instruction set than Motorola did.” Thanks to a review in the November issue of Byte magazine, the chips soon gained a larger following. Dan Fylstra, founder of the company that would someday sell the legendary VisiCalc spreadsheet, wrote an article entitled, “Son of Motorola.” People were soon convinced the 6502 chip was a legitimate microprocessor. In the months that followed, engineers and hobbyists began demonstrating products with the MOS microprocessors. MOS Technology began selling the KIM-1 in late 1975. Buyers who sent away for their KIM-1 were pleasantly surprised to have a rectangular cardboard box

arrive from “MOS Microcomputers”, a short-lived division of MOS Technology. The KIM-1 circuit board arrived sealed in a black static-proof bag, surrounded by thick foam padding with manuals and documentation on top. The documentation included with the KIM-1 went beyond other computers of the day. There were three manuals—a 200-page 6502 programming manual (written by Peddle under duress), a 100-page KIM-1 manual, and a 150-page hardware manual. The writing was friendly, concise, and detailed. Most importantly, it did not assume the knew everything about computers already. The KIM-1 manual promised, “You should be able to achieve initial operation of your KIM-1 module within a few minutes.” Of course, this assumed you had access to a 5 volt, 1.5 ampere power supply. A 12 volt supply was required if the cassette tape was to be used. For those in doubt, the manual contained complete instructions and a parts list for building a power supply. Once the power supply dilemma was solved, the hit the RS (reset) key to start using the system. This started the TIM monitor program running from ROM, which displayed numbers and accepted input from the keypad. Unlike today’s systems, the KIM-1 contained no on-off switch. s then began the exacting process of entering code into the machine in order to make KIM do something. After entering all the data, it was simple to run the program—just set the computer to the address where the program began and hit the GO button. If the program misbehaved, the KIM-1 also had a switch on the keypad labeled SST (single step). This would cause the computer to execute the program one instruction at a time. s appreciated this important feature, which greatly assisted in debugging. The built-in cassette-tape interface of the KIM-1 proved indispensable for early hobbyists because it allowed them to save and load their work. In contrast, s of the MITS Altair had no way to save programs with their basic system. They would sit in front of their machine, laboriously flipping switches to enter their program into memory. If someone happened to trip on the power cord, the programmer had to start all over again. The tape-interface alone made many KIM-1 owners fall in love with the computer, and many praised it for its reliability. Tape storage was the perfect medium for a 1-kilobyte computer. Programs loaded and saved rapidly, and dozens of programs fit onto a single cassette. Of course, a cassette recorder was not included with the KIM-1, so it was up to the to find one. It was also up to the to connect it to the KIM-1 by interfacing the microphone input and

speaker output jacks to the gold-plated I/O pins of the KIM-1. Another advanced feature of the KIM-1 was its ability to connect directly to a Teletype machine or computer terminal through a built-in serial interface. Teletype machines were large electromechanical devices with the ability to enter data through a keyboard, print hard copy, and load and save data via punched paper tape. A noteworthy feature of the KIM-1 was its ability to automatically adjust to the speed of the teletype connected to it. People were amazed to see the tiny KIM-1 operating a massive piece of hardware normally connected to minicomputers or mainframes. This helped to convince skeptics that microcomputers were true computers. As it turned out, however, many people preferred using the LED display and keypad to the noisy, messy, and costly teletype machines. s received the small computer enthusiastically. Al Charpentier recalls, “They sold a lot of those. It was sort of the first fully packaged microcomputer that you could take out of the box, throw a power supply on, and do something with. It was hell, but it educated people on the processor.” Engineer Robert Yannes recalls KIM-1 engineer John May showing the machine at his college. “I had a lot of familiarity with the KIM-1,” he says. “The guy who designed that was actually a friend of Al Charpentier’s, and he was a Villanova graduate too. He had brought it to Villanova University when it first came out and I had gone to that presentation. They had KIM-1’s at Villanova too, so I ended up playing with them.” “The KIM-1 had one characteristic that everybody always commented on,” says Peddle. “It was a packaged, complete, plug-it-in-and-start-using-it product. You could sit down and learn to program using my manuals.” * * * While trying to interest customers in the 6502, Peddle met with Allied Leisure Industries in Hialeah, Florida, a city in the densely populated area around Miami. “I went to Florida because this guy from Allied Leisure said he had a fun application,” he says. The company assembled pinball machines and electronic games from a large warehouse at 245 West 74th Place. At the time, Allied Leisure was larger than Atari. “He had knocked-off Pong when it was in the arcades and bars, and he had been very successful,” says

Peddle. Allied Leisure created two coin-operated Pong clones in 1973, Paddle Battle and Tennis Tourney, which reportedly sold in excess of 30,000 units. “He actually out-Ponged Atari as far as the games go.” Like most companies Peddle visited, Allied Leisure had no expertise with microprocessors at the time. “That was a new frontier. Embedded micros were fairly new,” says Bill Seiler, an Allied Leisure employee at the time. Allied Leisure’s engineers created the logic in their electromechanical games by wiring together masses of transistors, called Transistor-Transistor Logic (TTL). “We had a pinball machine that had a bunch of boards in it and it was all just discrete TTL logic,” says Seiler. “The engineers who did it didn’t know anything about clocked logic. It was awful design. Some boards wouldn’t work with other boards because they had logic race conditions in them. It was really crazy.” Allied Leisure wanted an easier, more reliable and less costly manufacturing process for their game machines and they began looking at adopting microprocessors. “Everybody came by to sell us their microprocessor,” recalls Seiler. “They said, ‘Just buy the demo boards and do it yourself.’”

Assembling electromechanical games from Allied Leisure’s warehouse.

Peddle went a step further than his competition. “They were into electromechanical games in a big way,” says Peddle. “We got into a discussion and I volunteered to build them a pinball game.” “Chuck came by and he not only talked us into using his micro but he sat there for about a month writing the code,” laughs Seiler. “So we definitely went with that deal because we didn’t know anything about programming in the company at the time.” Peddle saw pinball as a good opportunity to demonstrate the capability of the 6502 to control multiple mechanical devices. “I wanted to do the game because it was a fun thing and it used lots and lots of I/O devices. We could do some fun stuff and we could keep up with the speeds. I wanted to prove to everybody that you could make the microprocessor do all that stuff well.” While there, a scruffy young engineer befriended Peddle. “There was this kid— ponytail, shorts and everything. He kind of got involved and hung around while I was doing this development.” The young engineer was Bill Seiler, who also designed devices for Allied Leisure. “He was deg analog circuits that made gaming noises,” says Peddle. “Cars would sound like cars, airplanes would sound like airplanes and that sort of thing.” Allied Leisure created electromechanical games, which differed from videogames because they did not produce a digital screen image. Instead, the games used mechanical devices, such as film projectors and motors, to produce images. One of their electromechanical games, Monte Carlo (1973), was essentially the classic wooden-maze game of Labyrinth where the player tilts the playfield to control a steel marble through the course. Allied Leisure added racing sound effects, a steering wheel, and lights to enliven the game. Another game, Ski (1975), immersed players in a downhill ski race. Players stood on a pair of rotating foot pedals and grabbed hold of ski poles to steer down the hill.

“We made a lot of mechanical stuff. That’s what they had a good forte at,” says Seiler. “I think their favorite game was a Snoopy airplane game. It projected pictures on the screen and you sat in this chair and flew it around with the joystick. It was all mechanical. The little kids liked it because the seat would move when you moved the joystick around.” Allied Leisure required technicians to fix their broken circuit boards when arcade operators sent them in for repairs. “I got hired on as a technician,” says Seiler. “They had electronics using the old relays but their electronics were really badly designed. I was using a scope and a schematic, and going through the logic.” Back in 1968, engineers Ron Haliburton and Dave Braun founded the company as a spinoff from Allied Leisure’s slot machine business. Cuban exiles supplied most of the low-cost labor in their warehouse on 4th Avenue. “I really had a lot of respect for the Cubans that came back then because they came over after the Castro thing with nothing,” explains Seiler. “They worked hard and moved into the slums. They fixed up the homes, painted them and cut the grass and the poor blacks had to move out because they couldn’t pay the increased taxes on them. They ended up getting into the government and now all of Miami is politically run by the Cubans.” The methods used by other technicians bewildered the university-trained engineer. “There was this Cuban guy who didn’t know anything about logic,” he says. “He had this little book, and if a videogame like a Pong game was doing something funny, he would just change three chips and it would fix it. He did this other thing where he would rub on a chip with a pink pencil eraser and it would start working again for a few seconds, and then it would go away and he knew that was the bad chip. I said, ‘This is voodoo, man.’” Although the Cuban technician lacked Seiler’s university background, he was often very effective at repairing the boards. “He fixed more boards than I did,” says Seiler. Dozens of low-skilled workers assembled the game machines, turning out dozens every day. “It was all these crazy Cubans building these pinball machines,” says Seiler. “I think part of the Cuban language is spoken with hammer.”

The high failure rate of many electromechanical games was the result of over complexity at the design level by Seiler’s mentor. “The guy I was working with down there was a Cuban guy. He dressed like John Travolta in Saturday Night Fever. He was a swinger but he was married,” recalls Seiler. “He had really good ideas but he wasn’t a very good electrical engineer. His designs worked but they weren’t that good.” Seiler’s first design work came on a game called Daytona 500. “It was mechanical and the pictures were projected on the screen. You’ve got to the cars and swerve around them.” Seiler had enough audio experience from his university days to qualify as the resident audio expert. “Mostly they stuck me with the sound because they knew I knew a lot about audio,” he says. On his very first design job, Seiler decided to outdo himself. “I noticed the cabinet looked like a folded horn speaker,” he recalls. “I put a pretty big audio amplifier in it and it was loud. You heard screeching, crashing and motor noises and all this stuff. Half the time the guys in the arcades turned it down.” Seiler loved computers, but after university, he had difficulty obtaining computer time. Peddle began to sense a demand. “People used to be able to get their hands on computers. Then, in the late ’60s and early ’70s, there was a big revolt against technology. People were attacking computer centers with axes, claiming computers were taking over our lives. We’re talking about serious hippy-type stuff. So all of the computer rooms locked the doors.” The need for security drastically reduced the freedom people previously enjoyed. “If you wanted to get a computer run, you walked up with your punch cards and left them on someone’s desk,” says Peddle. “They went from these time-sharing friendly, I-can-do-everything systems to having zero access to the computer.” Peddle noticed the demand for small computer kits like the TIM. Seiler ordered a computer kit from a fledgling company called The Digital Group, run by a Dr. Robert Suding in Denver, Colorado. “I got a Digital Group, which was a kit like the Altair or IMSAI,” recalls Seiler. “It was a neat thing because you could plug in different U boards like the 6502, 6800, Z80, or 8080.” As Chuck Peddle worked on his pinball machine, Seiler assembled his computer

and worked through the inevitable problems inherent in early computer kits. “I was putting that together in the engineering lab because they had scopes and soldering irons,” he says. “Chuck was sitting there writing the code for the pinball game and I was playing with my Digital Group computer. He was standing there next to me when I finally got it working.” The computer was notable for booting from a ROM chip when powered on. “The very first program you get going is a little text-based display,” recalls Seiler. “You would stick an AM radio next to it and it would play the Star Spangled Banner while it was slowly printing this American flag on the screen.” The moment was a revelation for Peddle, as he realized there must be thousands of people like Seiler who wanted their own computer. “[Seiler] was the first guy I actually saw bring up a personal computer,” recalls Peddle. “He was all proud of it.” Seiler re Peddle telling him, “Someday Bill we are going to build one of those. We’re going to do that.” It was a promise the older engineer would keep. After the impressive debut of The Digital Group computer, Seiler ran out of ideas. According to Peddle, “Basically, he didn’t have a clue what he was going to do next.” Peddle saw firsthand how computer s became stuck due to the lack of proper applications. A picture started forming in Peddle’s mind. “Seiler triggered in me that people who were building this stuff didn’t have a clue what they were going to do with it, and when they were all done they didn’t have anything that really worked,” he explains. “The programmers said, ‘I don’t know how to build my own kit. I can’t get anyone to build one for me, and when they work they don’t work very well.’” It was as if Henry Ford introduced the world to the automobile by selling a box of parts. “I realized that we were selling KIM-1’s because a lot of people didn’t want to build their own kit,” explains Peddle. “But the KIM-1 didn’t do enough. It just wasn’t a strong enough product for anything and wasn’t intended to be. It was supposed to be a way to play around with my processor.”

In time, Seiler acquired a copy of BASIC on a paper tape roll that hobbyists freely shared. “There was a guy in Miami who had an Altair,” he explains. “I got Bill Gates’ original 4K BASIC from him.” Seiler’s friend in Miami had no means for loading the BASIC code into the Altair’s memory, other than a series of switches. “It was on a punch tape, one of those little paper tapes. On an Altair, you could switch in the bytes on the front , kind of like the old minicomputers could. You could enter in the address and write what you wanted in the byte. He put the whole 4K in by hand.” Once the BASIC resided in RAM memory, any disruption in the power supply would cause the Altair to lose the program. “If the power went off, he had to reenter his whole BASIC in, so he just left his machine on all the time so he could have BASIC. This guy was ruthlessly crazy.” Seiler lent his expertise in electronics to aid the man. “I helped him build a paper tape reader we found in some junkyard so he didn’t have to do that every time.” In return, the man gave Seiler a copy of BASIC. Seiler thought nothing of the pirated copy because there was a weak concept of software ownership in 1975. Then in January 1976, Gates addressed the nascent community in the Homebrew Computer Club Newsletter. “Bill Gates wrote some kind of open letter to some magazine and Bill [Seiler] read it,” explains Peddle. “Gates was saying, ‘You guys are taking my livelihood. I developed this thing and you’re stealing it. I’m not making any money and we’re going to go out of business.’ Bill Seiler felt so bad he sent Gates the money for the copy he stole.”[4] Seiler wanted to pay for his copy of BASIC, but he also felt Micro-Soft (as it was named) was asking an outrageous price. “At the time I think he was selling it for $500,” recalls Seiler. “I wrote him a letter back and sent him $35 and said, ‘This is what it should cost. You want too much money for it.’ I eventually met Gates and he said he still had that check. He said it was such a riot there at Micro-Soft.” Observing the difficulties Seiler and others encountered with their hobbyist computers, Peddle believed he might be able to construct a better personal computer. He went back to MOS Technology to propose a new product. At the time, Paivinen was ive of the idea and agreed to fund development. “I had a commitment from the company that as soon as I got to a certain point, I

would get to do my computer,” he recalls. Peddle’s interactions with Seiler started him thinking about personal computers, but for now he was too busy to do anything with the idea. “In the course of deg my [pinball] game, I used to work all night and sleep on a mat at their place,” he says. He completed the design and showed it at the National Computer Conference. It was the first pinball game to use a microprocessor. While in Florida, Peddle also spent time deg a photocopy machine for a company called Saxon. “I sold one of the copier companies on the idea of doing a copier using my microprocessor.” Throughout late 1975 and 1976, Peddle travelled the country promoting the 6502. His microcontroller became part of an important new technology called the ARPANET (later renamed the Internet). A company named Telenet (later acquired by GTE) created its own commercial network. “GTE had their Telenet packet switching,” recalls Mensch. “Telenet was one of the early fiber providers for the early Internet.” Packet switching required fast processors, and originally the company used costly minicomputers as switches, but then decided to move to something less expensive. “Telenet was a very high performance packet switching for communications systems,” says Mensch. “They tried all the other chips and said the only one that can do it is the 6502.” GTE settled on the 6502 and incorporated the processor in its switches throughout the network, making the 6502 an integral part of the early Internet. During Peddle’s travels, the idea of a real personal computer began to crystallize in his mind. He could see that programmers wanted their own computer at home, similar to a mainframe computer terminal, except without the bulky mainframe. He planned to begin work on a computer for MOS Technology once he finished marketing the 6502. * * * In December 1975, the coveted inside front cover of Byte magazine contained a two-page ment for “the world’s lowest cost computer system.” Though it contained a 6502 microprocessor, it was not the KIM-1. It was the Jolt computer, sold by Microcomputer Associates, either as a kit for $249 or fully assembled and tested for $348.

Jolt, named after Ray Holt, who helped develop the TIM and KIM-1 code, was technically similar to the TIM computer. Although Jolt competed with the KIM1, Peddle did not object. “Manny just said, ‘We want to do this board of our own’, and I said ‘great’”, explains Peddle. “I was looking for anything that would help customers design with the [6502] product. We gave these guys the license.” Nonetheless, Jolt did not have lasting popularity with the hobbyist market. Jolt’s most notable achievement lies in its use as the platform for the Atari 2600 VCS prototype system. The Jolt ment in Byte influenced MOS Technology’s marketing approach. A few months later, in the April 1976 issue of Byte magazine, a new product announcement appeared for the KIM-1 titled, “What’s New, KIM-osabee?” There was also an ment from MOS Technology itself. The lowkey ad included a clip-out order form for a $245 KIM-1 microcomputer system. Anyone who understood computers recognized the potential immediately. The ment in Byte caught the attention of the hobbyist market. A month later, Byte ran a feature article titled, “A Date with KIM.” Byte contributor Richard Simpson gushed about the low price and quality of the feature-packed KIM-1. He identified it as the ideal system for anyone who did not want to assemble a kit. The KIM-1 subsequently became a favorite of Byte and other popular homebrew publications, such as Dr. Dobbs Journal, Kilobaud, and Interface Age. Articles and projects appeared in these magazines well into 1979. MOS Technology released the KIM-1 in 1975, the same year as the Altair 8800 computer. The Altair has come to be known as the first personal computer system in North America to herald the new microcomputer revolution. The differences between the KIM-1 and the Altair computer illustrate a split in design philosophy within the computer world. The KIM-1 was a single-board computer, with all components mounted on a single printed-circuit board. It had room for expansion, but there were no slots to insert adapter cards. This design philosophy reduced production costs and thus gave the KIM-1 a major pricing advantage over the Altair. The Altair 8800 used an Intel 8080 chip, which retailed for $360, but inventor Ed Roberts was able to negotiate the price down to $75 each in bulk. Still, he needed to sell his computers for $439 in kit form, and $621 assembled to make a profit. MOS Technology was able to sell KIM-1 systems profitably for $245.

Though it was not a true personal computer, MOS Technology soon discovered the KIM-1 had a large market. “That was one of the things that took MOS by surprise,” recalls Bob Yannes. “Throughout the early days of computers, one of the most successful computers introduced in that timeframe was the KIM-1.” Although the KIM-1 was a reference system, meant to instruct engineers, it often became an integral part of products. “They had developed the KIM-1 as a sort of sales tool for the 6502 processor,” explains Yannes. “They would say, ‘Here’s a development system for you. You can design your own computer system and develop your software on the KIM-1 and help understand the hardware architecture and so forth.’ And people would use them and say, ‘Why do we want to design our own computer? We have one right here and it’s only [$245], which is cheaper than we can build it for.’ They would just buy KIM-1’s and bury them in their products.” Hobbyists began enthusiastically calling and writing for the kit. Some hobbyists, such as future Commodore employee Andy Finkel, came straight to the source to purchase the computer. “I showed up there with my check to buy a KIM-1,” he recalls. Scrambling to find him a computer, the MOS Technology receptionist made some calls. “I was in the reception area in the front, sitting on one of the couches waiting while they got a KIM-1,” he says. “They really weren’t set up to sell something like that, but they managed to do it.” Eventually, someone appeared with a kit. “Chuck came out, said hello and shook my hand,” recalls Finkel, laughing. “He was the one who wanted to give the kid the KIM-1. I told him how much I liked his processor and that was pretty much the end of it. It was just a nice thing.” As hoped, the do-it-yourself nature of the kits spawned familiarity with the products, and once hobbyists had invested time learning about the chip, they often remained loyal to the 6502. Many hobbyists ended up using their TIM computer as a small development system, since it was ideal for creating small programs. Though the goal had been to drum up interest in the 6502 chip, it soon became apparent that microcomputers would also be a valuable source of revenue for the company. According to Kilobaud magazine, MOS Technology sold over seven

thousand KIM-1 computers by June 1977. At $245 each, revenue was in the millions, which helped MOS pull through a tough financial period. “They sold a lot,” says Charpentier. “By God, they sold thousands of them—ten thousand or something like that. It was a big number of processors back then.” There was an obvious demand for computers. The appeal of the KIM-1 was not lost on Chuck Peddle. “It was a complete package, and there are a lot of people who bought it just for that reason and learned something, and then said ‘Okay, that’s all I can do.’ But we were seeing those people and talking to them and getting .” [1] In the microprocessor market, the manufacturer doubles the manufacturing cost when selling to a dealer, who then doubles the price again to sell to the consumer. Since MOS Technology would sell the microprocessor directly without an intermediary, it only doubled the manufacturing cost once. [2] In September 1995, as part of its 20th anniversary edition, Byte magazine named the 6502 one of the top twenty most important computer chips ever, just behind the Intel 4004 and 8080. [3] Even pop-culture recognizes the 6502 chip. The animated show Futurama revealed that one of its characters, the robot Bender, has a 6502 microprocessor for a brain. Futurama, “Fry & the Slurm Factory” (Season 2, Episode 4). [4] Peddle had a chance to ask Bill Gates if he ed receiving the check. Gates replied, “Yeah, and we really noticed it because almost nobody did [send us a check].”

CHAPTER 4

The Acquisition 1976

In early 1976, Allen-Bradley, the financial backers of MOS Technology, began to reconsider its investment. “Somewhere about the time we launched the 6502, we spent a lot of money fighting Motorola,” recalls Chuck Peddle. “AllenBradley decided to part ways partially because of the Motorola lawsuit and partially because they were in a very unhappy working relationship with some of management at MOS Technology, who were difficult to work with from time to time.” Furthermore, the decline of the calculator business made Allen-Bradley nervous. “[Texas Instruments] was in the process of just fucking the calculator business big time,” recalls Peddle. “A lot of the customers for MOS Technology were dying. Allen-Bradley looked at it and said, ‘One of the reasons we got into this business is we thought there would be some synergy. There’s no synergy.’” Inconceivably, rather than liquidate MOS Technology, Allen-Bradley gave the company to the three founders for almost nothing. “They decided to just not be involved anymore,” says Peddle. “Allen-Bradley basically handed the company back to the founders.” Why would a large corporation like Allen-Bradley give away a company that just launched a hot microprocessor? “Nobody at Allen-Bradley thought about microprocessors being hot,” explains Peddle. “It was just not even in their mind. Microprocessors became strong because we made them strong over that time. But six months before, right after we launched it, it wasn’t a big deal in of the volume.”

With Allen-Bradley out of the picture, the three founders now owned a company largely built with other people’s money. John Paivinen, Mort Jaffe and Don McLaughlin must have felt giddy. “The founders knew they had a winner on their hands with the microprocessor,” says Peddle. “They figured if they could get the thing back for free, they would do that. I suspect they paid a little bit for it, but effectively free.” Without Allen-Bradley, the financial backing suddenly stopped. “Now they had the company in their hands, but they didn’t have the financing,” explains Peddle. “The semiconductor business tends to eat money even though it generates a lot of money. All the things, like the KIM-1 and all that stuff were coming along. And the Motorola suit had hurt.” The three owners attempted to solve the problem of financing. “I knew during that time they were thinking about raising money by some kind of public offering,” recalls Peddle. Unfortunately, their efforts to put together a public offering failed. As a result, “they were ripe for some kind of a deal,” says Peddle. * * * Peddle helped anyone with an interest in developing for the 6502, including two kids working out of one of their parents’ garage. “While we were out visiting Atari, my west coast [sales] manager [Petr Sehnal] said, ‘Hey, there’s some kids working on a machine in his garage and it’s not working. We have a development system with us, so why don’t we go over and help them?’” Steve Wozniak, the young man who visited Chuck Peddle in his hotel suite, was a smart, inquisitive, impish young man with a penchant for electronics. His sidekick was Steve Jobs. Wozniak worked for Hewlett-Packard at the time and he began deg a computer while at work. Other engineers influenced Wozniak when he created his first computer. Ray Holt claims Wozniak purchased a 6502 based JOLT computer from him, which Wozniak undoubtedly studied. He also learned from the MOS Technology manuals, included with the 6502. “My books had a lot of influence on him,” says Peddle. Both Peddle and Sehnal visited the young entrepreneurs. For Steve Wozniak, it must have been a thrill to meet the designer of his favorite microprocessor.

“These guys were in the Bay area and their product didn’t work, so we took my development system over and spent an afternoon with them in the garage helping them bring up their system,” recalls Peddle. Peddle used his MDT650 system to program the ROM code in real time. “It was a great big bulky thing with a big old heat sink on the back,” says Peddle. “You plug it into the board you were working on and you could program it right there.” Peddle finished his work and the two kids promised to keep in touch. Surprisingly, Peddle and Steve Jobs formed a lasting friendship, despite Jobs’ lack of engineering background. “Steve [Jobs] is a hustler. He was a hustler back then and he’s a hustler now.” In February 1976, Wozniak brought his 6502 based computer, the Apple I, to the Homebrew Computer Club. The reception was lukewarm since most s were more favorable towards Intel and Motorola processors. Without a printed circuit board, the computer looked even less friendly than MOS Technology’s KIM-1 computer. Wozniak attempted to interest Hewlett-Packard in his design, but it was not interested. In April 1976, Wozniak obtained official release of the technology from Hewlett-Packard. In July 1976, after refining the Apple I to include a keyboard connector and display adapter, Wozniak and Jobs made their sales pitch for the Apple I at the Homebrew Computer Club for $666.66. The price included just the circuit board alone, and did not include a case, a keyboard, or even a power supply. In all, Apple sold about 200 Apple I computers—far less than the KIM-1. * * * The previous years with Chuck Peddle had been exhilarating for all of the 6502 team. They made headlines when they departed Motorola for MOS Technology and they even had their group picture featured on the front page of Electronic Engineering Times, which they put on the wall in the MOS Technology lab. The 6502 had revolutionized the microprocessor industry. Anything that came after was bound to be anticlimactic. Peddle recalls the high of working with the 6502 team. “It was the best project team I have ever worked with and it was the best product I’ve ever worked on.

All the other guys who worked on it feel the same way.” Unfortunately, the success created a dilemma. “The problem was, it was a very successful project, with all kinds of hype. Then—and I went through this with my second wife—there’s a time when that’s over and all you’ve got to do with life is business.” After the product launch came the tedious legal issues with Motorola. “We went from there into what most of the guys saw as the lawsuit and all the shit we had to go through for that,” recalls Peddle. Suddenly engineers were spending time giving affidavits to lawyers and gathering information, usually accompanied by excruciating explanations to educate the lawyers and the courts on microprocessors. Peddle also blames some of the management for not valuing its engineers enough. “Paivinen is a wonderful guy but he’s not a people guy at all. He had his favorites and everyone else he treated badly.” “There were serious problems within management too,” recalls Peddle. “McLaughlin and Paivinen didn’t like one another very much.” Peddle believes the three owners made the engineers feel they were not integral in the direction MOS Technology was taking. “The top three guys got very selffocused as a threesome,” Peddle explains. “They were so focused on getting the company back [from Allen-Bradley].” While Peddle busily promoted the 6502, his engineers were without a goal. “I went from driving the program to marketing and things like that,” explains Peddle. “Some of us were busy and very much focused, but the rest of the team had assignments that weren’t as exciting.” “When Commodore took over, that was kind of like the end of the road because Commodore was really buying it for the calculator business,” explains Peddle. It was anticlimactic to go from microprocessors back to calculator technology. “The calculator guys had real assignments, but these guys didn’t want to fool around with calculators,” says Peddle. “They had done that earlier in their lives, so it was just easy for them to melt away.” Arguably the most valuable member under Peddle, Bill Mensch, began to plan

his escape. “Fundamentally Bill was a very young guy at Motorola, right out of school, but he was a real talent,” recalls Peddle. “At MOS Technology he definitely took bigger and bigger pieces, and he had matured, if you will. He picked up a lot of the things he wanted to do with his life and he had proven to himself he could do almost anything.” Mensch and his best friend and coworker, Rod Orgill, soon became bored and started missing work. “Rod is a magical soul. He didn’t want to design any more chips there because he realized Chuck’s agreement wasn’t going to work out,” says Mensch, referring to the promised employee ownership plan. “We would frequently go out for lunch at the local pub and have turtle soup and we’d have a beer and then we’d have another beer and then a pitcher or three.” With all the hype surrounding the 6502, Peddle had little time to get his team started on a follow-up project, such as a 16-bit version of the 6502. “We didn’t have a concrete ‘This is what we are going to do next, let’s go get them,’ plan because we were so busy trying to get the first thing circled,” says Peddle. “We didn’t really do anything on the 16-bit,” says Mensch. “There was never any real discussion.” According to Mensch, the reason Peddle did not consider a 16-bit version is because he still thought of the 6502 as a microcontroller for mechanical devices, and a 16-bit microcontroller would be overkill. Mensch began planning a new direction for his career. “For a year before I actually left, I was thinking about leaving. I wanted Will Mathys and Rod Orgill to work with me,” he reveals. “I wanted to start my own company on July 4, 1976. It was going to be founded in Pennsylvania, in Valley Forge, on the second century anniversary of our country. It was idealism.” Mensch’s plans went awry one afternoon away from MOS Technology. “We’re at the local pub and Chuck didn’t know where we were. He would often look for us because we’d go there and get drunk, and come back and go home because we didn’t have anything to do.” At the time, Rod Orgill was still in mourning over the ing of his father. After some beers, Mensch made a less-than-tactful jest. “It was just a bad comment from me,” he explains. “I said something to him that related to his father that pissed him off.” Mensch had gone through a painful period following the suicide of his father,

which had toughened his emotions, so he assumed a layer of rawhide also protected the feelings of others. Although an extrovert, he was still somewhat socially awkward. Unfortunately, Orgill was still sensitive over his father’s death. “The next thing I know, he has his hands around my throat banging my head against the wall next to a dartboard. I’m thinking, ‘What’s going on here?’” As a result, Mensch’s dream of his own business started to crumble. “I had to mothball the business plan for the Western Design Center because nobody wanted to work for me,” he laughs. “Rod said ‘No, that ain’t going to work.’ So he’s out of there. Ray [Hirt] said, ‘No, I don’t want to do it.’ Nobody wanted to do it so I just said, ‘Well, it’s not the right time.’” Gradually the 6502 team left MOS Technology. “Most of them just said they missed the west coast,” says Peddle. “The whole thing was coming apart. Effectively the microprocessor team vanished all at one time. Mathys went to DEC. Orgill went to HP in Colorado. [Harry] Bawcom went somewhere. All of the people just went away.” One by one, the remaining employees blacked out the faces from the photograph on the wall. “All of them left in a period of about three months,” says Peddle. Soon, the picture began to look eerily black. The very people who would be capable of taking Commodore microprocessors into the 16-bit era were disappearing. * * * With most of his coworkers gone, it was up to Peddle to help resolve the lawsuit with Motorola. Today, the courts would dismiss the lawsuit immediately, but high-tech matters were unknown to the legal system at the time. The actual nature of the lawsuit changed since the original filing. “In their discovery process, they realized they didn’t have a leg to stand on other than patent violation,” explains Mensch. “It started out as a trade secret thing but went to a patent violation.” Motorola named all four inventors of the 6502 in the lawsuit: Chuck Peddle, Will Mathys, Bill Mensch and Rod Orgill. During the discovery process, Peddle felt confident Motorola’s lawyers would not uncover anything damaging to the case. Unfortunately, one of his engineers

from Motorola had failed to heed his instructions. “Mike James cost us months of discovery because he refused to give up his copies of the work he did at Motorola,” explains Peddle. “He had all his original work from the 6800 and hid it from Motorola when he left even though they and I both told him to come only with his shorts and his mind. He claimed that he was an artist and he kept the work as his original art.” The discovery of the chip layouts made it almost impossible for MOS Technology to prevail. “That was one of the key things they lost the lawsuit on,” recalls Seiler. Part of MOS Technology’s strategy was to use the 6501 as a sacrificial lamb. Mensch claims they told Motorola, “We want to steal your sockets. We want your sales team to build the market for us and we’re going to replace your microprocessors.” He explains, “You usually don’t say that to somebody that’s already pissed off.” In fact, the defendants were hoping to use the 6501 as bait to make it seem like MOS Technology was offering up something valuable to Motorola in the settlement. It was an easy decision for Motorola. “Motorola didn’t like the fact we were going to rape and plunder them,” says Mensch. “We already knew we didn’t want to stay with the 6501. We already had the 6502 and that’s what people are buying anyway.” According to Peddle, the cancellation meant nothing to MOS Technology. “I didn’t mind throwing it away,” he says. “If they had called up and said we want you to kill the 6501, we would have killed it. They didn’t need to sue us.” Unfortunately, Motorola wanted to extract some money from the lawsuit too. “They never were able to prove any trade secrets [were stolen] but they found our N-channel device violated one of their patents,” he says. “In order to get the license to the patent, we had to settle the suit.” MOS Technology did not have money to fight the patent, so it agreed to the settlement with Motorola. “It was a big fucking lawsuit and ended up costing us a million dollars,” says Peddle. Peddle was reluctant to go along with the settlement. “We settled in such a way

that implied we had done more than we had,” he says. “It was a good business decision but it just pissed me off.” In the end, the courts settled Mensch and Orgill’s bet when Motorola squashed the 6501. “As it turns out, I won the bet, not because we weren’t stealing sockets, but because part of settlement was to remove 6501 from the marketplace.” The lawsuit had even worse consequences for Peddle’s personal computer, which he had been planning since late 1975. “It was obvious they were not going to keep their word on that. They called me in one day and said, ‘We don’t have the money.’” Peddle began looking for other companies who might be interested in developing a computer product, including Allied Leisure. “I talked to the guy who was the president of the company, showed him Bill’s [Digital Group] computer and said, ‘I’m convinced that we need to build a complete product and I know how to build it.’ This guy said, ‘If you want to do that, come down here.’ So I basically finished up what I was doing [at MOS Technology].” Peddle was now committed to Allied-Leisure and began restructuring his life around Florida. “My youngest daughter moved from Phoenix back east with me with the promise that we would move to Florida.” * * * Commodore was a longtime customer of MOS Technology, having purchased calculator chips since 1972. Steered by CEO Jack Tramiel, the company had sales of $49.8 million in 1974, but fierce competition resulted in a glut of calculators on the market, rapidly falling prices, and masses of returned product. In 1975, Commodore had a surplus of $6.5 million worth of obsolete calculators, which it had to sell off below cost. Although it had sales of $55.9 million for the year, it ended with a loss of $4.4 million. By 1976, Commodore became MOS Technology’s largest customer by virtue of having survived the calculator wars. Peddle recalls, “Jack effectively saved their bacon that year, because they weren’t doing that well, and Jack basically just bought processing time for his own designs.” This made MOS Technology dependant on Commodore for revenue. As the three owners would later learn, being dependant on Commodore also made them susceptible to Tramiel’s will.

According to Peddle, an acquaintance of his named Andre Sousan was responsible for instigating the MOS Technology acquisition. Sousan was corporate VP of engineering and a board member of Commodore Electronics Ltd. (the operating subsidiary of Commodore International). He had been with Commodore since 1971, and frequently traveled to Europe and Asia for marketing and manufacturing purposes. “Sousan spotted two things,” explains Peddle. “One, they were the right partner for them to make the calculator stuff. Two, Andre felt I was on the right track with the microprocessor.” Tramiel was an easy sell, since the acquisition of MOS Technology would help achieve his master plan. “His primary drive was he needed vertical integration,” says Peddle. “So in effect, what started out as ‘Gee, we’d like to buy a bunch of chips and then maybe you can process some chips for us,’ went to, ‘How would you like to just let us buy you?’” Bob Yannes, a MOS Technology engineer, speculates how Tramiel took control of the company. “He ordered a bunch of calculator chips from MOS and apparently there was a bug in it,” he says. It was not a major bug. “I think they did one calculation wrong. You took the cosine of some bizarre number and it came out 0.000001 off.” Yannes continues, “Then MOS got sued by Motorola for possible design infringement with the 6502 over the 6800. So they were embroiled in that, spending money on lawsuits. Then, as I understand it, Jack came in and said, ‘Okay, well I’m not going to pay for those chips. If you want, I’ll buy your company from you for almost nothing.’” The questionable methods Tramiel used to obtain MOS Technology have become legend, making it difficult to separate fact from fiction. But is the legend true? “Having met Jack and dealt with him on a low level, I would not at all put it past him,” says Yannes. Peddle gives a similar explanation. “Jack is the kind of guy that will look for any lever,” he says. “It is undoubtedly true he was squeezing them on payables. Whether they found a bug or didn’t find a bug—whatever. As a standard practice, Jack would rack up a bill and then squeeze people.” If Tramiel was doing business with your company, he needed your parts and

supplies. However, he also wanted vertical integration, so he wanted not only your parts but also your whole company. Normally, if Commodore became beholden with debt to a supplier, the pressure was on Commodore to repay it. However, Tramiel did not play by normal business rules. “If Jack owed you a lot of money, you were in trouble,” says Peddle. Tramiel realized the best way to acquire MOS Technology under his was to hold back payments, making MOS Technology starved for cash. With money running out, MOS Technology would be unable to pay its employees, making them amenable to almost any deal. “He would sometimes use it for leverage,” recalls Peddle. “He’s a very clever, very ruthless businessman.” Kit Spencer, a UK marketing manager at the time, believes Tramiel did not intentionally withhold payment to MOS, although he definitely took advantage of the situation. “Commodore was in a horrendous cash squeeze at that time,” he recalls. “Most people in the calculator business were going bankrupt and we were close to it too. At this stage, we were already cash desperate around the world. We were sending money from the UK to the US and Japan to meet payroll because we were reasonably profitable [in the UK]. The rest of Commodore was in a desperate state.” In order to have enough cash for the acquisition, Tramiel had to call in some favors from his European managers. “I getting a phone call in the UK late one night from Jack. He didn’t call me up that often. He said, ‘Kit, we have to have another half a million dollars by the end of the month. I don’t care what you do, we have got to have cash.’” The sudden request caught Spencer off guard, but he soon understood the urgency. He recalls, “Jack understood the business. He said, ‘There will never be an opportunity like this to buy a chip plant at this price. And to be where we need to be in two years we have to do this deal.’” The next day, Spencer surveyed Commodore’s books and found a possible solution. “I looking down our inventory. We were in the watch business as well. I had about 25,000 digital watches in inventory, and they don’t sell too much in August. They were a Christmas item.” Spencer went to his most promising retail customer. “I ringing up the head buyer for Boots, which is a 1000 store chain in the UK. They were

probably the biggest market for that in the UK. I said, ‘Can I come and see you? I’ve got a proposition.’” Kit was experienced in negotiations with purchasing managers and knew he could not expect any big favors without offering something in return. “I asked him to take every single watch we had into inventory in August saying, ‘You’re bound to sell them by Christmas. We’ll give you good now and a special promotional deal around Christmas to promote it if you take them now.’” The request made the buyer suspicious, and he asked, “Are you going bankrupt?” “I crossed my fingers underneath the table and said, ‘We’re buying a company and we need the cash.’” In fact, Commodore was on the verge of bankruptcy. Spencer laughs, “I didn’t answer the other half of the question.” Spencer immediately shipped all 25,000 watches to Boots. Because payment was not immediately expected, Spencer used a financial technique called factoring; he showed the signed order from Boots to the National Westminster Bank, and within the week, the bank advanced Commodore UK the cash. Unfortunately, it would be even more difficult to ship the cash to the US due to restrictive regulations in the UK. “I had a problem, because I couldn’t just ship cash to the States unless it’s against something,” he recalls. “If Commodore did go bankrupt and I had shipped money out of the UK without a valid reason, then I could technically be liable.” To legitimize the transaction, Spencer would have to buy product from the United States. “Tramiel used to say this stuff is like fruit—it just rots on the vine real quick,” says Dick Sanford, the financial controller of Commodore at the time. “So you’ve got to be on top of new technologies continuously.” At the time, Commodore had an excess inventory of outdated calculators. “I calling up the US and saying, ‘What have you got in inventory that you can ship me that I can ship you a quarter of a million dollars for?’ They had a load of bum stuff. I said, ‘Don’t worry. Ship it all to me and I’ll take it.’ They shipped it and I transferred the cash to the US and basically helped make the cash for the deal to buy MOS. In a lot of corporations you couldn’t work like

that.” In September 1976, Commodore fully acquired MOS Technology. “He basically bailed them out by buying the company out from under them,” says Yannes. Although Commodore could not raise the full price of MOS Technology in cash, they made up for it in a stock transaction. “Jack bought the company for stock and some cash,” recalls Peddle. “The founders actually got reasonable amounts of Commodore stock, which were very depressed at the time.” According to Spencer, “We bought MOS Technology for three-quarters of a million dollars, which was unheard of for a chip plant, but everybody in the calculator and semiconductor business was going bankrupt. It was one of those bloodbath times. We were certainly one of those candidates who might have gone bankrupt.” When Spencer received his shipment of outdated Commodore products, he quickly sold them off at a massive loss in the UK. “Four or five months later Jack said to me, ‘Why did your profits go to hell this month?’ I said, ‘ a few months ago you wanted cash? I had to take all this rubbish inventory and sell it for a loss over here.’ He said, ‘Okay, that’s fine.’ That was typical Jack.” Now that Commodore had chip manufacturing capability, it could become competitive once again. “Commodore had the calculator line, so when they acquired MOS Technology, they morphed into more like what TI was or Hewlett-Packard was,” says Mensch. “They had their own manufacturing of the chips, so they could compete by getting the cost down on the product.” Did the MOS Technology owners receive a bad deal from Jack Tramiel? “Beauty is in the eye of the beholder,” says Peddle. “Considering the financial state they were in, and their inability to pursue what they needed to do without a significant amount of outside financing, they got a good price.” The relatively depressed price of Commodore stock at the time made the total value of the transaction small. “It wouldn’t have been much but it didn’t need to be much,” says Peddle, estimating that it would have been worth a few hundred thousand dollars in 1976. “They got an intangible for their intangible. They had a processing capability, they had a factory, they had everything else, and Jack had a company that was public. In effect, he offered them a way to exit from their investment in a public way.”

Over time, the few hundred thousand dollars of stock had the potential to turn into millions. “I think one of the reasons they all held it was because they did not get ed stock,” ventures Peddle. “They were forced to hold it for two years, which turned out to be highly, highly advantageous to them.” The engineers at MOS Technology did not fare as well as the owners. Some of the ex-Motorola employees received the promise of stock when they first ed MOS Technology. “They cashed out all of us that might have had some ownership in the company,” explains Peddle. “We just got cash. I was not a real owner of the company, except for a deal that we were going to get paid off.” According to Peddle, the deal with Commodore occurred without their knowledge, which caused some friction between the engineers and the owners. “This was done without all of us having a clue,” he says. When Tramiel bought MOS Technology, he could not have imagined the riches it would bring to him and his company. The acquisition had been undertaken solely to achieve vertical integration of calculator products, but it was about to transform Commodore into a true computer company. * * * In 1976, Peddle developed a third high-profile source for the 6502. “After we were bought by Commodore, we were approached by Rockwell to buy rights to the product,” he recalls. “Rockwell came along and gave us a bunch of prestige.” With a third, stable source for the 6502, hardware makers were more likely to use the 6502 in their products. Rockwell released a “KIM-like” single-board computer called the AIM-65. The 6502 system was similar to the KIM-1, but contained a 20-digit LED display and a tiny thermal printer mounted directly on the motherboard. It also included a full-sized keyboard, which attached to the board by a ribbon cable. The system sold for under $500. Synertek, the existing second source for the 6502 microprocessor, also released a KIM-1 clone called the VIM. Fearing confusion with the cleaning product, the company soon renamed it SYM. The system contained a more extensive operating system and more memory than the KIM-1. Overseas, a Dutch company named Elektor developed a European KIM-1 clone, called the ElektorJunior. In , a series of systems called the ALPHA were developed.

Commodore actively marketed the KIM-1 system against their competition, with ments appearing in 1978 asking, “Honestly: How many reasons do you need to make sure your next microprocessor is the original, genuine KIM?” The ments appeared in magazines like Kilobaud, but curiously, all Commodore ments in Byte magazine ended and would not reappear for almost a decade. * * * Before the acquisition by Commodore, management at MOS Technology had rejected Peddle’s personal computer proposal. “I was getting ready to bail and go to Allied Leisure to make my own computer when Commodore bought the company,” says Peddle. Peddle had his desk packed up in a cardboard box and was ready to leave when Commodore executives began arriving on the premises. “When we acquired MOS, Chuck was basically kissing the girls goodbye in the front lobby and he was leaving,” recalls Dick Sanford, Commodore’s financial controller. “I introduced myself and he told me he was an inventor and an engineer.” Sanford listened as Peddle explained the 6502 processor and the KIM-1 computer. “I said, ‘So you’ve invented a computer. We’ve had computers all over the place here, you know.’ He said, ‘How much time do you have?’ I said, ‘I have all the time in the world.’ He literally went to the front door and picked up a cardboard box and said, ‘Come on, let’s go up to the lab.’” Sanford was about to learn Peddle’s vision of the computer industry. “We went up to the lab and he opened up this box with bubble wrap and out comes a motherboard,” says Sanford, referring to the KIM-1. “I look at this and said, ‘What is this?’ He said, ‘This is a computer.’ So he plugged it in and up came blackjack. I said, ‘What are you going to do with this?’ He said, ‘How much time do you have?’ I said, ‘I have the afternoon.’ And he told me his dream about the personal computer world that had not been born yet.” Peddle described the advances and breakthroughs which would soon happen with personal computers. “Apparently he tried to convince management of MOS at the time that this was a fantastic piece of equipment or at least it had a future, and I guess they were in financial difficulty and just didn’t take the project on.” Sanford realized Tramiel should be aware that MOS Technology’s greatest

engineer was about to walk out the door with a potential product. “Tramiel was in the building somewhere, and I got Jack,” recalls Sanford. “Once I introduced Peddle to Jack, we chatted a while and said, ‘Hey this makes a lot of sense. Why don’t we put an agreement together?’” By mere minutes, the personal computer would now stay with Commodore. “Chuck calls the wife and says, ‘Hey honey, get the Mayflower truck and stop it. We’re staying.’” To Peddle’s surprise, his old acquaintance Andre Sousan now worked at Commodore as VP of engineering. “We had known each other from that period, although we were not bosom buddies or anything,” says Peddle. Sousan possessed a wealth of knowledge, with a doctorate in physics from the University of Paris. He also knew the practical side of engineering, having received a master’s degree in electronics from the Superior School of RadioElectronics in Paris. Previously he had worked for five years with Texas Instruments as a marketing manager before ing Commodore in 1971. Peddle recalls the soft-spoken Sousan. “He was a really gentle, intelligent Frenchman. He had long wavy hair and a moustache. He was an older guy with a stunningly beautiful wife and they lived in Palo Alto.” “Andre called me up and asked some questions,” explains Peddle. “He said, ‘Get on an airplane, bring your wife, come out to the west coast, and let’s talk about it.’” “So we flew out and I told Andre my idea for the computer. Andre said, ‘I’m convinced we can go do this product. Oh, by the way, we have our hands on a spec from Radio Shack for a similar product.’” Radio Shack wanted a computer system for hobbyists and it sent a specification to some of its suppliers. The company hoped one of them could design and manufacture a computer for its customers. According to Peddle, “The Radio Shack sales manager came down and said, ‘Sign us up for a whole bunch of calculators for the Christmas season. By the way, we’re going to give you a bid at CES for the computer.’” Tramiel was understandably wary of the unproven market. Instead, he saw it as an opportunity to sell more calculators. “The plan was, we would make this

computer for Radio Shack and that would get us instant distribution to sell calculators,” says Peddle. Peddle had met other tycoons in Silicon Valley, but none so magnificently coldhearted as Tramiel. “Jack was a terribly smart guy,” he says. “He picked up a lot of things that the rest of the world didn’t. He was one of the best minds I’ve ever met, but totally ruthless.” At the time, Tramiel focused on scientific calculators. Kit Spencer explains, “Calculators had gone from simple four-function calculators to memory calculators. Then you got scientific calculators and then you got keyboard programmable ones. The next step was the HP65 with a little card reader to make it more programmable. We were asking, ‘What’s our product development doing on this?’” Sousan and Peddle pitched the product as an evolution of the calculator which would sur the HP65. “In our meeting, Chuck and Andre Sousan said, ‘No, our next generation is going to jump that. Instead of a card reader, we’re going to have a cassette deck. Instead of an LCD display, we’re going to have a TV monitor.’ And that was the concept of the PET,” recalls Spencer. “It was very much an evolution from the same technologies from the calculator business.” Peddle described a computer small enough to fit on a desktop, yet with more than enough power for an individual . However, Sousan took the lead in convincing Tramiel. “Sousan was pretty persuasive that we should go do that,” recalls Peddle.

Jack Tramiel, founder of Commodore.

The idea appealed to Tramiel and he listened attentively. “Jack understood where the business could go, but he was not a technical man,” recalls Spencer. Undercutting the competition was in Tramiel’s blood, and this was an opportunity to undercut the biggest competitor of them all: IBM. Peddle received the response he was looking for. “They basically told me, ‘We’ll set you up in the computer business. It’s your business and you can run it, because it’s not part of our calculator business. If it goes, it’s yours. We want to own it, but depending on how well you do with the computer, we will give you significant stock.” Before shaking hands, Peddle made a special deal concerning his wife, Shirley. “She got to travel with me. That was one of my deals when I went to work for them,” he explains. Under the deal, Commodore would pay for her transportation and lodging. “She was totally involved with everything I was doing at MOS Technology and early Commodore. She was free help for Commodore and Jack didn’t pay her anything.” “I said to Jack, ‘Okay, we got a deal.’ We shook hands on the deal. They moved me across the country and gave me a loan so I could buy a house in California.” Chuck Peddle would now build computers for Commodore. Peddle returned to the east coast to inform his employers he was leaving for Commodore. He knew they would be unenthusiastic, since they had lost almost the entire 6502 team already. “I went back to MOS Technology and said, ‘I’m going to the west coast to do a computer.’ By this time Jack owns the company, so they’re not going to tell him to fuck off,” he says. Peddle’s fellow engineers at MOS Technology the move. “Chuck went to the west coast because that is where the power was at,” explains Al Charpentier. “Jack lived in California, so where would Chuck go?” Unfortunately, Peddle had to break his promise to his youngest daughter to live in Florida. “She gets in my house in Pennsylvania and I tell her, ‘No, we’re moving to California.’”

In October 1976, Peddle, his family, and his three cars traveled to California. “I just packed up, put one car in a van and put another car behind it and my wife and daughter drove another one. We packed up and drove across the United States.”

CHAPTER 5

The Prototype 1976

In 1976, Commodore underwent a major restructuring. Company chairman Irving Gould created Commodore International Limited (CIL) in the Bahamas. Underneath CIL were the divisions that made up the Commodore International army: Commodore , Commodore Canada, Commodore Switzerland, Commodore Japan, Commodore Hong Kong, Commodore Business Machines (US), and Commodore UK.[1] A general manager headed each country. “Commodore was a European company when I ed them,” says Chuck Peddle. “They were not doing that well in North America from a calculator standpoint, but they were the definitive player in Europe.” Kit Spencer had been with Commodore UK since 1974, when the brutal calculator wars threatened Commodore’s existence. Spencer often received job offers from Commodore’s competition, but he always turned them down. “You don’t jump ship when things are going badly; you try and keep it going,” he says. “When the times are tough, that’s when you start earning your salary.” At the time, he was in charge of the UK marketing operation. While Commodore watched most of its calculator market shrink in the US and Japan, he steadily built up market share in the UK. “The UK at that time was the most profitable of the Commodore companies and probably the biggest sales too–more than the US,” he says His innovative dealer promotions and retail advertising were largely responsible for the success. The Commodore brand name, which Spencer had steadfastly built up over the years, was a significant asset in the UK. Commodore had an

excellent reputation for quality, innovation, and low prices. In fact, most people either owned a low-cost Commodore calculator or knew someone who did. After the MOS acquisition and plans for a full computer system, Spencer knew he would stay with Commodore to see through its next product. “The acquisition of MOS technology was certainly very key. You could see the possibilities and I had faith in what could happen.” Commodore employee Dave Haynie explains the company structure. “Commodore was run like a cellular company,” he says. “Each cell did its thing, and ran fairly independently of the parent, CIL. This is why every company did marketing differently.” The international general managers met four times a year with Jack Tramiel to discuss marketing and strategy. They enjoyed almost complete autonomy in their efforts to market products. Each country had control over its product distribution system, advertising, and packaging. The cells were so autonomous they decided which products they would market. If Commodore made a product and the cells did not feel strongly about it, they could ignore it. “To get their product, each marketing company places orders, and Commodore fills them as best as they can,” explains Haynie. “Commodore says, ‘Here’s a new product!’ The marketing departments would order it or not order it. It was survival of the fittest and I think it worked well.” Products came from one of four divisions: consumer products (calculators and watches), components (semiconductor chips and microprocessors), metal products (office furniture), and systems (computers). Tramiel wanted complete vertical integration, including the reselling of Commodore calculators at the retail level. To test this concept, Commodore opened about a dozen Mr. Calculator stores, which were to become authorized resellers of Commodore products. These stores were small high-tech stores located in strip malls throughout the United States. The chain of Mr. Calculators would remain a subsidiary of Commodore International. Commodore Japan had a special role within the Commodore empire. “Irving got Jack to believe he had to manufacture in Asia,” says Peddle. “Jack set up the company in Osaka. He was getting all of his product made in Japan. He was buying the printers, the screens, and particularly keyboards in Japan.”

Months earlier, the Japanese operation emerged from a corporate shakeup. “There was a company called CBM in Japan under Commodore Japan,” says employee Yash Terakura. “CBM was actually the engineering part of Commodore Japan, deg all the calculators for Commodore.” A general manager named Mr. Fakagi headed CBM. “In Japan, everything goes by the last name, so there is hardly a first name basis,” says Terakura. In fact, due to this social convention, coworkers rarely knew each other on a first name basis. Fakagi was a traditional Japanese manager, and he demanded obedience from employees. All decisions came from the top, leaving very little free choice for employees. Approximately five engineers worked under Fakagi: Aoji, Fujiyama, Nishimura, Tokuda, and a 30-year-old engineer named Yash Terakura, who clashed with Fakagi and was subsequently terminated. When the calculator business began faltering, Tramiel closed down the Osaka offices. “Pretty much what they did was Jack and John Calton kicked everybody out,” says Terakura. Although Tramiel was unhappy with the management, he still valued his Japanese engineers. “They moved Fujiyama, Nishimura, and Aoji to the United States and they made one small Japanese group developing calculators and other stuff based on Japanese parts and Japanese information,” explains Terakura. “At that time Fujiyama was the head honcho of that group. He was a very good engineer—a typical Japanese-style engineer.” According to Peddle, “Fujiyama and Aoji were in our factory in Palo Alto, to coordinate with the people in Osaka. So they were moved over from Japan to the US to act as the bridge, because they could talk Japanese.” Tramiel wanted to reopen the Japanese division in Tokyo under more Westernized leadership. He chose a new manager named Tony Tokai. “Jack and Tony go back a long way, back to when Jack was still doing typewriters and all that stuff,” says Terakura. At the time, Tokai was working for the Ricoh Company, which had recently sent him to its California offices. Since both men were in the typewriter industry, their paths often intersected and they got to know one another. “Somehow Tony and Jack hit it off pretty well and they kind of got to be friends rather than just business acquaintances,” recalls

Terakura. “Jack’s kids were small, so Tony Tokai was babysitting them and everything, and Jack Tramiel started to trust Tony Tokai.” When Tramiel explained to Tokai that Commodore was entering the computer industry, Tokai saw an opportunity for himself. “Tony said, ‘Let me do it.’ And Jack said, ‘Go ahead and do it.’ Tony has no background in computers or anything, but he has a business background.” The new Commodore Japan would also attempt to market products in Japan. “That’s when they started the new Commodore Japan and they decided to move the office from Osaka to Tokyo. Tony Tokai was the successor of the old Commodore Japan.” During the Pong craze, Commodore Japan had found a company to manufacture Pong machines. “Tony sourced them and put them through to Jack,” says Spencer. Spencer ended up selling the machines off in the UK. “It gave us a taste of what that market could do.” In the United States, the general manager was not a defined position. “Tramiel would be considered the general manager,” says Dick Sanford. Sanford himself was a tough executive who began his career as an ant with Arthur Anderson until Commodore snatched him away in 1976, right around the time Commodore International established its headquarters in the Bahamas. “I believe the idea came from Arthur Anderson,” says Sanford. The 33-year old Sanford started at Commodore as the company financial controller, responsible for producing a healthy cash flow. “That’s what he laid on me,” says Sanford. “How do I make payroll tomorrow? How do I make payroll next week? How do you get the vendor to keep the line going? How do you do that? That’s what you’ve got to learn, because if that line shuts down, a lot of people don’t eat next week.” Sanford quickly rose within Commodore. “Dick was very capable and for quite a long time was the nearest Jack had to a second in command,” says Spencer. It was his job to act as the company hatchet man, laying off employees or management when the need arose. Sometimes he would ask managers if they wanted to resign, sometimes he would fire them outright if they were incompetent, or if there was a personality clash and Tramiel just wanted them to go, he would prepare them for a layoff. This meant setting them up for a fall, or

as one Commodore employee put it, “giving him enough rope to hang himself.” One way or another, if Tramiel wanted someone out, Sanford could arrange it in a legal manner. “The States was always a difficult area to operate in because too many people had Jack and Irving’s ear,” says Spencer. “It was potentially a bigger operation, but it was always a much more political area and less happy. We had much more stable management in Europe.” When Tramiel promoted Sanford to vice president of finance, he moved back to the east coast, where he was born and raised. “When we acquired MOS, I moved from the west to the east about nine months later.” Tramiel wanted Sanford closer to New York, where it was easier to talk to financial people. * * * Based on Chuck Peddle’s experience with the Dartmouth time-sharing system, he believed s needed the BASIC computer language to do anything useful with a computer. Peddle recalls, “These people wanted a computer they could use to write BASIC because they loved doing it.” When corporations began locking the doors to computer rooms, s had nowhere to program. “There was no way they could do it at work so they went out and bought their own machines just for that purpose. I knew there were at least two million people that had been trained that I could sell this machine to.” The inclusion of a built-in programming language seems quaint, but in the 1970s, it was perhaps the most important aspect of a personal computer because there was no software industry. s had to develop their own software, and BASIC was the simplest path. To acquire BASIC, Commodore would make a deal with a small company called Micro-Soft. At the time, Micro-Soft did not own an operating system. Instead, it sold programming languages, principally its well-regarded version of BASIC. A young Bill Gates led the company. Some people mistakenly believe Gates invented BASIC, but John Kemeny and Thomas Kurtz invented it in 1964 at Dartmouth College. Peddle finds the misconception amusing. “I knew BASIC when Gates was still in goddamned grammar school,” he laughs.

Bill Gates and Paul Allen were still new to business. Their company had grown to five employees and their BASIC language was quickly becoming the industry standard. Micro-Soft was located in Albuquerque, New Mexico.[2] In late 1976, Peddle received a phone call. “A guy calls me up from this place Micro-Soft and says, ‘I’ve got a BASIC for the 6502 and we’re not finding any customers. Would you like to start marketing it for me?’ It turned out this guy’s name was Rick Wyland.” Wyland had taken the original Intel 8008 code created by Bill Gates and, just for fun, converted it to the 6502 microprocessor. “He had done it on his own because the place that Bill had built up was kind of loosey goosey,” explains Peddle. When Wyland showed his ad-hoc project to Gates, he did not receive an enthusiastic response. According to Peddle, Gates reaction was, “That is such a shitty [microprocessor] because of the stack. I don’t want us to waste any more time on it. Get some money out of it and get out of it.” Wyland and Peddle continued talking back and forth until Peddle and his family began their trek to the west coast in late October. “I was taking the southern route because it was getting ready to snow up north. So I organized to come through Albuquerque and sit down with Gates and Wyland in their two-room office over a bank.” Peddle had a positive first impression of the young Gates. “Gates is a tough competitor and he’s been pretty ruthless about stuff, but down underneath he is a real good guy.” Peddle told Gates and Wyland the features he wanted in his BASIC language. “I’m building a computer. I want to build BASIC into ROM. I want to make it very friendly and there’s some things in BASIC I have to extend: I need an I/O structure and I need the ability to draw characters on a screen.” Gates replied, “If you can afford it, we’ll do it.” Peddle was impressed. “Gates is a very honest businessman. He made a commitment that we would have a working BASIC.” With an informal agreement between Micro-Soft and Commodore in place, Peddle and his family continued their journey.

* * * Once in California, Peddle reunited with his west coast 6502 sales manager, Petr Sehnal. They had very little time to complete the prototype computer. “We’ve got just enough time to do the computer before the Consumer Electronics Show (CES), which is in January,” he recalls. “But we have to start with some equipment and we have to kind of hack our way to it.” To complete the computer in two months, they decided to start with an existing piece of hardware. Peddle called on Steve Wozniak and Steve Jobs. “We kept in with them because they were a 6502 customer,” he recalls. “One of the ways to get it done quickly was to start with the Apple I and do our own version of the machine that met my criteria.” Through 1976, Wozniak gradually improved on the Apple I. “His definition of a machine appealed to the people that Woz knew; the hobbyists and the guys who were excited about doing something better. Graphics was one of the main things,” says Peddle. In September, Wozniak improved the display to allow color graphics. Color was nothing new at the time and even the lowly Altair had a color display adapter called the Cromemco Dazzler. Wozniak’s adapter could display four colors at once—two better than black and white. The text was muddy, with color streaks seeping into the white characters. Wozniak also created six expansion slots for his improved Apple. This brought it up to date with the functionality available to other computers, including the Altair and the KIM-4 expansion module. In October, Peddle approached Apple in the hopes of acquiring Wozniak’s technology. “We tried to get them to sell us the Apple … as the basis for our PET,” says Peddle. The person negotiating on behalf of Apple was Steve Jobs. The two partners came to Commodore headquarters with the intention of selling their fledgling company. “Andre sets up a meeting with Apple to have a discussion about how we could get to the CES show together,” recalls Peddle.

He and Andre Sousan told the two Steves, “If you work with us we can put together a deal, we can go to CES, and we can have a product that meets our new spec.” During their discussions, Peddle and Wozniak had several disagreements over the design. “There were two fundamental things that blocked it,” explains Peddle. “One, Woz had written this really nice little thing called Sweet-16, which made it real easy to program in machine language.” Peddle liked the machine language assembler, but he believed it was too complex for ninety percent of computer s. He believed s needed a BASIC programming language, which was already in development by Micro-Soft. “Woz and I had a big argument about it.” The second disagreement was with the color. “They felt the thing that their customers wanted was some audio and color graphics,” recalls Peddle. Their instincts were accurate, but Peddle felt the machine should be simpler. “My argument was, color graphics are wonderful, but people can’t just instantly pick up and use it. If I do character graphics and teach them how to use it, everybody can do their own pictures in a hurry.” Although there were disagreements, Wozniak and Jobs were still willing to sell, as long as the price was right. According to Wozniak, “We went to Commodore and talked primarily with Andre Sousan. Steve Jobs was trying to talk Commodore into buying the Apple II for a large amount, like hundreds of thousands of dollars. … Steve Jobs also wanted Commodore to hire us along with the proposed deal. The deal was never on paper and never concrete, as to how much.” “The discussions never got beyond a meeting with Jack and Steve,” says Peddle. “Jack’s view was that Steve wanted much too much for the company and in the fall of 1976 he was right. I him laughing about Steve Jobs and his view of his company’s position.” Tramiel was willing to purchase Apple, but he wanted the lowest price possible. To do that, he put pressure on Jobs by refusing his initial offer. “Basically Jack decided to go along with it, but he tried to squeeze Steve,” explains Peddle. Steve Wozniak tells a different story. “We were told that Chuck wanted to do his own thing and that he could do better than us at reaching the cheap needs of

customers. Jack Tramiel, the Commodore president, was talked by Chuck into shooting for cheapness. We were told that Chuck Peddle was the instrumental force in ing on this deal.” However, Peddle denies turning down Apple. “Petr [Sehnal] and I were more interested in our short term problems of getting prototypes done. I was never asked, nor gave an opinion, except that using the existing Apple I board would help us get the original prototypes done.” As it turns out, a retired semiconductor executive and millionaire had his eye on Apple. “What Jack didn’t know is that right about that time Mike Markkula had fallen in love with the idea of doing his own computer,” explains Peddle. “He effectively came out of retirement from Intel and he threw money at them at the same time Jack was trying to squeeze them, so Jobs just walked.” If the transaction had gone through, the world would be a different place today. “The incident didn’t change either company, but had they merged, it would have changed the industry,” says Peddle. Commodore possibly would have released the Commodore Apple and Wozniak would have become one of Peddle’s engineers. More importantly, Commodore would have eliminated an early rival at the bargain price of only $100,000. However, Apple did not look like a formidable rival at the time. Dick Sanford believes two separate companies was better for everyone. “You know what, it wouldn’t have worked anyway,” he says. “It would have been a tough play with Jack with those two guys. I just don’t think the personalities would have clicked.” * * * Tramiel was always looking for new products to diversify Commodore’s calculators and digital watches. “When Jack took over MOS, he wasn’t interested in microprocessors,” says Bill Mensch. “He was interested in calculators. Then, because of video game popularity, he was trying to figure out how to do a video game.” With MOS Technology’s success in selling the complete chipset to Atari, they decided to create a graphics chip in the hopes that Atari would purchase it too. “The [Atari] 2600 chipset that made that videogame system had all of 128 bytes,” recalls Bill Seiler. “That’s all the memory it had. Atari was moving onto

bigger and better things with more horsepower. So MOS designed a thing called the VIC: video interface chip.” The inspiration for the graphics chip came from designer Al Charpentier’s disappointment in the Atari 2600. “The 6502 in that thing was actually running the display,” explains Seiler. “The chip would only display one scan line; that’s all it could do. So the next scan-line had to be set up by the 6502. As the picture was coming down the screen, the 6502 is very, very busy setting up each scanline. When it got to the bottom of the screen and the CRT was flying back, that’s when it actually played the game. So it only played it on the refresh time of the display, every 60th of a second.” The engineers began to figure out how to do audio and video graphics. Bill Mensch proposed using his new 6522 VIA chip, but he met with resistance. “Chuck didn’t think it was possible to create the timing for driving the black and white TV video and audio with just the 6522,” he says. Mensch assembled a proof of concept, a game he called Barn Storming. The prototype used a 6502, a 6522 and some memory chips, all running at 4MHz. Mensch loaded the game into memory using paper tape. “As I recall, it only took a couple of weeks to design, wire the board, and write the software of a bi-plane diving around a field with a cow. Wherever the plane hit the barn, a piece of the barn was removed. All the while, a clicking sound could be heard as the plane’s engine sound.” Although Mensch proved his point with the 6522 chip, the quality of the sound and video was not high enough for the marketplace. “Chuck actually didn’t think I could get anything to work at all and when I did he didn’t really care and was not impressed,” he recalls. MOS began looking into a dedicated video and sound chip. A young engineer named Albert Charpentier began work on the chip. Charpentier was born in Philadelphia in 1952. Since his youth, he wanted to be an engineer. “I was fascinated from the age of seven about how the picture on the TV arrived in my home,” he recalls. “I wanted to understand how you control a TV and that ultimately led to the video chips I designed.” Charpentier attended the University of Pennsylvania in the early 1970s, but there were no courses in semiconductor design. “There was semiconductor physics

and stuff like that but nothing like VLSI [Very Large Scale Integration]. It was all being developed as we did it.” In 1975, he graduated with a Bachelor of Science in Electrical Engineering. Immediately after graduation, he began work at MOS Technology. Engineers John May and Don Schneider mentored him in semiconductor design. “The engineering staff at MOS continued my education. They basically taught me how to go about the process.” Charpentier spent his early years at MOS Technology deg calculator chips, always veiled in his own cloud of tobacco smoke. “He always had a cigarette in his mouth, even if it wasn’t lit,” recalls Bob Yannes. “Sometimes he’d have lit cigarettes in three or four ashtrays scattered around the office.” His boss, Charles Winterble, was impressed with the chip designer. “Charpentier was a rail-thin guy and a chain smoker. He was just a hyperactive kind of guy. He was also a clean-cut guy. Charpentier would show up frequently wearing a tie to work.” During the Pong videogame fad, MOS Technology helped companies develop semiconductor chips. “MOS Technology had made video game chips for a couple of companies, like Coleco,” says Yannes, referring to the Telstar game console. At the time, most of the electronics resided in the game cartridge. “They looked like game cartridges that you plugged into the game [module], but in fact the game module itself was just a power supply and an R/F modulator. The game cartridges weren’t software; they were whole custom chips to play that particular game.” Charpentier found the game design inefficient. “Al basically thought that was a pretty crazy way to do that,” recalls Yannes. “Why not have a microprocessor and a video chip in the box, and the cartridge could just be software, which is where Atari ended up.” After the MOS Technology acquisition, Charpentier worked for Tramiel. “I liked him,” says Charpentier. “He had a tough way about him and he was gruff. He had some pretty big flaws, but if you could look past those flaws, he had a lot of things he could teach you.” Charpentier offered to design a video chip for Commodore. “When Commodore took over [MOS Technology] in late 1976, that’s when I basically made the

proposal. I started on the chip in 1976 and didn’t have it working until one year later.” The video chip would become part of the 6502 family of chips, so Al gave it the numerical designation 6560. He called it the Video Interface Chip, or VIC. Incredibly, Charpentier had no experience with video displays. “He told me the reason he did [a video] chip is because he wanted to learn how TVs worked,” says Yannes. Bill Mensch was MOS Technology’s microprocessor manager at the time and he worked on the video chip in the early stages of development. “Al and I worked together on what became the VIC. I reluctantly contributed to it, answering questions or suggesting things to him.” Not all of the features Charpentier examined would make it into the final chip. “Al Charpentier was also looking at some tricky circuitry to have sprites and characters and things,” says Mensch. The VIC displayed black, white, red, cyan, purple, green, blue, and yellow. It could also display bitmapped graphics if the chip had eight kilobytes of memory available. The chip even had the ability to work with a light gun or light pen. Charpentier planned to add sound synthesis to his video chip too. According to Chuck Peddle, he knew Atari was developing a more advanced chip at the time. “We kind of kept telling them, ‘Don’t bother because you’re not going to be able to compete,’ because we had a relatively good insight into the Atari infrastructure,” he says. Unfortunately, a non-disclosure agreement with Atari prevented Peddle from telling Charpentier what he knew. “We weren’t allowed to communicate it to them.” It was for the best. Charpentier continued working on the chip through 1976 and into 1977. * * * Without the Apple computer to use as a foundation, Peddle had to look elsewhere. “My buddy on the west coast [Petr Sehnal] was working with a customer deg some piece of equipment that was small enough to fit in the case we wanted,” recalls Peddle. “It had a 6502 in it and enough other stuff in it that we could hack it around and make it drive the screen and run the keyboard.” What kind of hardware was it? “I don’t ,” says Peddle. “We started with a card that they were selling for something like a machine controller, I

think. We just adapted it.” When pressed further, Peddle thinks it was, of all things, a computerized sprinkler control system. “It may have been controlling solenoids for watering your lawn. It was a little general purpose thing that was designed for that kind of control application.” Peddle had been nurturing his idea of a computer system since his days at GE. He wanted to build a microcomputer system with all the elegant functionality of the Dartmouth system. “I’m trying to build a product people will perceive as professional, as opposed to hobby,” says Peddle. “I wanted them to be proud of what they bought.” He wanted a mainframe computer without the mainframe. Physically, the computer would resemble a mainframe terminal. Peddle also received inspiration from Hewlett-Packard. “HP was selling a BASIC computer for $5,000 or something like that, with an IEEE interface on it.”[3] He decided to add an IEEE-488 interface on his computer. “I was trying to make everybody believe we were an HP but at $500.” Peddle modeled his system software on the HP 2000 line of timesharing minicomputers. These systems came with BASIC integrated right into the operating system. “The definition of the product was lots of capability from a standpoint—BASIC in ROM and some demonstration software and manuals,” explains Peddle. “It was a system that worked when you plugged it in and you never had to think about it. That was totally different from what anyone was shipping at that point in time.” Peddle and Sehnal worked out of the tiny offices attached to Commodore’s LED and LCD plant at 901 South California Avenue in Palo Alto. “Jack and Andre [Sousan] had an office there,” says Peddle. “At this time Jack was only tirekicking in [the computer] business. His focus was on turning around his calculator company.” The PET engineers moved into the cramped quarters alongside the existing calculator engineers from Japan, and their lead engineer Shiraz Shivji. “They had this little building on California Avenue,” says Seiler. “They were building LCDs, watches and calculators.” As part of his vertical integration strategy, Tramiel had recently acquired an LED factory for manufacturing the glowing red calculator displays. “The LED technology was purchased from a company called Optical Diodes Inc. and used

in the LED calculator line,” explains Commodore engineer John Feagans. At the time, calculator technology was transitioning to the more efficient LCD displays. “They had a clean room where they were making LCDs for watches and calculators,” recalls Seiler. Decades later, he still re the stench in their work area. “It was stinky. The fluid they put in LCDs is really smelly.” Rounding out the team were two Japanese calculator engineers, Mr. Fujiyama and Nobuo Aoji. “I always called him Mr. Fujiyama because that’s what Jack called him,” says Peddle. “Fujiyama was more of a mechanical guy than an electrical guy, though he had both skills. He helped me with the packaging a lot. [He was] just a spectacular little Japanese guy. Fujiyama just made things happen.” “Aoji was the technician,” says Peddle. “Aoji was a great tech and really dedicated; just really nice guys to work with.” Petr Sehnal, who was most familiar with the sprinkler system controller, was crucial in assembling the prototype computer. “Petr had really good mechanical skills, because he was an ex-mechanical engineer, as well as an electrical and computer science engineer,” says Peddle. Peddle knew it would take time to develop a case for the computer, so he ed Larry Hittle, the man who fabricated the KIM-1 line. In early November, Hittle created some prototype cases. “We knew an industrial designer and we had a couple of buddies who did some sketches and helped put the thing together. We ordered two wooden prototypes of the product we wanted, nicely shaped—a real pretty box.”[4] A Courier terminal inspired the shape of the computer case. “If you look at it pretty carefully, it’s a knockoff of a terminal,” its Peddle. “I would say it looked a lot like a Courier system. It has very similar lines.”

Creating prototypes at Commodore’s 1976 headquarters.

Commodore attempted to associate their computer with the future through the case styling. Molded-plastic epitomized the future in the 1970s. The look was evident in everything from the Jetsons, to George Lucas’ THX-1138 and Woody Allen’s Sleepers. One of the most controversial design decisions originated with Tramiel. “I’m having trouble making a regular keyboard fit in the case,” explains Peddle. “I’m wandering around and Jack is talking to the Japanese engineers. I wandered in and showed Jack the problem.” According to Peddle, Tramiel asked him, “How do you know they really want a typewriter keyboard?” Peddle replied, “Well, I really don’t.” Tramiel said, “I’ve got a calculator here that we built that’s not going anywhere. You can have that keyboard. It’s done and it’s yours.” Tramiel clearly wanted to use Commodore’s vertical integration to supply a keyboard, rather than purchasing a professional keyboard from another supplier. “It’s very indicative of Jack’s thought process,” says Peddle. “He was willing to gamble on a new idea.” Against his better judgment, Peddle decided to follow his boss. “I shut up and built it the way he suggested.” The keyboard design came about under guidance from Mr. Fujiyama. “Mr. Fujiyama was a real serious short little Japanese guy,” says Seiler. “He was really good. He stuck together a couple of calculator keyboards and sent it off to Commodore Japan. They went off and somebody made us some prototypes.” The engineers delivered a rubberized 69-key keyboard. “We called them chiclet keyboards because the keys were Chiclet sized,” says Seiler. “They had these

little metal faces on them that had the alphabet character and the graphics characters on them.” It used the popular QWERTY layout, along with uppercase and lowercase characters, plus a numeric keypad on the right side. “It was very calculator like; you couldn’t touch type on it, but it fit in our case,” says Peddle. From an ergonomic standpoint, the keyboard was impractical for prolonged typing. Since the plan was to market the computer to businesses who would undoubtedly want to use it for typing documents, the decision seemed misguided. Peddle also decided the computer would have a built-in monitor. “We made the decision—independent of Radio Shack—that building the monitor in was the only way to sell the product.” Unfortunately, no one at Commodore knew anything about television electronics. “We’d never done that,” says Peddle. “I had a guy who was kind of kicking around Commodore at the time and I gave him the assignment.” The PET needed a small, cheap video monitor. Peddle realized the best way to find such a product was to go to the nearest electronics store. “Chuck had gotten some display people to help him on the wooden PET design a display,” says Seiler. “They went down to the TV store and bought the smallest, cheapest black and white TV they could get and then they took it apart and decided what part of the circuit would work for them.”

Wooden PET prototype with unreleased printer.

The team constructed their circuit on a separate breadboard with TTL components. “Thanks to my buddy Adam Osborne’s book on how to connect a TV, we got it to work,” explains Peddle. “He read Adam’s book and we just hacked. We took the TV board apart and we built something up so we could demonstrate the TV.” In December 1976, Commodore leaked information and photos of the mockup computer to Electronic Engineering Times. The same month, Commodore’s stock rose from $4 ½ to $7. It was a sure sign the personal computer could help offset Commodore’s financial woes in the calculator business. To help motivate Peddle, Tramiel offered him one dollar for every computer Commodore sold. Peddle accepted, though he seemed unsure if Tramiel was sincere about the deal. In spare moments, Andre Sousan tossed around potential names for the new computer. It would be three letters, as the TIM and KIM had been before it, and it would sound friendly. Sousan might have pulled some inspiration from a Commodore calculator. The box for the 776M called it a Personal Electronic Calculator. Sousan might have thought Personal Electronic Computer also fit in with this scheme, but the acronym PEC did not sound like much of anything. He contemplated the name further. Sousan soon received inspiration from a fad. Peddle explains, “In the fall of 1976, a guy in Los Gatos, California started packaging rocks with cute sayings called the Pet Rock, and he sold tons of them.” Legendary Californian advertising man Gary Dahl had caused an inexplicable mania by selling Pet Rocks. He packed a simple Beach Stone in a gift box shaped like a pet carrying case, accompanied by an instruction book titled Pet Rock Training Manual. In just a few months, Dahl sold over a million rocks at $3.95 each. Sousan loved the name. “Both he and I used the words ‘warm and fuzzy.’ I gave quotes about how computers are supposed to be warm and fuzzy,” says Peddle. To Sousan, the Pet Rock was warm and fuzzy. “That was absolutely the image we were trying to get. Once we picked it, we tried it on some people and they

kind of liked it.” The designers had an acronym, but it had no meaning. “I was making the words mean something so that we didn’t get caught up in arguments about stealing the Pet Rock,” recalls Peddle. Unfortunately, there were not many T-words in computer vernacular. Typewriter was a candidate, but it would mislead buyers into thinking it was an electric typewriter. Terminal was another candidate, but Peddle disliked the name since it strictly implied a brainless keyboard and monitor. Peddle finally hit upon transactor. “The Personal Electronic Transactor is because I was in the intelligent terminal business and transactions were something we used there,” he explains. “So the Personal Electronic Transactor doesn’t really fit, but it fit my company background.” It was a vague description, but it worked. Sousan also wanted a futuristic name. The movie 2001: A Space Odyssey (1968) seemed to epitomize the future. “If you think about it in 1977, that was the futuristic thing people talked about,” recalls Peddle. “That set a tone of futuristic. There wasn’t a Star Wars movie yet and there wasn’t a Star Trek [movie]. Go back in that time and you’ll see that 2001 implies a successful future.” Sousan appended the number 2001 to the PET name. “2001 was absolutely picked because of the movie. It wasn’t inspired, it was absolutely ripped off,” says Peddle, proving Picasso correct that good artists borrow, great artists steal. The computer now had a name. [1] Due to strong economic nationalism in at the time, Commodore was unable to create Commodore . Instead, a local company called PROCEP rebranded and marketed Commodore’s products. [2] Micro-Soft resided at Ed Roberts’ MITS, manufacturer of the Altair computer. The two companies were so close Roberts believed MITS owned the BASIC language, since Paul Allen was an MITS employee and director of software. In 1977, Micro-Soft painfully separated from an angry Ed Roberts. [3] At the time, Hewlett-Packard called it the GPIB (General Purpose Interface Bus). The IEEE later accepted the interface as a standard for minicomputer peripherals, such as printers and disk drives.

[4] The fate of the original wooden prototype is unknown. Leonard Tramiel comments, “I wish I could find that machine. I’m not quite sure what happened to that one.”

CHAPTER 6

The Demo 1976-1977

In late 1976, Chuck Peddle’s engineers continued troubleshooting every aspect of the PET. “We decided how to make the BASIC have I/O capability and we made several changes that we could get in on time [for CES],” explains Peddle. “Some we couldn’t get in on time, but we could get them in later before we went into production.” With his general specification for the BASIC language completed, Peddle brought in Micro-Soft. “We flew Rick [Wyland] out and he did the deal,” he recalls. “We negotiated a one-time paid-up license for the thing. Wyland was happy because he got money enough to cover all of their development expenses. So he did what Gates told him to do: he got rid of it.” Kit Spencer happened to visit California in late 1976. “I’d gone over for a meeting in Palo Alto with Jack and Chuck was around,” he recalls. “I got to know Chuck a little bit then and fortunately I managed to get on quite well with Chuck. I think we both respected each other and accepted there was something we could both learn from each other.” Spencer arrived at an opportune time to sit in on an evening dinner meeting between Commodore and Micro-Soft. “Before the meeting Chuck said, ‘You are going to meet someone tonight who looks like a 16-year-old nerd, but he is really quite bright and shouldn’t be underestimated.’” The description intrigued the Brit. Commodore paid very little for the rights to use Micro-Soft BASIC. “The price was actually negotiated by Sousan,” recalls Peddle. “He just took a flat deal and

a negotiated price that wasn’t very heavy because we were selling zero computers. Apple is selling 10 or 20 [Apple I kits] a month. The industry is totally in its infancy. Numbers like twenty-five and fifty thousand dollars are big numbers. So we bought BASIC with a onetime license, paid up.” According to Jack Tramiel, Gates originally offered royalties of $3 per copy, but Tramiel replied, “I’m already married.” Instead, he offered to purchase it outright for $25,000.[1] The deal stated Micro-Soft would keep the modifications made to 6502 BASIC and sell it to other companies. “The deal was, we would extend it, but he had the rights to sell whatever we did,” says Peddle. Peddle was not worried that other companies might use his innovations against him. “I was going to that because obviously I was still a MOS Technology guy. His [6502] BASIC was going to be available to other people,” he says. “They liked the extensions I put in, they thought they were a good idea, so they agreed to keep them in as part of their normal BASIC.” At first, Bill Gates was not very involved with the 6502 BASIC project. “We really didn’t work with Bill hardly at all on the product,” says Peddle. “It wasn’t until after we got going and started doing some stuff that Bill started getting involved, although he actually did write some of the code for the final product. Bill wrote good interrupt driven code that really worked.” In a 1993 interview with the Smithsonian, Bill Gates recalled his early start with Commodore. “…they started with us from the very beginning. Because we helped Chuck Peddle, who was at Commodore at that time, really think about the design of the machine,” he says. “Adding lots of fun characters to the character set, things like smiley faces and suit symbols. That was the first machine we did that had this wild extended character set.” Peddle explains the origins of the character set. “One of my specs was that I had to be able to do card games because that’s one of the things people did with computers.” Bill Gates is famous for his shrewd business style, but in 1976, he was still learning his craft. The deal for BASIC would ultimately turn into one of the biggest missteps of his early career. From the moment Commodore purchased BASIC from Micro-Soft, Commodore could include it in all computer models

they happened to make in the years to come, without paying royalties. “As far as I know, Jack Tramiel was the only one who ever got the upper hand on Bill Gates,” says engineer Bob Yannes. “They wanted to charge him a perunit fee and Jack would never agree to a licensing fee. He would only buy stuff right out, because that way he knew what it would cost him and he didn’t have to worry about what it was going to cost him in the future. I guess Micro-Soft didn’t think that Jack was likely to be very successful, so they agreed to it, and he bought Micro-Soft BASIC.” According to Tramiel’s son, Leonard, the BASIC programming language remained the same for years to come. “It was a one-time, royalty-free purchase,” says Leonard. “The ROMs went through different versions and there were changes in the operating system, but I don’t think there were any changes to BASIC itself. The BASIC part was identical.” The license allowed Commodore to improve the operating system, however. “As far as new features and functions [to the operating system], I don’t think those came from Micro-Soft. Those were done internally at Commodore,” says Leonard. Bob Yannes believes Micro-Soft underestimated Tramiel. “I think, probably in the back of their mind, Micro-Soft was thinking, ‘This is a primitive form of BASIC. Two years from now he’s going to need a new version anyhow, so who cares.’ Well, Jack didn’t care what kind of software his machines had,” he says, laughing. “He was putting the same BASIC in every machine, even though it was obsolete.” * * * With CES nearly upon them, Peddle and his team had little to show. “We are fucking hacking,” recalls Peddle. “We had two months to do something and make this absolute hodgepodge. We stuffed it into a case and nothing worked.” It was an anxious situation. “Jack had made it clear that if I didn’t perform, he was going to kill the project. This was wonderful, after I moved my family to the west coast and everything. But I love challenges, and usually if you start insulting me you just piss me off and I work harder.” The team boarded an airplane for Chicago—without a working computer.

Tramiel hoped to display the PET on the show floor, along with his calculators, but a bug prevented the PET from booting up properly. As Leonard recalls, “When the machine was hooked up to the development system, it worked great, but when you plugged the processor in, it didn’t work.” John Roach of Radio Shack would view the computer and make his assessment on the last day of the show. With only days remaining, there was intense pressure to eliminate the bug. Peddle recalls, “We got a suite, and I worked two groups of engineers 24 hours a day. I slept maybe one or two hours a day.” Tramiel toured the CES show floor to see if his competition had beaten him to the personal computer. The disco era was almost at its peak and throngs of attendees with trendy bell-bottom pants waded past the display booths. Companies used disco-balls and colored lights to attract attention while playing hits like the Bee Gees “You Should Be Dancing.” Tramiel sweated his way through the displays and was relieved there was no other competition for the personal computer. As the show progressed, the increasingly strung-out engineers searched for the source of the bug. In the midst of this turmoil, Peddle found beauty. “I a vision that I will carry with me until my death. I was sitting there, getting the computer sort of working, looking across the Chicago lakefront. It was 50 degrees below zero outside and there was this ice fog all over the lake, and the sunrise on it was just spectacular. I will always have the image from that hotel in my mind.” The sunrise was a good omen. Leonard Tramiel recalls, “It took a while to find that error, and unfortunately it was found the last day of CES. It turned out there was just a microscopic technical issue that was absolutely trivial to fix that took a surprising length of time to find.” The computer booted up, but the image horrified the engineers. “To give you a feel for how dumb we were about TVs and everything, the day before we were supposed to show it to John Roach we got the goddamned thing working but the image was upside down,” recalls Peddle. “We didn’t know how to fix it, so we go back and read the Osborne book. We were really not very sophisticated about some of that stuff. We finally got it kind of hacked together.” The Radio Shack meeting could make or break the PET. “On the last day of the

show, Jack brings John Roach up to the room,” recalls Peddle. “It’s the first time I met John Roach.” Roach was about to see his star rise in the business world, but he was an unknown in 1977. “John was not the head of Radio Shack at that time,” explains Peddle. “He was an operations guy. He just happened to wind up with the assignment of getting a computer to work.” Nervously, Peddle revealed the PET computer. “I show him this thing, sort of working and kind of interesting. John basically says, ‘Well it’s obviously not totally working, but you’ve got people that sound pretty smart. I know your company and I know you’ll make it happen. Let’s sit down and talk about a deal.’” The deal would have given the PET computer immediate and massive distribution all over the world in Radio Shack outlets. Sensing interest from Roach, Tramiel went ahead with his plan to ask for a larger order of calculators. “Jack wanted a big commit on calculators if he did it,” recalls Peddle. It was a big mistake, and Roach started to have second thoughts. “John was convinced that he could do it himself because he knew we hadn’t worked on it very long,” says Peddle. “He kind of looked at it and said, ‘You know, if those guys can do it that quick, I can probably get a team to do it.’” Commodore would not supply computers to Radio Shack. It was a devastating blow to Peddle. “So he basically went back to Dallas and commissioned a team to design the Radio Shack machine.” Peddle thought his dream was lost. Fortunately, the rapid rise of Commodore stock made Tramiel take notice. “We came back from the show,” he recalls. “Jack said, ‘I saw you bust your ass. I saw you get something done. I’ve decided we are going to spend the money and make this happen.’” It was an uplifting moment for Peddle, who had worked so tirelessly the past few months. * * * In January 1977, Commodore became the first computer company to announce a personal computer for the mass market. Commodore would unveil the PET at the Hanover Fair in March, followed by the West Coast Computer Faire in April. The features sured everyone’s expectations, with a built-in keyboard, monitor, cassette storage unit, BASIC programming language and four

kilobytes of programmable memory, all for only $495. The announcement sent shockwaves through the nascent microcomputer companies of the time. Some microcomputers sold for several thousand dollars. The Cromemco Z-1, an S-100 compatible computer, was selling for $2,495 without a keyboard or monitor. No company could offer such a complete system at such a low price. The hobbyist market had anticipated and dreaded this moment ever since Commodore acquired MOS Technology. Dan Fylstra, a KIM-1 and eventual PET owner, described the reaction in the February 1978 issue of Byte magazine:

…Commodore Business Machines Inc, a well-known maker of pocket calculators, startled everyone in the personal computer industry by announcing its PET computer. … The announcement was greeted with considerable enthusiasm and skepticism, for Commodore obviously was stretching the state of the art in technology, manufacturing and mass distribution.

Tramiel wanted Peddle and his team to have the final production model ready for the June CES show. He gave Peddle the resources to expand his team and create Commodore’s first computer systems division. Almost two years earlier, while Peddle was working out of Allied Leisure, he had promised engineer Bill Seiler that they would make a personal computer. Seiler was born November 14, 1949, making him 27 years old at the time. “I was born in Miami Florida and lived there until about 1976 when I drove out to California to find fame and fortune,” he says. “My dad was a Miami Beach fireman and my mom was a housewife.” In 1976, he graduated with a university degree. “I have a Bachelor of Science in engineering from the University of South Florida,” he says. Like most engineers, Seiler was disappointed the university lagged behind the latest advancement in electronics. “I being in college telling people about microprocessors, because the 8008 was out,” he recalls. “I said, ‘This is

important, you have to pay attention.’ But the professors didn’t.” A Professor Sampson particularly impressed Seiler. “There was one guy who was really hip. He was one of my most inspiring professors,” he recalls. Many engineers use trial and error when deg parts, but Professor Sampson instilled in Seiler to follow a problem through logically. “He was such an energetic engineer. He’d pull out paper and start with Ohm’s law to solve a problem. He even got me a job one summer in St. Petersburg when I worked for Honeywell on the Space Shuttle.” Seiler had his first introduction to a computer with the PDP-11 minicomputer from Digital Equipment Corporation. Seiler’s roommate, Larry Bryan, introduced him to the system. “Larry was doing some classes but he was mostly working in the electronic music department in the school downstairs,” recalls Seiler. “They had a bunch of Moog synthesizers and all that junk so the crazy musicians could do electronic music. They had a PDP-11 down there in the basement hooked up to all these things so they could make digital music.” The music students were far from protective of their computers, and were glad to have engineers around to help them. “Of course the music students couldn’t program, so us engineers who were struggling to find any computer we could, were always hunting around,” says Seiler. “I found Larry and his stuff.” One of the most popular video games in the early 1970s was Spacewar!, originally programmed for the PDP-1 in 1962. “I had heard about MIT Spacewar! back on the early computers,” says Seiler. “We had a large screen oscilloscope hooked up to the PDP-11. The computer had D to A (Digital to Analog) converters and A to D (Analog to Digital) converters on it, because that’s how you control the Moog synthesizers with anaputs and voltages and things like that. So I hooked the D to A up to the X and the Y on the scope.” Once Seiler had a way to output video images to the oscilloscope, he began recreating Spacewar!, which simulated planetary gravity. “I wrote that all in assembler,” he says. “The PDP-11 was a great processor. It had to generate every little dot on the rocket. That was a real task. It was the first time I did any assembly language programming.” The hardware was in its infancy, and programming was an arduous job. “I had to use paper tape and compile the modules separately, and link them all together. It

was all done without any kind of mass storage. The PDP-11 had 8K of RAM and a paper tape reader, and a boot ROM that could boot the loader and assembler and linker off the paper tape. I think I still have those paper tapes somewhere.” After graduation, Seiler drifted towards the most exciting job he could find in Florida. “Most of the engineers out of that university went to work for Tennessee Valley Electric or Florida Powerline,” he says. Instead, Seiler wanted something more exciting and playful, which brought him to Allied Leisure. In early 1977, Seiler became disenchanted with Allied Leisure. When the company had more orders for game machines than they could immediately deliver, it took advantage of existing customers. “They did some real unscrupulous things,” he recalls. “People would send in their broken video games. I was fixing them and then they would put them in a new machine and sell it to somebody! They’d just keep telling the guy they hadn’t fixed his board yet.” Although Seiler never found direct proof, he suspected organized crime owned Allied Leisure. The links from Allied Leisure to the mafia came through the company’s links to gambling. “We were making slot machines too at Allied,” he says. “That company was so corrupt, I think it was run by the mafia. It was just amazing. The guys were these guys that you’d swear were from Atlantic City.” On a trip to New York, Seiler met with upper management within Allied Leisure. “I took some trip up to the Coney Island area. We were kind of surveying the scene out there to see how our games were doing and getting from the arcade guys,” he says. “It was pretty scary. The guys looked like they were from the Sopranos. I would swear I was in that kind of scene with the characters I was meeting.” The unscrupulous nature of the company contributed to Seiler’s decision to leave. The young Seiler was about to embark on a journey, which would eventually lead to a new life. “After working at Allied Leisure, I decided to come to California for the fame and fortune,” he says. “I putt-putted across the country in my little yellow car.” Seiler’s journey took him over 3,200 miles across the southern United States. “I stopped to visit a friend of mine in New Orleans, drove up through Texas and decided to go to Colorado to visit Dr. Suding,” he says, in reference to the maker

of his Digital Group computer. Seiler had developed an interest in computer video, and had a special hardware project for his kit computer, which he wanted to show Dr. Suding. “I had designed a video card by that time that plugged into his machine to do video graphically. I wire wrapped stuff up and it didn’t work very well, but sort of worked.”

Bill Seiler (standing) in 1977.

The headquarters of The Digital Group were smaller than Seiler imagined. “I actually went and found him and was at his house. He was running the business out of his garage. I was sitting in his kitchen, talking to him about his design.” After his pilgrimage to the home of his beloved computer, Seiler continued to his destination. “I kept on going and came out to California. I had a friend that was living down in Santa Monica that I knew from college. I stayed with him. Then I visited my brother in Hawaii and came back.” Unsure of what he would do for a living, Seiler picked up a copy of Electronic Engineering Times and saw an article about the PET computer. “Chuck had gotten this front page story about the PET computer that was going to be $495. I called him up because I ed him from Allied. He said, ‘Bill, get up here. We need you,’” recalls Seiler in a pitch-perfect impersonation of Peddle’s voice. Those seven words made his journey worthwhile. “I jumped in my little yellow car and drove up.” Seiler came to California with a distinctive philosophy. “When he first moved to California, he came in this little Honda,” recalls Peddle, referring to the original Honda Civic that did not compare favorably to Peddle’s Porsche. “I asked him about it and he said, ‘I’m like a turtle. I don’t want to own anything I can’t put in the Honda.’ He wanted to pack up and go anywhere. That was his view about life.” Seiler spent his first few months in California living with Peddle’s family. “He was living in Saratoga with Shirley, his new wife, and her crazy family,” recalls Seiler. “I stayed with them for a while. There was a mix of two sets of kids: Shirley had kids and Chuck had kids. There was a real sizzling sibling rivalry going on but generally they were pretty cool.” “He lived with me in my house,” says Peddle. “Of course, the kids loved him. All of us loved him.”

Every morning, the two engineers commuted 25 minutes to Palo Alto. At the time, Commodore was struggling to remedy a chronic defect in its watches. “They had some real bad problems with the early ones with the alarm speaker,” says Seiler. “If you whacked the watch hard enough, you would reset the chip.” As a result, owners frequently had to reset the time on their watches. Shiraz Shivji, who had designed the watch, was trying to identify the problem. “I coming in the first time. I saw him working,” says Seiler. “They were sitting there whacking these watches on a table. I was watching them saying, ‘These guys are nuts.’” After much experimentation, the engineers isolated the problem with the watch speaker. “Turns out the little beepers in watches, piezoelectric speakers, generate a tremendously high voltage if you whack them,” explains Seiler. “The lighter that starts your barbeque with a click is a piezo being whacked. The electricity gets back into the chip and it resets the watch.” To redesign the watch circuit, Shivji made a prototype of the electronics on a large circuit board. “They had a breadboard that must have been two feet by three feet, wire wrapped with the real basic CMOS components you could buy from RCA. It was huge for a watch,” recalls Seiler. “They finally had to put a transistor outside of the chip that sort of buffered it so the [piezo discharge] didn’t get back in the chip. That was the only solution they found at the time.” Peddle knew Seiler had the potential to become a key member of the PET development team. Aside from The Digital Group computer kit and video card, he was involved in other hardware projects. “He was a blue boxer,” says Peddle, referring to the illegal practice of using an electronic speaker device to hijack telephone calls. “He even got busted for having a blue box. He was everything the hobbyists were and he was an absolute loner. No friends; nothing.” When Seiler arrived on the west coast, he signed up for the Homebrew Computer Club, an acclaimed nexus for early computer hobbyists. “That was the one that was meeting up at SLAC [Stanford Linear Accelerator Center],” explains Peddle. “Woz was in that, Gordon [French]—a whole bunch of people were in that.” Seiler’s only acquaintance on the west coast was Peddle, so the Homebrew Computer Club was a welcomed activity outside of work. He also spent his free time reading issues of Byte magazine and Dr. Dobb’s

Journal, the only publication at the time dedicated to microcomputer programming. “I Byte was an inch thick way back then,” recalls Seiler. The other new addition to the PET team was Leonard Tramiel, who had attended CES in January. “Leonard is the scientific son of Jack Tramiel,” says Bill Seiler. “He was a real math type and he was always checking the calculator stuff to make sure the algorithms worked right.” Leonard had been an invaluable product tester for Commodore’s calculators while in university. “The first calculators really worked in decimal but later on they went to binary and converted when they put it on the screen, but that created problems,” says Seiler. “They had weird problems where you could invert a number twice and it wouldn’t come back to the original number. That really bothered people.” Leonard was now 23 and had just earned his physics degree at Columbia University. “I graduated from the university in the spring/early summer of 1976,” recalls Leonard. “I took a year off from the middle of 1976 to the middle of 1977.” During this time, Leonard decided what he wanted to do with his life, and it brought him into a disagreement with his father. “[Leonard] loves his dad, but it’s kind of funny because there was a time when his dad didn’t love him,” says Peddle. Leonard fell in love with physics and now wanted to earn a PhD. “His father thought he was making a mistake,” explains Peddle. “Jack didn’t want him to do that, so he was squeezing him economically because Leonard was going off to be an astrophysicist.” Tramiel adamantly refused to pay for Leonard’s tuition. “Leonard is a very funny guy; very docile and very inventive at that time in his life, but a very mildmannered person,” says Peddle. “Jack was definitely not mild mannered. This was kind of like his first rebellion, and I helped him.” Peddle recognized Leonard as someone who could evaluate and test the project. “Leonard was involved in the QC [quality control] work,” explains Peddle. “He was very critical about what we did and he was very fair about it. He would run tests and figure out what we were doing wrong, bring it to our attention, and

demand we fix it. I put him in that role of being the in-house critical .” “Jack threatened to fire me because I hired Leonard,” says Peddle. “Jack didn’t hate me, but he gave me shit for it.” Tramiel was probably also grateful to see his son in the company. “I put him on the payroll and he managed to get through the summer, so Jack couldn’t squeeze him.” * * * Chuck assembled his team and began planning a better design for the PET. “It was a very small group,” recalls Leonard. “Chuck was a real bright guy who knew what he wanted the machine to do and explained it pretty well. We got together and did what he wanted. It was a pretty simple design and we made it work as best we could.” The engineers would be deg a circuit board, jammed with integrated circuits and processors. “We started building up two or three prototypes,” says Peddle. Bill Seiler entered the project as Peddle began attempting to transform the prototype circuitry into an actual production design. “By the time I got there, they had the wooden PET,” he says. “It was loading and saving programs on the cassette tape. The BASIC was running. They had the little chicklet keyboard Fujiyama had worked out with the Japanese guys at Commodore, and the graphics characters were on the keys.” Seiler soon learned that the electronics devised by Peddle and Petr Sehnal were in a rough state, as is typical of prototypes. “It was two or three boards from some other project that they had ginned together. It might have been a printer or something else they used as the base breadboard to build on,” explains Peddle. “It was a pretty hodge-podge thing and it was pretty flaky.” Peddle gave simple jobs to Seiler in order to familiarize him with the PET before tackling more complex jobs. “Chuck assigned a few things for me to do, like to write the power-on self test,” recalls Seiler. Like the KIM-1 before it, the PET circuit board would contain a MOS 6502 U running at 1 MHz, situated in the center of the board. To one side were eight sockets for the RAM chips. Peddle used MOS Technology 6550 Static RAM (SRAM) chips, each containing 512 bytes of memory.

Unfortunately, MOS Technology was having very little luck producing reliable memory. “We tried to design RAMs but that’s a true specialty. They could never do it well,” says Peddle. From the start, Peddle designed the PET as a self-contained system without expansion slots. “We were trying to give them the whole solution,” says Seiler. “We looked at the IMSAI Altair expansion board. This was a market that Apple looking at too. We decided you needed some way to talk to peripherals and not much else. Just wham-bang, there’s the whole thing.” Additional peripherals, such as printers and disk drives, would connect to an IEEE-488 slot at the back of the computer. “It was designed to be a selfcontained unit, and it had the IEEE interface to control all the peripherals,” explains Peddle. “That’s what made Commodore so much money.” s could connect up to 15 peripherals through the IEEE-488 interface. In many ways, the interface was an early predecessor of USB ports. “The expansion port has morphed into the USB plug,” observes Seiler. It was up to Seiler to implement the interface. “Chuck wanted me to work out a way to do some kind of I/O to external devices,” says Seiler. “So we picked the IEEE-488 bus, because that looked interesting. HP had this bus they had developed to control all their fancy meters and measurement devices. It was an 8-bit bus that was parallel, using pretty fancy cables, so you could stack them up. Each device had an address. It was really neat.” It was up to Seiler to implement IEEE-488 in the limited hardware of the PET computer. “We figured out a way to implement it cheaply all in software as much as we could,” he says. “The devices have to respond in a certain amount of time, so it created a little bit of a headache for us. We kind of blew it a little bit because it needed some hardware assist, but we didn’t do that.” If he had a new design worked out and needed it prototyped quickly, he turned to the Japanese technician Nobuo Aoji. “When we would have to build a prototype or something we’d give Aoji a schematic and he’d be like a Tasmanian devil in the lab,” recalls Seiler. “In about an hour, this breadboard would be built. He would get real energetically working on it. He was a real good technician.” Although Aoji was a hard worker in the engineering lab, he was equally energetic after work. “Aoji was a fun guy,” says Seiler. “He had a crazy lifestyle.

He’d go out on Saturday night and get real drunk and end up in jail. He was wild.” It was up to a Japanese-American engineer named Takamori to refine the video electronics hacked together during CES. He would need to conceive new circuits to control the cathode ray tube and display text to a screen at 60 times per second.

Nobuo Aoji working on a schematic.

The PET would contain all 128 characters defined by the ASCII standard, plus an additional 128 graphical characters, which allowed s to create pictures. s later whimsically dubbed the character set PETSCII. “I never really liked that term since it was never much of a standard,” reveals Leonard. “The part of PETSCII that coincides with ASCII is standard but the rest of it is not.” Leonard created the symbols through experimentation. “While the PET was being developed, I designed most of the graphics character set. It was a t effort, but I put most of that together.” The biggest test for his character set was a graphical version of a popular KIM-1 game. “What Leonard worked on, which he should take a lot of pride in, was Lunar Lander,” says Peddle. “It was the program we put in every book on the PET.” The graphical symbols contained arrows, lines, corners, curves, playing-card suits, and screen control codes–over 50 graphic symbols in all, most of them accessible from the keyboard. “It turned out to be a very useful and flexible way to do graphics,” says Leonard. Leonard contributed other games for the PET, all the while testing the computer and searching for bugs. “He wrote a lot of the games we had on the PET,” says Seiler. “Leonard put a lot of input into the graphics characters because he was writing all these games. He said, ‘You’ve got to have this card and we need this thing to make a little house or rocket.’” Bill Seiler, who was quickly becoming the most knowledgeable engineer on the PET architecture, also helped define the graphics characters. “The graphics didn’t all fit together right to do boxes and borders and things. They had already had all the graphics characters and I kind of massaged them and cleaned them up. I wrote a lot of BASIC games and stuff too.” The PET would require two kilobytes of memory to store all the characters. Peddle included a second text mode, which did not have many of the character graphics but ed upper- and lower-case text. This required another two

kilobytes for the text mode definitions. Chuck used a Commodore Semiconductor Group 6540 ROM chip to store the characters. “One of the things that made my personal computer work was that ROMs had just come into real strong being, and MOS made good ROMs,” says Peddle. “We took advantage of the fact that ROM had made a breakthrough in density. We were able to make a full operating system and a full BASIC and everything else and stuff it into ROM, just like today everybody uses flash [memory] for that function. Nobody had done that before.” For months after CES, the Commodore engineers refined BASIC with MicroSoft. “Micro-Soft did the debugging, we did the testing,” says Leonard. “I did more testing and not much interacting with the people at Micro-Soft. That was done largely by Chuck.” Ray Holt, designer of the SYM-1 and JOLT computers, also acted as a beta tester for Micro-Soft BASIC. Peddle chose a nine-inch Sony picture tube at random, but it fit the case so well he decided to strike a deal with Sony. Tramiel relied on Mr. Fujiyama and Aoji. “They were the Japanese leads and s,” says Peddle. “They helped negotiate the deal with Sony.” The two engineers also helped choose the Sanyo cassette unit for the PET. “We went out and bought a reasonably decent Japanese cassette,” says Peddle. “We took it apart and looked at the signals.” Peddle programmed his code for the cassette unit using the same ICE development system he used to make the Apple I operational. “It was a very powerful tool,” he recalls.

Video circuitry for the PET monitor.

“I literally sat there with a development system and a scope and wrote the code.” He throws all modesty aside when describing the code. “There is really some smart shit in there. It worked much better than anyone else’s cassette. It had error recovery and lots of stuff.” Lack of memory forced Peddle to leave his code undocumented. “The problem was that the development station that I was using didn’t have enough space in the memory to write line comments for the code I was writing,” he explains. “I wrote them down somewhere, but ultimately, this code was literally written on a machine with a scope and it wound up being undocumented.” The undocumented code would haunt Commodore engineers for years to come. The PET also allowed a second cassette unit. “The Commodore PET had a cassette built in it but there was a third connector on the back that was a small edge connector that you could plug an auxiliary cassette,” says Seiler. “Chuck was always of the school that to do standard business practice, you need to read data from one device and write the new data after you’ve worked on it to the second device—like reel to reel.” The final PET design had a simple, modular design. The main board had a video signal connector, which allowed the nine-inch monitor to plug right into the board. The cassette recorder and keyboard also had connectors on the motherboard, making it easy for s to upgrade the keyboard. However, before the final product release, the hand of Jack Tramiel would change much of the design. [1] Estimates vary. Engineer Robert Russell thinks it was $10,000. Kit Spencer thinks it was $12,000.

CHAPTER 7

Igniting an Industry 1976-1977

Leonard Tramiel unveiled the PET 2001 to the world. “The first showing of the PET was at the Hanover Fair in in March 1977,” he recalls. “It was shown first at the Hanover Fair in that hand-carved wooden case.” Though Peddle wanted to show off his computer, Jack Tramiel wanted him to continue development in California. “Jack felt I didn’t have time to do that,” says Peddle. Leonard was unhappy taking the PET to Hanover. “We stuck him with a product that didn’t work very much and stuffed him in a booth with the calculator guys. He was never happy with that assignment,” says Peddle. A month later, they would unveil the PET in the United States. * * * In early 1977, Bill Mensch began to feel he did not belong at Commodore. He had little enthusiasm for the video chip project, headed by Al Charpentier. “I realized, ‘This isn’t really the way I wanted my career to go.’” Mensch preferred working on general microprocessors, not specific applications such as video displays. “What Al was doing was trying to speed up a display problem.” “I’m realizing this is not the right place for me. I have two children under the age of three. When I’d go to work, it wasn’t very good,” he recalls. “So one day me and my buddy Ray Hirt went out and got drunk at lunch and drank from noon until five in the afternoon. It was a snowy day. I did a donut in the field next to MOS Technology in the snow. Unfortunately there was a ditch there and I got stuck.” Hirt drove Mensch home, and on the way, he paid a mechanic to tow his van out

to avoid having to give an embarrassing explanation to his employers. The next morning, Mensch came into work hung over. “I’m standing there on the second floor near the engineering offices looking at this green van outside still stuck in a ditch.” In that moment, he made a decision. “I’m thinking, ‘I need to be gone.’ I talked to the guy who was the CEO at the time and said it’s time to move on.” Mensch left in March 1977 to work at ICE (Integrated Circuit Engineering) in Phoenix, Arizona. Shortly after his departure, someone used a felt-tipped marker to black out Mensch’s picture from the group photo on the wall. Except for Peddle, no one from the original 6502 project remained. Commodore management did not understand how vital the 6502 engineers were to the continued success of MOS Technology. Unfortunately, Commodore would repeat this pattern many more times. Commodore engineer Dave Haynie later dubbed this pattern the “Commodore Curse.” * * * On the cusp of success, the Commodore Curse was about to strike again. Andre Sousan had spotted MOS Technology and the 6502 processor, instigated Commodore’s first personal computer, enticed Chuck Peddle to stay with Commodore and named the PET computer. It was time for the curse to take effect. According to Peddle, it was inevitable Sousan and Tramiel would part ways. “Andre was a Jack loyalist but Jack had a way of eating up people like Andre because Andre was relatively straight. A lot of guys around Jack had their shifty side.” According to Peddle, Sousan played fair. “He had a good mind and always used his mind well, in spite of his association with Tramiel. I’m sure a few times, because he was associated with Tramiel, he got on the odd side of a deal because Tramiel did that.” Trouble began in early 1977 over a seemingly insignificant disagreement over product and management philosophy. “Andre and Jack got into this fight over something that was going on in the calculator business, and Andre was defending his position one day,” says Peddle. “Basically Jack said, ‘Get out.’” It was the end of the line for Sousan at Commodore. “This guy had been totally

loyal and built his life around him.” Now he would end up helping Commodore’s rival achieve success. Peddle explains, “Andre basically walked across the street [to Apple] and said, ‘I want to be in the personal computer business, but I’m not going to get to do it with Commodore.’” In April 1977, Sousan approached Apple, the company he had hoped Commodore would acquire, and offered to set up European operations. At the time, Apple was looking for investors and they had no firm plans for marketing products in Europe. “They didn’t have a clue how to do Europe,” says Peddle. “Markkula didn’t know anything about it. Nobody knew anything about it and Andre did.” According to Peddle, Sousan had ample wealth from his Commodore days. “He was making serious money,” he says. “He went to Apple and said, ‘Listen, I will invest a bunch of money in the company. What I want is European distribution rights.’” Apple gave Sousan full distribution rights to all of Europe for Apple products, which Sousan marketed under the name Euroapple. “It was at a very appropriate time for him with Apple because Apple was just getting started and they loved having a guy to take over Europe. It was a perfect match.” * * * In early 1977, “Big” Jim Warren, the editor of Dr. Dobbs Journal, ed Chuck Peddle. Warren, a tall, bearded computer hobbyist, had an idea for a computer show just for microcomputers. “Warren is a funny guy,” recalls Peddle. “He dreamed up the idea of the show. He went to a whole bunch of us and charged us relatively little money to set up booths, because he knew all of us were just getting started.” Warren would make his profit on ission fees. “He was charging everyone twenty-five bucks to come by and look at what was effectively a computer store,” says Peddle. In April 1977, the West Coast Computer Faire was upon the engineers. The three-day event took place over a weekend. This was to be the first major

gathering of hobbyists and microcomputer companies, and many saw it as finally legitimizing the nascent microcomputer industry. “That is a landmark decision in the world of personal computers,” says Peddle. “Jim Warren deserves all the credit he ever got for dreaming that thing up.” There were no large computer companies in attendance—no IBM, DEC, or Hewlett Packard. Commodore was the heavyweight of the show and rivals eagerly awaited a glimpse of the much talked about PET 2001. In the months leading up to the West Coast Computer Faire, Warren became a promotion machine. He created a flyer proclaiming, “You can be a part of it.” He copied the ment on bright yellow paper and handed flyers out to anyone who would take one. The same flyer also appeared in full-page ads in computer publications like Byte and Dr. Dobbs Journal. The gathering took place at the Civic Auditorium across from City Hall in downtown San Francisco. On Friday, April 15, the doors to the underground Brooks Hall opened to the public. “There were important moments in history, and that was one,” says Peddle. Hobbyists flocked to the event. “The fucking lines outside of the building g up for the show were phenomenal,” says Peddle. Attendees received a tantalizing 42-page, black-and-white program with heavy gloss covers packed with descriptions of exhibits and presentations. They also received a brown paper bag with string handles to carry paraphernalia. Warren fulfilled his promise of “You can be a part of it.” “He was right, people wanted to be part of it,” remarks Peddle. “By that time, everybody in the Bay area wanted a part of the action. It had been talked about and written about, and everybody wanted something to happen.” The auditorium was gleefully chaotic. Warren expected 7,000 to 10,000 enthusiasts to attend, a figure many thought overly optimistic. Instead, 13,000 attended. Saturday was the big day for the convention and the aisles were jammed with people. The floor overflowed with tables and booths, separated by yellow-orange curtains. People waving dollar bills crowded around the booths. “It felt a lot like Woodstock. Everybody was there in fucking jeans and there was not a suit in the place. These were the guys with the thick glasses and the slide rules in their pocket. This was their thing.”

Peddle was clad in a leisure suit, which was considered stylish in the seventies disco era. Peddle’s picture would later appear in Personal Computing, much to his embarrassment today. “There’s this article inside and I’m wearing this really hokey leisure suit,” he recalls. “I weigh about 250 pounds, totally out of shape.” Above the din of the crowd, attendees could hear strange sounds they had never heard before. These were early demonstrations of computer music and sound effects. The perfect, unerring rhythm captivated early listeners even though the buzzers producing the sound were primitive by modern standards. There were even early demonstrations of speech synthesis, slurred almost to the point of incomprehension. Other booths featured a chance to have portraits digitized and printed on computer paper. One young artist attempted to sell computer-produced artwork for the outrageous price of six hundred and fifty dollars per printout. Compared to industry gatherings today, the first West Coast Computer Faire was pandemonium. It was about the money, but it was also about learning and exploration with a distinctly fun element. Organizer Jim Warren wore roller skates to get around the show. If a new complication arose which demanded his attention, Warren would press the walkie-talkie to his ear, give a quick reply and skate off in a new direction. Computer s brought their nerdy science fiction enthusiasm to the convention. There were at least three unlicensed Star Trek games displayed in various booths. The games were unsophisticated, but they gave onlookers a chance to understand the potential of computers in a fun way. Attendees would type in commands and most of the time received the response, “Captain, I’m afraid your last command made no sense.” Despite the shortcomings of these early games, people received a distinct thrill from interacting with a computer. Around the edges of the auditorium were side rooms, where presenters gave unorthodox presentations. Among them was Ted Nelson, author of Computer Lib, a book hailed as revolutionary for the time. The book, written in 1974 before the KIM-1 and Altair 8800, envisioned networked computers with libraries of information linked together. At the conference, Nelson gave an imioned speech about software, criticizing the current state of affairs. In the center of the auditorium were the larger microcomputer companies,

including Commodore (with their Mr. Calculator booth), Processor Technology, IMSAI, MITS, Sphere, Cromemco, Digital Research, and Byte Publications. Attendees glided from booth to booth, watching demonstrations and stuffing their brown paper bags with product handouts. Many of the attendees used the show to decide which computer they would eventually purchase. In the center of the anarchy was Chuck Peddle with his PET, which had increased in price from $495 to $595. “We kind of got the prototypes working at the West Coast Computer Faire,” says Peddle. The PET could power up and run BASIC programs, but it was not the final design. Physically, the PET computer impressed spectators. Other computers were a tangle of wires and cables. Wires ran from the keyboard to the computer, from the monitor to the computer, and from the cassette recorder to the computer. As well, every device had its own power cord. It was wire spaghetti. In contrast, the PET had one single power cord. Spectators loved the beautiful, smooth styling of the case. Many commented that it was exactly how a computer should look. Every half-hour, Peddle gathered spectators around and began his presentation. “We had to develop the market for computers and tell people why they wanted a computer. We told everybody they could checkbook and all that shit.” Peddle boldly predicted something that even computer fanatics had a hard time believing. He said, “You’re not going to stop with one computer. Most everybody is going to have two or three.” “They thought I was fucking crazy,” says Peddle. He concluded his presentation with a memorable comparison. “The idea was to plant ideas in people’s minds. If you get a big audience, you try to leave them with things they don’t forget. I used to tell people that computing was an urge. It was a need, the same as the hierarchy of oxygen first, water second, and all that stuff. And computers fit in there.” Where did computers fit in Peddle’s hierarchy? “Right about sex,” he says. “Once you are hooked on computers, you will get back to using your computer whether you like it or not.” Between presentations, Peddle had time to tour the show floor and inspect other products, such as the Apple II from his friends, the two Steves. Peddle liked

what he saw. “Steve [Jobs] was focusing on a really neat package,” says Peddle. “It was done very cleanly.” The Apple II did little to stand apart from other computers. It lacked lower case characters, which made it unsuitable for word processing, one of the major uses of a computer. It had no cursor keys. s were required to hold down the control key with a letter combination to move the cursor, such as Esc-D.

Chuck Peddle displaying the PET at the WCCF.

Spectators who turned on the Apple II saw a screen full of random alphabetic characters and symbols, and possibly some colored blocks. s had to press a reset key in the upper right hand side of the keyboard to get things started. After this, an asterisk would appear in the bottom left-hand corner of the screen, indicating the computer was in monitor mode. To get into Apple-BASIC, the had to press Ctrl-B, which would bring up a different prompt indicating BASIC was ready. Compared to the PET, it was not -friendly. Peddle received free publicity from the show. “Magazines were just starting to happen. You know this argument that rock music and all that stuff was pulled together by Woodstock. That show kind of brought together the suppliers, the customer, and the reality that this was going to be a real business.” One writer for Personal Computing magazine produced a series of articles detailing the beginnings of the personal computer industry. Peddle was interviewed and, along with others, gave predictions for the industry: some realistic and some far fetched. “I had made certain statements of where the industry was going and what the computer was going to do. This guy went off and he talked to a whole bunch of people, who later on came into the business, asking, ‘Is this going to be a real business?’ Everybody’s saying, ‘Yes, it’s going to be a real business. It’s going to happen.’” Personal Computing later published an article titled “Birth of the Personal Computer.” “The picture on the cover was literally an operating room with a doctor assisting birth, and he’s pulling this personal computer out of this blanketed form, and it’s the PET.” Inside, along with a photograph of Peddle, resplendent in his leisure suit, is an article naming Peddle as the father of the personal computer. Years later, Apple revisionists attempted to paint the West Coast Computer Faire as Apple triumphing over the competition.[1] According to author Michael S. Malone, this depiction could not be further from the truth. In Infinite Loop, Malone claims most attendees other companies such as Commodore making a stronger impression. Magazines from around this time strengthen this viewpoint, with Byte completely failing to mention Apple in their July 1977

coverage of the West Coast Computer Faire. Computer buyers in 1977 had a difficult choice to make. Among the deciding factors was price, where Commodore was poised to dominate their rivals. In contrast, Apple wanted $1,298 for its 4-kilobyte computer. Those lured in by Steve Jobs’ sales pitch would have to invest even more money to obtain a cassette recorder and monitor. The Commodore PET sold for less than half the cost of an Apple II. For most people, the PET was an easy choice over the competition. Three personal computer titans were about to rise from the West Coast Computer Faire and dominate the computer scene for the remainder of the seventies and into the eighties. “I wasn’t sure if Jack was going to screw me about the computer before we got to the West Coast Faire,” says Peddle. “Jack goes to the show, he sees the people lining up, and he says, ‘This is a business and I am going to play in it.’ He backs the program.” “John Roach [from Radio Shack] comes around scouting the show and says, ‘I am going to make a business out of this’. He goes back and finishes his machine, and launches his machine at the East Coast show in New Jersey later that year.” “Apple had their prototype there, I had prototypes there, and that’s really and truly the genesis of the personal computer,” says Peddle. “Before that, we were all trying to get there. All of a sudden, here’s this guy [Jim Warren] proving to everybody that there is a real market by bringing the technical people together. It was a really, really important deal.” The West Coast Computer Faire awakened Tramiel to the possibility of a new computer business. Commodore was barely surviving the calculator wars and he was ready to use every tactic he knew to dominate this new industry. He would use resourcefulness, cunning, flexibility, secrecy, speed, positioning, surprise, deception, and manipulation. Any company not prepared to match Tramiel would find itself slowly fading into oblivion. * * * As the hardware design of the PET neared completion, it occurred to Peddle that he needed a dedicated software engineer. “I had to have somebody who could write the fundamental architecture.”

Finding a skilled engineer in the wild, like Bill Seiler, was rare, so Peddle tapped into a steady source of engineers. “When I was at Motorola doing the first computer class, there was this balding guy sitting at the back of the room. He introduced himself afterwards as Dr. Roger Camp from Iowa State University.” After Peddle’s class ended, the two talked. According to Peddle, Dr. Camp told him he was on sabbatical. “It’s a fucking professor!” exclaims Peddle. “He and I got to be friends somewhat at that point.” After Peddle left Motorola, he maintained with Dr. Camp, who continued to teach at Iowa State University. “I call him up and tell him I’m working on this personal computer.” Camp had already heard of the PET project when Peddle asked for a software architect. “I’ve got the perfect guy for you,” replied Camp. “He calls this guy up, but this guy says, ‘I’m packed to go to IBM and I’m leaving this weekend.’ Roger said, ‘Listen, don’t go until you’ve flown out to see this guy. He’s crazy but I think you might like what they’re doing.’” When the young engineer, John Feagans, arrived in California, Peddle and Seiler tried to convince him not to go to IBM. “Bill Seiler and I sat down with him and we told him what we are doing.” The sales pitch worked, and Feagans ended up staying far longer than a weekend. “I think he stayed in that hotel for about three weeks.” “It was funny because about three weeks later IBM calls Dr. Camp and they said, ‘We are really worried about this young man John Feagans because he was supposed to leave on this weekend to come and see us. We think he’s been kidnapped!’ Roger said, ‘Well, he sort of has been,’” recalls a bemused Peddle. Feagans was a brilliant addition to the PET team. “Every year after that, Roger would pick the smartest, weirdest kid in the class and send him to me.” Even though Feagans was reclusive, he was everything Peddle wanted in an engineer. “Feagans was an absolute loner. He got into his code and spent all his time doing that. He stayed friendly with everybody but was really and truly distant.” Feagans’ task was to write the code that resided in the ROM chips. “He wrote

the system,” explains Peddle. “He was the guy who wrote the glue.” In essence, Feagans was creating the operating system. Commodore’s programming development tools were in a primitive state early on, and Feagans’ first task was to develop programming tools that could actually run on the PET computer itself. Feagans took his 6502 assembler, which he had earlier developed for the PDP-11. “When I ed Commodore I ported the assembler to the PET and we used it to build everything,” he says. Feagans also created the compiler, which took BASIC code and compiled it into machine language. “The BASIC compiler was created from a compiler-compiler I had written as a grad student at Iowa State,” he says. Feagans also tackled some obscure problems in the original PET code. “We had a character for PI,” recalls Seiler. “You’d type the character and you’d get the number but it wasn’t 3.1415926 something. It was off a little bit because it was a binary representation of PI and there were some rounding effects. So Feagans fixed that, fudged it so it looked closer to the real PI, which made people happy.” During the intense development, the engineers worked 16-hour days to get the PET to production in time. “We’d always carpool up there. I those crazy nights when John Feagans and I were sitting in the back of Chuck’s Ford, and he was driving home,” recalls Seiler. “It’s after midnight because we had been up all evening working on the PET. Chuck’s hardly awake and he’s trying to stay awake, so he would roll down the windows in the wintertime.” Often the California air at night was not cold enough even during the winter months to keep Peddle awake. “Feagans and I are looking at each other ready to grab the steering wheel because he’s nodding off,” laughs Seiler. Fortunately, the three dedicated engineers escaped any serious accidents. In order to cut down on the commute time, Seiler soon moved closer to Commodore’s headquarters. He was young and mobile, and not ready to set down any roots yet, so he rented. “I moved out of the Saratoga place and got an apartment in East Palo Alto because we were working in Palo Alto on California Avenue,” he says. The PET had many more functions than the KIM-1. As a result, the ROM was an enormous 18 kilobytes stored on seven 6540 ROM chips. It was up to Feagans to fuse the BASIC editor, file commands, character set, and other sub-systems into

a seamless operating system. The operating system would accept keyboard commands such as LOAD, RUN and SAVE. Feagans was determined to do things right. He even inserted a one-kilobyte diagnostic routine into the ROM. Commodore technicians on the factory line used the routine to identify errors and validate the operation of newly assembled PET computers. The diagnostic routine ended up squeezing out the machine language monitor, which Peddle planned to include in ROM, but that was fine since Peddle believed machine language was beyond ninety percent of his s anyway.[2] Contrasted to the random characters of the Apple II startup screen, Feagans’ operating system was friendly as soon as the turned on the power switch. The PET started up by communicating to the that BASIC was ready, with the message *** COMMODORE BASIC ***. It then listed the available memory, told the it was READY, and presented a flashing cursor. * * * During this hectic period, Peddle encountered health problems. “Every so often I would go into what was the equivalent of a kidney stone being ed. It paralyzed me.” He visited a physician, who informed him he was having gallbladder problems caused by the formation of large gallstones. “I had to have my gallbladder out.” Unfortunately, the doctor would not operate unless Peddle lost weight. “The doctor diagnosed me and said, ‘I’m not going to touch your body until you lose 50 pounds.’ He got me running and changed my life.” Peddle began a strict regimen of jogging, accompanied by the athletic Bill Seiler. “I took up running seriously because of the [impending] operation.” The two jogged through the nearby Stanford University campus, where students saw two men out of breath as much from running as from their animated PET discussions. “We used to go on these ridiculous long runs,” recalls Seiler. “I’d go on these long runs with him and his little brown shaggy dog. We’d be out running for five hours. The poor dog; you can’t run dogs that long. I a couple of times I got dehydrated and almost into hypothermia, until we found a front lawn with a sprinkler to get some water.”

* * * One day, Peddle learned there was very little money to develop the PET. “Commodore was having serious, serious financial problems when we started off with the PET,” he explains. “The calculator stuff wasn’t doing well and they were getting killed all the way around. Jack smartly never itted he was broke.” One of the most expensive parts was the molded plastic case, which the engineers loved. “Everything was tapered just properly so they could mold it out of plastic,” says Seiler. “It was supposed to be designed so it would be sort of like a Styrofoam cup; it would pop out of the mold.” As the production date neared, Tramiel began to have cold feet about producing a costly production mold. “Even though he had made all these brags to me, he really couldn’t afford the money to go tool a case that looked as pretty as [the wooden prototype],” says Peddle. “But he wouldn’t it it at the time.” According to Seiler, “A big mold like that was an investment of $300,000.” It was inconceivable to spend so much on an unproven product. Instead, Tramiel tried to convince Peddle to use a folded-metal case. “I said, ‘I want to build the case out of plastic,’” recalls Peddle. The rounded, molded plastic case would have easily made the PET stand out from their competitors. It was a chance to take the lead and cultivate a reputation for Commodore as a stylish innovator. “He said, ‘Listen, we’re going to build the case out of metal.’ I said, ‘But Jack, it doesn’t fit what I’m trying to do at all.’ He said, ‘You don’t know what people will buy, but I’m the boss, and we’re going to build it out of metal.’” It was a difficult change to accept but there was little Peddle could do. Commodore’s metal products division started earlier in 1977 with the acquisition of two companies: Nortex Products and Gildon Metal Enterprises. “The metal case was because Jack owned an office furniture factory up in Canada,” explains Seiler. “They made file cabinets and desks. I don’t know if it was that busy so they sent the wooden PET up there.” Both Peddle and Tramiel made the trip north to see if a metal case was possible. “We get on an airplane, go up to Toronto, and sit down with these guys,” recalls

Peddle. “They start tooling the metal, and they convinced Jack and Jack convinced himself that they could make something that looked almost as good as my pretty one. But it didn’t.” Tramiel and Peddle also worked with the Canadians picking out color schemes for the computer. “The guys in Canada said, ‘What color do you want?’ I said, ‘Fuck, I don’t know. I think I want light colors.’ Jack said, ‘You know, we can paint them anything we want.’ So we actually built 25 of about 10 different colors.” “There were black ones, and red ones, and yellow. I don’t think we ever did pink ones.” Yet another case featured wood- trim around the monitor and keyboard—a popular look for seventies’ electronics. “The vote came down on the light blue with the light white.”

Nortex Products in Toronto. Note the PET computers in the window.

Tramiel’s instincts were either brilliant or shortsighted, considering he did not have funds for such a large investment in an unproven product. “They thought they saved money, but I think that was probably more expensive than the plastic molding thing in the long run,” says Seiler. For Tramiel, the move fit in well with his strategy for vertical integration. “His file cabinet business wasn’t doing that good. So he got some business for that,” says Peddle. Peddle had already compromised once on the keyboard. Now he was giving up one of his principle design goals. “[I wanted] something friendly; spend the night with it; cuddly. That’s why we didn’t like the metal very much.” Secretly, Peddle was becoming frustrated. Although Peddle believes similarities between the PET 2001 and 2001: A Space Odyssey were incidental, the look of the new PET case was very close to the production design of the film. The fictional ship, Discovery, is stark white, with trapezoid shapes, including the computer monitors. This matched the trapezoidal profile of the new PET case. The associations of the PET 2001 and 2001: A Space Odyssey became even more apparent when the Japanese designers presented their final keyboard. In the film, the buttons on the Discovery are mostly of blue and red, exactly the same coloring as the PET keyboard. The PET keys are also in a perfect grid, just like the button s in the film. During the scene where HAL ejects astronaut Frank Poole into space, the screen displays the words ‘computer malfunction’ in upper case surrounded by a red background. Other displays, such as the vital signs of the astronauts, are in white surrounded by either red or blue rectangles. This is identical to the look of the space, shift, and return keys on the keyboard. Even the font of the keys is identical to the fonts used in the movie. The effect it had on buyers was undeniable. Richard Mansfield, an early PET 2001 owner and editor of Compute!’s Gazette, later described the PET as “a

lovely, futuristic piece of work: a sleek gray cabinet, a glossy keyboard, and a nine-inch black-and-white monitor jutting above like the entrance to a space station.” Peddle believes no one intended to emulate the look of the film. “We were not copying 2001. We hadn’t seen it for a long time,” he says. “Probably we just both picked a popular font.” However, it seems improbable the similarities could have occurred by chance. Since the keyboard design went to Japan, it is likely the designers noted the number 2001 in the name and decided to use the film for inspiration. Seiler found it difficult to open the machine to inspect the motherboard. “When they made the first couple of metal prototypes, the whole top came off when you took the screws out,” he recalls. The team began considering options for easier access into the computer and noticed a novel solution in the Silent 700 terminal in their office. “TI had a TTY terminal that we used as our printer,” recalls Seiler. “The Silent 700 was a series of TTY terminals that TI made. It was hooked up to our MDT650 development system that we did the software on.” The terminal used an arm to hold the case open. “It was like the hood on your car,” explains Seiler. The engineers adapted the same mechanism to the PET case. “It hinged across the back,” says Seiler. “You’d have to take two screws out and it would tilt up and you could see the whole motherboard and everything. The hinging thing with the arm was much better, and it allowed you to get in there and fiddle around.” Although the final metal PET case was not what Chuck Peddle wanted, it was still a remarkable design. It was simple and functional. In contrast, both Apple and Radio Shack had seals on their cases, and breaking the seal voided the warranty. Now Commodore had to figure out how to make thousands of PET computers. [1] The TV movie Pirates of Silicon Valley (1999) inexplicably shows everyone flocking around the Apple booth, while all the other booths are deserted.

[2] Commodore US included the machine language monitor on a cassette tape.

CHAPTER 8

Building a Factory 1977

The West Coast Computer Faire was a show for hobbyists, allowing Commodore to connect with the grassroots of the computer movement. However, for massmarket penetration, Jack Tramiel would take the PET to the summer Consumer Electronics Show. CES first debuted in 1967. Crowds at the show consisted largely of electronics manufacturers and dealers, and it was all business. The manufacturers created elaborate displays to highlight their latest electronics and set up orders, while the retail distributors walked the floors looking for exciting new products. Organizers hosted the show twice yearly, with the January show used to display new technology and the June show to take orders for the Christmas season. No one had ever displayed a microcomputer at CES. Commodore also entered the video game market at CES. Pong machines were popping up in living rooms all over the country, and Tramiel hoped to capture a part of this market. Commodore released two Pong consoles, the TV Game 2000K and 3000H, both manufactured in Hong Kong. “All that gaming stuff was from [Al] Charpentier,” says Chuck Peddle. The systems were slightly ahead of others at the time since it included four-player Pong complete with four controllers.[1] “It was very much a short-term product,” recalls Kit Spencer. “We sold the original Pong-type machines we had at Christmas quite happily.” Peddle wanted to transform the microcomputer into a true consumer product, right beside televisions and VCRs. By attending CES, Tramiel would do more to

popularize the image of personal computers than any other figure in the industry. Attending CES was no small expenditure. The West Coast Computer Faire cost Commodore about $100,000 in flights, registration fees, and displays. In contrast, attending CES would cost close to half-a-million dollars. Unknown to everyone except for Tramiel and his inner circle, Commodore was almost out of money. “He never itted it to me,” reveals Peddle. “Chris [Fish] finally itted it later. He said, ‘Jack was absolutely broke! We were just dead broke. We needed every penny just to stay alive.’” Commodore’s vice president of finance, Dick Sanford, recalls the difficult period. “It was tough going back then. We were a company trying to build this thing without cash,” he says. “The banks didn’t understand at that point, ‘What is this new industry?’” It was up to Sanford to raise the funds. “I dragging one of the three computers that Chuck had created to the banks in New York and trying to convince them to give us funding for this thing,” he recalls. “At that point, they were thinking they were funding this thing called a watch and calculator company up against guys like TI. Before you go to Wall Street with this thing, you have to prove to Wall Street that you’ve got something real. And here you’ve got this guy who’s walking around with this new dream. Is this really a venture? Is it something Wall Street wants to get behind?” Unfortunately, Peddle and his team did not have the PET ready for release in June as they had hoped. “We took the computer to CES in June with just a prototype working,” recalls Peddle. “It really wasn’t ready to go.” He also brought the different colored steel cases from Canada. Anyone who attended the show saw a rainbow of PET computers, including the wood-ed PET. On June 5, 1977, as Three’s Company made its television debut in North America, the doors to CES opened. Most of the Commodore booth displayed calculators with a small section for PET computers. Commodore was the only personal computer company at CES. The PET was a curiosity at the show and garnered attention from the crowds, if not sales. The show also generated publicity for the PET. During the course of the show, Commodore’s stock rose to 9 ¼, a clear signal that the personal computer held promise.

Being the first microcomputer company to explode onto the scene, Commodore was the key beneficiary of attention directed at microcomputers. Within a short period after CES ended, the media began generously covering the PET. Small news items appeared on local television shows, radio stations, and in newspapers and magazines. However, for all the interest generated, the PET was still just a curiosity. Most of the large retailers were unsure if it was a marketable product and were unwilling to risk their careers by ordering shipments of the computer. “We had shown the product at CES but the computer dealers weren’t really there, and the other guys weren’t sure,” says Peddle. “It didn’t get a great response.” However, the interest they received from one person inspired Tramiel to survive the cash crisis. “There was a guy in the Midwest who used to sell used computers. Big stuff,” says Peddle, referring to bulky mainframe computers. “He saw the [PET] computer at the West Coast Computer Faire.” Later, the same man approached Peddle at CES. “The guy walked into my booth and said, ‘I want to be the first customer. Here’s a check for 25-thousand dollars. When you ship me that many, I’ll give you another check.” The man was Rick Inatome, who later founded the pioneering Inacomp Computers. “Rick was one of the major players in the Midwest computer business,” says Peddle. The check stunned Tramiel. Normally, retailers placed orders and only paid when Commodore delivered the product. Here was someone who would pay large amounts of cash up front for a non-existent product, just so he could ensure his supply. The transaction caused Tramiel to reformulate his business plan.

Jack Tramiel g orders at CES.

Days after the show, Tramiel approached Peddle with his new idea. “Jack comes to me one day and he says, ‘We’re going to be like HP.’ I said, ‘Yeah, Jack, that’s what I’m trying to do.’” However, like so many others, Peddle missed the meaning in Tramiel’s words. Tramiel told him, “No, you don’t understand. Listen to me. When HP first came out with their calculator, they took advance sales where people had to send the money in and then they shipped to them within 90 days. That’s what I mean about being like HP.” Peddle now laughs at the misunderstanding. “I’ve got this image I want to be like HP and I want my product to be professional. He doesn’t mean that at all! He means the selling technique like HP. It doesn’t matter to him what the product looks like. None of that matters to him.” Tramiel noticed the high demand, so he instigated a policy of advance payments, with a promise that customers would receive their computer in 90 days. The ment read, “Shipments of 8K units will begin in September 1977. Price $795.” To make the contract legally binding, it included the phrase, “I attach my check for $795 in full payment. I understand that delivery will be within 90 days or I will get my money back.” In truth, many would wait for up to six months for delivery. Tramiel’s idea saved the company. “We were so broke we could not bootstrap the company without that money,” says Peddle. * * * At Apple, chairman Mike Markkula had already invested $250,000 of his own cash in the fledgling company. He began transforming Apple into a professional company. “Mike Markkula is a marketing genius,” says Peddle, who credits Markkula over Steve Jobs with Apple’s initial success. “He was the guy that was tying the company together and he brought in Regis [McKenna].” Regis McKenna, a former Intel advertising executive, was responsible for

Apple’s early advertising. “Regis hit on this great idea of making these two guys in the garage and how wonderful they were,” explains Peddle. “Mike Markkula and Regis McKenna created the Apple legend in a very careful, controlled, PR way.” Without the two former Intel executives promoting Apple, Jobs and Wozniak did not stand out from other computer hobbyists of the time. “There were six guys who did kit computers at the same time or ahead of them,” says Peddle. “Woz and Steve would have wandered off into the sunset and you never would have heard of either one of them, except Mike Markkula fell in love with them.” Of the two founders, Steve Jobs was easy to sell to the public. “Steve [Jobs] is a very, very charismatic guy,” recalls Leonard Tramiel. However, the mediocre sales of the Apple I computer, a period before Markkula and McKenna, testifies to his lack of results compared to other kit computers. “It’s really easy to get carried away with what he is saying, but as far as actually producing sales and making money and selling machines, Commodore did a far better job than Apple.” Although Steve Jobs was a natural, Steve Wozniak lacked a compelling persona. “If you were to read the literature during that time, you will discover that [McKenna] probably took Woz to two places and then dumped him because Woz just didn’t come across as smart and interesting,” recalls Peddle. In May 1977, Apple moved into their first company headquarters in Palo Alto, close to Commodore. McKenna launched Apple’s first ment in Byte magazine in July 1977, a month after Commodore sold its first PET computers. In comparison to the amateurish advertising of other computer makers in the 1970s, early Apple magazine ments are clean and refined. Apple’s marketing impressed Leonard Tramiel. He recognizes the company as a pioneer in personal computer marketing. “As far as I’m concerned, Apple invented marketing,” he says. The Apple II manufacturing process was comically stone aged compared to Commodore. “They didn’t have funding,” explains Peddle. Since Apple did not have the money to set up a factory line, workers assembled Apple II motherboards from their homes. An enterprising woman named Hildegarde Licht stopped by Apple in her white van to pick up boxes of parts and

components. During the day, she drove around Silicon Valley dropping off the boxes and picking up completed circuit boards. This haphazard system probably ed for the high failure rates of early Apple computers. In August 1977, Byte had a full cover picture of Isaac Newton sitting under an Apple tree. Unfortunately, it had nothing to do with Apple computers. When people mentioned Apple and computers together, they were usually talking about APL (A Programming Language), which many people just pronounced Apple. In 1977, customers greeted the debut of the Apple II with silence. While Commodore was receiving dozens of dealer inquiries per day, Apple was selling less than five computers per day. In the months that followed, Apple received about three hundred orders for the Apple II. The results were promising compared to the sales of the Apple I, but disappointing compared to Commodore and Radio Shack. Part of the reason for the lack of Apple II sales was due to the unfinished state of the Apple II. Bob Yannes explains, “The Apple II was fully assembled, but it did not have a TV output because they couldn’t get around the FCC emissions problems.” The Federal Communications Commission regulated, among other things, radio signals. Unfortunately, the Apple II emitted strong radio interference signals, which the FCC does not allow in consumer devices.[2] To get around the rules, Apple pretended it did not intend the Apple II for the home market. According to Yannes, “They basically said, ‘Okay, this is a data processing device and therefore the Class A FCC rules apply to it (the more relaxed rules) because this is going to be used in an industrial environment and not in a home.’ They couldn’t put an R/F modulator in it to hook to your TV set because obviously that was something for the home.” Without an R/F modulator, the Apple II was too complicated for inexperienced s. “The PET and the TRS-80 both came with their own monitors, so they were a more appropriate solution for most people than the Apple II was,” says Yannes. The original design by Steve Wozniak also had several flaws. “Right after the Apple II came out, Electronic Engineering Times wrote a story about the three major design flaws that Woz made on the Apple II,” says Peddle. “He didn’t understand the ways the [6502] chipset worked and some other electronics

stuff.” In response to these problems, Apple hired an engineer to redesign Wozniak’s motherboard. “There was a guy who was hired at Apple to redesign the Apple II and make it real engineering without offending Woz,” explains Peddle. “He basically went through and redesigned the product. Woz was a technician—he wasn’t a real engineer.”[3] The other disappointing feature of the Apple II was the lack of a decent BASIC programming language. The Apple II relied on Integer BASIC, which was inadequate for business use. “Three months after we are both in production, Markkula comes up and they buy Apple BASIC from Microsoft,” recalls Peddle. “Gates again got a fixed price, but higher [than Commodore],” explains Peddle. How much did Apple pay for its BASIC? “Gates at this point effectively took our number and doubled it for Apple.” Peddle believes the computers appeared at roughly the same time. “Both of us were in similar timeframes. They had to redesign and put the Apple BASIC in after that,” he says. Perhaps most telling is Byte magazine’s conclusion in a November 1982 article, which credits the PET as the first personal computer. * * * A week after CES, Peddle was in Dallas, Texas for another exhibition. This time it was the National Computer Conference, held on June 13, 1977. The conference took place at the Dallas Convention Center, which had the equivalent of five football-fields worth of floor space for the 300 exhibitors. Displays were extravagant. The Data General exhibit comprised an entire planetarium capable of seating 50 people. The conference was three times the size of the West Coast Computer Faire, drawing in over 36,000 attendees. Peddle re the experience. “It was in the middle of summer in Dallas. The first two floors of this big convention center are set up for all the big computer guys. It’s a dog-and-pony show with IBM limousines and all that shit.” The very first National Computer Conference was four years earlier, in New York City. The conference regularly attracted such distinguished speakers as Douglas Engelbart, the inventor of the mouse. Normally the conference was the domain of the larger computer companies. The industry commonly referred to

the big eight computer companies as “IBM and the seven dwarfs,” which included Univac, Burroughs, Scientific Data Systems, Control Data Corporation, General Electric, RCA and Honeywell. Other major attendees at the show included Memorex, National Semiconductor, Datapoint, and Sperry. “NCC was a big computer show,” recalls Bill Seiler. “They had big mainframes and all those big companies were upstairs in these huge booths with salesmen in threepiece suits.” This year, however, the conference was trying out something new. “The guy came up with this idea for what he called the personal computer boutique,” explains Peddle. Organizers created an exhibition hall just for microcomputer companies, which they called the Personal Computing Fair, held in a separate area below the main floor and away from larger computer companies. “The personal computer thing just started happening and they stuck us in the basement,” recalls Seiler. “Diagonal across from us was Apple. There were maybe 50 little vendors down there.” If the big computer companies had not been aware of microcomputers before, they would surely take notice now. Prior to the show, Tramiel and Peddle discovered a problem with the business model of the 4K PET. “It turned out that we really couldn’t make much money on that one,” says Seiler. In response, the engineers created a PET with eight kilobytes of memory. The new model was identical to the four-kilobyte model, except for the additional memory. Commodore stopped taking preorders for the 4K PET. “So they sold an 8K PET, which was $795. Everybody upgraded to that one, they wanted a computer so bad.” Before the show, Tramiel had instructed Peddle and his wife Shirley to take preorders for the PET. It was the first trial of Tramiel’s idea. “We were down in the basement and I’ve got four machines working literally the day before the show.” Peddle also managed to persuade an editor from Electronic Engineering Times to cover the show. “We had gotten a lot of good publicity when we did the 6502. I knew one of the editors real well and he said, ‘What are you doing?’ I told him about the computer and he said, ‘It doesn’t exactly fit with what we are doing, but why don’t I do an article on it?’”

“So he comes out and he takes this picture of my wife,” explains Peddle. “She is built like Zsa Zsa Gabor. She’s holding the computer and it wasn’t intended to be a pinup, but it was.” The magazine later published the picture, which became popular with computer s. According to Peddle, “People would come to the show and ask her to please sign this thing, saying, ‘I keep it in my office. Everybody will be totally jealous.’ She used to love it.”

Shirley Peddle demonstrating the PET at a computer show.

The Commodore display was a simple counter with a white tablecloth surrounded by blue curtains. “We’ve got these four machines lined up on this counter top. I had three or four people all operating the same program on the machines. I had Leonard there and people like that who were showing what you could do while I was giving the pitch.” Curiously, Commodore was still displaying the wooden PET prototype. “We had the wooden PET and the two metal ones where the lid opened,” says Seiler. “With the metal ones, we didn’t have the cassette thing working. We had to type in the BASIC program if you wanted a demo. So we’d spend an hour before the show and put in some type of stupid program that would do something.” Software was non-existent in 1977. “We had nothing; just trivial stuff,” recalls Peddle. “We would show them you could program in BASIC. This was way before software.” Only the wooden PET had a working cassette unit, so it ran the most sophisticated software. “The wooden one was the one we would play blackjack and do some real things because the cassette worked on there,” recalls Seiler. “I [Steve] Jobs coming over and us arguing about the merits of our cassette.” Although the PET had an IEEE-488 interface, Commodore had yet to make any peripherals such as printers and disk drives. Instead, Peddle connected the interface to an oscilloscope. By the time he attended the National Computer Conference, Peddle had refined his sales pitch to an art form. In many ways, the engineer felt more like a performer. “In the beginning we just kind of showed them how the machine worked and they had to gather around. I’m giving these demos, standing up giving these pitches and getting them all excited.” The attendees saw the opportunity to own their own computer. “You’ve got to

understand the hysteria that had built up over time,” recalls Peddle. “They saw a packaged machine they could use.” At the end of his pitch, Peddle enacted Tramiel’s preorder plan. “This is just like a medicine show. I would say, ‘If you want to buy one, the pretty lady over here will take your money and we will ship to you in 90 days or we’ll send you your money back.’” To Tramiel’s surprise, the crowds surged forward. “These people are five deep, handing checks forward, saying ‘Put my check in first!’” recalls Peddle. “Jack was absolutely right. These people just fought to give my wife their checks. She would give them a receipt with a number on it that said that was the order in which we were going to ship to them.” One of the first orders came from Personal Software founder Dan Fylstra, who ordered his PET at the National Computer Conference. Months later, Fylstra received unit number 17 and wrote a positive piece for Byte magazine. “The crowd is so big and she’s taking so long to collect the money that I got a half-hour break,” recalls Peddle. “So I take this opportunity to walk upstairs on the last day of the show.” The third floor contained the exhibits for the big computer companies. Peddle was shocked at what he saw “The entire floor above me was empty of customers. All the big computer guys are sitting there talking to one another. There is nobody on the fucking floor.” “Meanwhile, you can’t get into the boutique. Downstairs, the guys who are supposed to be at the show are fighting one another to give me a check so they can get in queue. It was the death knell of the big computer.” Early ments of the period show Commodore going directly after the big computer companies. One brochure reads, “In a league with IBM, HP and Wang minicomputers. The PET should not be confused with game products that hook up to household T.V.’s. It is a minicomputer.” “If you had taken a snapshot at that show, you knew what was going to happen to the computer industry,” says Peddle. “I was walking among the dinosaurs while they were dying. They had lost the momentum. The momentum was building under their feet and they didn’t have a fucking clue.”

* * * Tramiel’s preorder strategy worked better than even he could have imagined. In the months that followed, orders from magazine ments poured in, along with personal checks. According to Fortune Magazine, “The ads appeared, and a hugely encouraging $3 million in checks came back.”[4] Commodore now had more than enough cash. In early 1977, there were approximately 250 computer stores across North America[5] and the number was rapidly growing. Tramiel wanted to tap into these dealers to sell the PET. Commodore had been receiving a measured number of dealer calls, mostly from those who attended CES and received Chuck’s business card. After several articles, Commodore began to experience massive interest in the PET. On some days, the receptionist handled between 40 and 50 dealer inquiries. This was much more than Commodore had planned for, and soon Commodore could only promise shipments in four or five months. To the Commodore dealership program, dealers were required to have a service technician, a retail storefront, and a good credit rating. Once again, Tramiel hoped to leverage money any way he could in order to produce the PET computers. Dealers were required to send a cash deposit for their computers. Likewise, dealers collected prepayment for PET computers from their customers, just like Commodore. As a result, money flowed into Commodore before they assembled and shipped a single computer. * * * Tramiel had a reputation as a hard businessman. “A lot of people thought he was too damn tough,” says Dick Sanford. “But if there’s anybody you ever want in a foxhole with you, you want Jack Tramiel. You do. The guy might have been a tough dude to work for, but he was as smart as a whip.” Aside from his immediate family, Tramiel kept a circle of close associates. “There was a term called the inner family,” explains MOS Technology engineer Charles Winterble. “He cared about those people in the inner family. You went to the ranch, you met his wife, and you sat down for drinks after all the business was over.”

The Godfather movies, which were popular in the mid-seventies, inspired Tramiel’s concept of a family outside his own family. “It was the infrastructure that ran the company outside of any vice president or operational titles,” says marketing manager Michael Tomczyk. “These were the key people that he trusted. This family, they were trusted to do almost anything.” Tramiel’s inner family was composed of Commodore employees from all over the world. Some were general managers and some were rookie engineers. “Each country had a few people,” says Winterble. “Just because you were a GM didn’t mean you were part of it. You were sometimes in and sometimes out.” According to Tomczyk, “He invented this because he was Jewish and he had strong feelings about the concept of family. He had a group of the top 20 people in the company. He literally called them the family.” Tramiel only allowed entry to those who earned his trust. “That was just how Jack was,” says Bob Yannes. “As far as I know, he rewarded them well. The people who were in the inner circle were very wealthy.” Dick Sanford, Tony Tokai, and Kit Spencer were all inner family . With the success of the PET computer, Chuck Peddle seemed like he was on the verge of ing the inner family. “There were probably a dozen or more guys like that in Commodore who worked pretty closely with Jack,” explains Spencer. “I worked with Jack for well over a decade. After a while, you get a lot of ‘credit in the bank’. You get fairly close. You get what you call family with Jack, like a circle. He might haul you over the coals or he might give you a bad time, but you’re not going to get fired; you’re going to be around. You understood that after a while.” Inner family had much more freedom than other Commodore employees did. “They were trusted to just run, and once in a while he would yank our chains and ask what we were doing,” says Tomczyk. “But he didn’t micromanage us.” Those outside of the trusted group found less freedom in their jobs. “He micromanaged everybody else like crazy and drove them nuts,” laughs Tomczyk. * * *

In the summer of 1977, Star Wars appeared in movie theaters across North America. Suddenly, the future seemed cool and the public was primed for the PET 2001. With the production design finalized by July, Peddle and his team began manufacturing PET computers. “That’s the real work of it,” says Bill Seiler. “You’re trying to define the product so it can be built reliably, repeatedly. I Chuck once said, ‘The only thing engineering really generates is paper. They don’t make anything else, they just make the paper that defines the product.’ It’s all drawings and things.” Despite the incredible demand, there would be no large manufacturing plants dedicated to PET construction. Neither was there high profile advertising, and the reason was Irving Gould. Commodore lacked manufacturing capability at the time. “We didn’t have a production line at Commodore to build anything,” says Seiler. “They were pretty much just an OEM (Original Equipment Manufacturer) reseller at that time. They were buying all their stuff prebuilt out of Asia.” While Peddle and his engineers designed the PET, Petr Sehnal dedicated his time to figuring out the manufacturing problem. “Petr was more running around trying to think about what it means to build a production line and get that going,” explains Seiler. “They knew they were going to try to build PETs first in America. Petr was starting to get involved there more.” The prepayment plan allowed Commodore to survive but Tramiel wanted more than to survive. He wanted to dominate. Tramiel wanted to raise investments in Commodore to quickly ramp up production, which would have satisfied both the customers and the ever-important dealers. Unfortunately, Gould did not hold the same philosophy. Peddle recalls, “He was one of the things that frustrated Jack more than anything else in the world because Irving wouldn’t let Jack go out and raise money.” Without short-term capital, Commodore would limp out of the starting gate. With a little advertising, Commodore could raise demand to incredible heights. Instead, Gould’s restraint slowed down production. Commodore could not attain market share fast enough compared to their rivals. Irving Gould was about to lose Commodore’s chance to grab a massive early lead in the North American

market. Perversely, Gould seemed to take pleasure in exerting control over Tramiel. He later told Forbes magazine, “I wouldn’t give them enough money to take any real risks.”[6] According to Peddle, “Jack was not really justifying it. Jack was just saying, ‘You know, you need to give me more money.’ Irving’s thought was, ‘You’ve always done better when you didn’t have a lot of money.’” With the preorders for the PET, Dick Sanford was able to go back to the banks to secure loans for production. “We had to build up its nest egg before the banks would give you enough money to build the next line,” he says. After a hard battle, Commodore was able to secure a loan for a minimal factory setup in Santa Clara. “We basically took a little factory over on Reed Street near the airport in San Jose,” says Peddle. Engineering manager Bob Skyles managed the technicians at the Reed Street location. Manufacturing was new to the engineers and the new operation did not go well at first. “The very first production was down on Reed Street in some warehouse,” says Seiler. “It was a disaster. We almost died building the first PET.” Seiler made a list of parts Commodore would need in large quantities. “I had to get the schematics under control and make sure we got all the parts out on the bills—all the grunt stuff you have to do in engineering,” he says. To meet the low price-point, it was vital to get the best prices on all the parts going into the PET. The buyer for these parts was John Calton, who had been with Commodore since arriving as an Irish immigrant in 1954. As the longest-serving employee at Commodore, aside from Tramiel himself, Calton knew Commodore inside and out. He had absorbed Tramiel’s philosophy and had risen to one of the most important ranks in the company. His strategy for buying from outside vendors was simple: know their manufacturing costs before negotiating. He would meticulously research a company before even beginning to talk about price. As a result, Calton knew exactly how low the supplier could go.

Dick Sanford had the unenviable task of making sure Commodore had funds to buy the thousands of required parts. “In any organization that is growing very quickly, no matter who you are or what you are, you have cash flow issues,” he says. The biggest problem was often the lag between sending product out to retailers, and waiting for revenues to return. “You’ve got to ensure that the vendors are supplied. You’ve got inventories all over the world, and we were dealing with shipping and the timing of inventories. There were various component trees for our facilities.”

Commodore’s hand painted sign in front of the Reed Street factory.

Commodore already had plenty of experience with circuit boards, having manufactured millions of calculators. Always searching for a lower price, Calton found a way to obtain the circuit boards cheaply. “They bought the time at some PC board house after hours to get these boards built cheap,” says Seiler. “They usually replace the plates the next morning, and the plating was flaking apart by that time. The flakes were floating around in the emulsion and sticking to the boards.” Making a printed circuit board (PCB) shared similarities with making a chip. A light shone through a stencil, hardening a surface coating on the board. Workers then washed the board with chemicals, dissolving all the copper except for the designated electrical paths. Workers then drilled hundreds of tiny holes into each board, into which they would insert resistors, transistors, and IC chips. Every board produced by the vendor contained small flakes in the circuitry, which caused short circuits. “You couldn’t see the short, it was so fine. It was a little micro short,” says Seiler. The low-quality boards would haunt the engineers for months. Commodore’s computer manufacturing of the 1970s was relatively primitive, lacking computer-automated assembly lines. Dozens of round tables sat on the clean, white tile factory floor. “The one at Reed Street was just crazy hazardous,” says Seiler. “There were just people at round tables putting parts together.” Each table had tool-racks filled with pliers, screwdrivers and other equipment. Here, workers sat hunched over empty circuit boards, stuffing them full of components. “They didn’t have machine stuffing back then, they were hand stuffing the boards,” says Seiler. Parts came from all over the world. “He had a couple of manufacturing guys there,” recalls Peddle. “We went into that building and started getting the cases out of Toronto and building the boards one at a time.” Most of the factory workers were either Mexican or Korean immigrants, chosen

because they worked for cutthroat wages. “We slowly set up a small production line,” recalls Peddle. Workers would grab a resistor, bend it to shape, and insert the two wires into the appropriate holes on the circuit boards. Once the boards were full, they went to a wave-soldering machine, which quickly bonded the parts. “The PET was just one big board on the bottom,” says Seiler. “There was a wave-soldering machine on the motherboards.” When the assembly workers completed a board, there were often errors that technicians needed to identify and correct before the computer went onto store shelves. “Sometimes they would bend a pin under, and you wouldn’t see it. Instead of going through the hole, it would get bent under,” explains Seiler. “Nobody would see it because it was under the chip. It would make with the circuit and the board worked fine.” Because the pin was not properly anchored to the board, the connection would not last long. “It would get dirty or corroded or thermal expansion and contraction would make it disconnect and the board would crap out,” says Seiler. To ensure the motherboards did not stop working mere hours after customers turned on their computers, the boards underwent a burn-in period. “I worked on the burn-in rack,” says Seiler. “It was this great big rack with all these doors on the front. In the big burn-in rack there were all these big honking transformers just like in the PET. We just took all the PET parts and duplicated them. These burn-in racks would take hundreds of boards, They would sort of slide in on a thing and they would plug them into the power supply and they would burn in for 24 hours. We were real big on burning everything in and making them more reliable.” For additional reliability, Seiler’s burn-in rack constantly cycled the temperature inside the rack from hot to cold. “It’s better if you thermally cycle because you have all the expanding and contracting of all the components on the board,” he explains. “You want to stress that because that stresses the chips. Usually what happens in the chips is they break a bond wire inside.” After 24 hours, if the motherboard still worked, it was good enough to put inside a PET case. “Some would fall out. You would go through the thermal cycling and burn in and it would be a connector that would eventually stop working,” says Seiler. “Inevitably once a day, one of the motherboards would be built

wrong and we would stick it in there and they would all go up in smoke and it would smell up the place.” It was up to the technicians to diagnose and repair the boards that failed the burn-in period. “We had a real good staff of technicians fixing things,” says Seiler. The engineers soon learned that some MOS Technology components did not last very long in the PET. “In the first PET we used the Commodore RAM and it almost did us in,” says Seiler. “Normally you burn parts in for 24 hours and the ones that are going to die will die. You get down this exponential curve. There are a few way out there that are still going to die, but you pretty much eliminated all the ones that the customer is going to see.” MOS Technology’s RAM, on the other hand, never achieved a plateau of reliability. “That curve never got to zero with the parts from MOS Technology,” says Seiler. “I think there might have been a heat issue and they kept dying. Customers were always having memory failures out there. Chuck said they didn’t know how to build RAM very good.” Tramiel’s vertical integration strategy was in jeopardy. The manufacturing problems were a constant source of pain for Dick Sanford. “, we don’t have a lot of money to throw away,” he says. “Chuck and his team at MOS, these guys start making chips, and they’re always looking for perfection. It’s interesting. You make the chip and you start this inventory. You’re coming down this line with these chips, and all of a sudden, oops, there’s a design flaw. And they go figure out something to fix that. Well what happened to all that inventory that started last week or two weeks ago or four weeks ago? You’ve got to throw it away and start it all over again.” The delays in manufacturing resulted in delays shipping product, which compounded Sanford’s cash flow problems. “We’ve got deliveries to make by next week or two weeks from now,” he says. “What do you do now? Tell your customers, ‘Sorry, it ain’t coming.’ Now, if you’ve got all the money in the world, and you don’t care, that’s one thing. But when things are tight, and you’ve got payroll to make, those were tough days back then.” Once the technicians verified the board, workers inserted it into a metal PET case from Toronto. “There were people putting the chassis together, bolting in

the power supply and the transformer,” says Seiler. “The motherboard was just plugged into this big transformer. Then there was a top board in the display.” Slowly, Commodore engineers figured out the manufacturing process, but it would be weeks until they shipped computers. * * * Normally Tramiel drove his workers hard to produce computers, but he had become distracted by personal issues. In 1977, Tramiel’s daughter-in-law became ill. “She gets this cancer and they take her to a whole bunch of treatments and everything else,” says Peddle. “It’s supposed to get better.” During her treatments, Sam refused to voice hope for a cure. “Sam says that he never told anybody that, because he didn’t believe it,” recalls Peddle. Sam had good reasons for his pessimistic belief, since he had participated in discussions with the doctors that others had not. “He thought, from what the doctors had said, that the stuff they did was going to make her well for a year and then she was going to die.” “Jack and Helen went through this big depression with the daughter,” says Peddle. “I mean, it was a big goddamned deal to them.” In late 1977, it seemed like there was hope for the family. “She suddenly popped back, and everybody believed that she was going to make it,” explains Peddle. “When she got better, there was this big celebration and they took them all over the world. There was this whole thing of joy.” The world travels took place sometime in late 1977 and 1978. Curiously, Sam refused to believe his wife was safe. “Sam, apparently, totally believes that all he was doing was making sure she lived the last year of her life as well as she can,” says Peddle. [1] The Pong systems did not catch on. Pong fell out of favor following the release of the Atari 2600 and Commodore ended production. [2] The FCC has two requirement levels, Class A and Class B. A computer that meets the FCC Class A requirements may be marketed for use in an industrial or a commercial area. A system that meets the more stringent FCC Class B requirements may be marketed for use in a residential area in addition to use in an industrial or a commercial area.

[3] Wozniak failed to complete his degree at the University of Colorado. He later went back to university and became a full-fledged engineer. [4] Fortune magazine, “Jack Tramiel is Back on the Warpath” (March 4, 1985) p. 46. [5] Based on the number of Byte magazine’s direct dealer sales outlets. [6]Forbes magazine, “Albatross” (January 17, 1983) p. 48.

CHAPTER 9

Trinity 1977-1978

The biggest entry in the early microcomputer industry was the TRS-80 from Texas based Tandy Corporation, owners of the Radio Shack chain. John Roach, the man who viewed Chuck Peddle’s PET 2001 prototype at the January 1977 CES, led Radio Shack into computers. According to the official Radio Shack history of the TRS-80, a Radio Shack employee named Don French initially conceived the idea of selling computers in 1976. By December 1976, the designers had the go-ahead to develop a computer for the chain. This version of history seems peculiar, since Roach gave Commodore a bid for the computer contract at the January 1977 CES. Perhaps John Roach had not intended to use the PET at all and just wanted to evaluate the competition. According to the history, engineers finished their handmade prototype by January 1977.[1] The prototype ended up costing Radio Shack less than $150,000 to develop, including case tooling. The marketing department assigned the computer an overly optimistic $199 price tag and conservatively predicted sales of 3,000 units annually. The Radio Shack computer had similar features to other computers of the day. The microprocessor was a Z-80, an 8-bit microprocessor made by a company called Zilog. The computer featured 4 kilobytes of RAM (expandable to 16 kilobytes), 4 kilobytes of ROM, a 53-key keyboard, a 12-inch black and white RCA video display, and a cassette recorder. According to Peddle, the design was lacking compared to the PET due to the

ROM size. “We built in a big enough ROM to get the BASIC, the operating system, and all that stuff in ROM, which was one of the key things to make it work.” Commodore achieved early success with its ROM chips and outclassed the competition. “We could get some really high density ROMs, so we gambled on a much bigger ROM than Radio Shack did, which ultimately caused them a lot of trouble.” As a result, the TRS-80 lacked many features in the PET, such as lower case characters. Radio Shack increased the price of the TRS-80 from $199 to $599.95 before release. Unlike the integrated PET, the TRS-80 was modular and buyers could purchase the components individually: the microcomputer for $399.95, the video display for $199.95, and the cassette recorder for $49.95. On Wednesday, August 3, 1977, company president Lewis Kornfeld unveiled the TRS-80 Model I in New York City’s Warwick Hotel. “The TRS-80 wasn’t a bad computer. It had a number of nice features,” says Bob Yannes. By the time Radio Shack released the TRS-80, Commodore had enjoyed sales since June 1977 and even had some software to demonstrate for the PET. “We showed Microchess at the East Coast Computer Faire when Radio Shack launched their computer,” says Peddle. At the time of the introduction, Radio Shack had only managed to build 25 TRS80 units, which was fine since they had predicted annual sales of only 3,000 units. Radio Shack planned to release the computer to a third of its 3,500 company-owned stores. Although there was virtually no news coverage of the release, the marketplace responded overwhelmingly. More than 10,000 orders came in within a month of the introduction. Radio Shack had a hit on its hands. The response to the TRS-80 not only surprised Tandy, but also Commodore and Apple. Apple had always thought it was in a race with Commodore. Commodore thought it was a race against IBM. Radio Shack showed both companies it was a race for mass distribution. “Radio Shack [success] was understandable because they had Radio Shack stores to sell them through,” explains Yannes. Radio Shack was in every city and mall where the entire population of North America could see it. Once shoppers caught a glimpse of the curious new machine, they could not resist walking into the store to learn more. After a few

minutes at the keyboard, many walked away dazzled by the promise of computers. It was a powerful marketing tool. Bill Gates, in a 1993 interview with the Smithsonian, concludes, “Radio Shack, with its distribution and its name, set the market on fire.” All Commodore had to compete with Radio Shack were Mr. Calculator stores. Mr. Calculator was not really working out for Commodore and it could not hope to match Radio Shack in both distribution and expertise. The only good news for Commodore was that Radio Shack was also having problems manufacturing the systems fast enough to match its orders. For a long period, each Radio Shack store received only one TRS-80 to sell. s affectionately called the TRS-80 the Trash-80 for some of its quirks and limitations. The tape system was a horror to use and the Z-80 microprocessor was not even running at full speed. The monitor also displayed static due to a lack of proper shielding. The BASIC language in the Radio Shack computer was also inferior to the PET. “John Roach introduced his computer with Tiny BASIC [a free BASIC developed in California] because his ROM wasn’t big enough,” says Peddle. Roach soon decided he needed to improve the BASIC language in the TRS-80. “He was trapped and he actually went to Gates,” says Peddle. “We were kicking his ass, and Apple was kicking his ass because BASIC was what people were buying.” “Now Bill has got John and he sees that things are going, so he figures he’s worth more, because he knows John has got to compete with us,” recalls Peddle. “So he doubles the [Apple price]. Roach basically says ‘I don’t want to pay you that number. I don’t know how many of these we are going to sell. Why don’t you take a piece of the action.’ Gates said, ‘No, I want a fixed price.’ That was the last time Bill took a fixed price.” Although Gates missed lucrative royalties, he was triumphant nonetheless, having sold his premier programming language to all three major computers released in 1977. “These machines drove the market and eventually, a year after they were out, all of them had our BASIC built-in,” recalls Gates.[2] * * *

The PET release coincided with much of the news coverage that had been building around microcomputers, since there was a three-month delay after CES for magazines to publish their coverage of the event. Byte magazine featured an early preview of the PET by Jeff Raskin. The fully integrated design of the PET must have influenced Raskin. Years later, he went on to design the Macintosh computer for Apple, one of the few desktop computers to include a built-in monitor. Peddle gave interviews to the media, along with his wife Shirley. “We were interviewed on TV and all kinds of crap. We were like movie stars.” Peddle also noticed a subtle attack on computers from the media. “I had made a statement that computers were going to change the way the classroom worked and that they were going to be a very important part of the classroom.” A San Francisco station heard the comments and decided to create a controversial news story. “The local news channel in San Francisco found out when I was arriving from Toronto and met me at the airport with a camera. They wanted to discuss this comment about schools. He said, ‘Are you on record that computers are going to replace teachers?’ I said, ‘It will augment teachers but not replace them.’” The news station turned the story into a conflict between teachers and computers. “It turned out to be total bullshit,” says Peddle. “They wanted to pick up on the line, so they took some pieces out of my quote. Then they ran this old school teacher saying kids still need the warm, loving attention of a school teacher to teach them how to do things.” The story provoked animosity against computers. “This was at a time when people didn’t believe in computers very much. They would take the piss out of computers on television. It was a hot topical comment.” The PET received better coverage in Popular Science, the magazine that launched the Altair 8800. The editors ran a feature listing the most popular microcomputers of the day, including the Altair 8080, the IMSAI 8048, the Processor Technology Sol, the CompuColor, and even the Radio Shack TRS-80.[3] The editors were impressed with the appearance of the PET and featured it on the cover. The image showed the PET sitting on a kitchen counter, with a recipe for Oriental Salad on the screen.

Although there had been many orders for the 4-kilobyte PET, it became clear the memory was severely limiting. Commodore wisely began discouraging orders for the 4-kilobyte version in favor of the 8-kilobyte version, which cost $795. The opportunity to purchase a computer for personal use seemed incredible. Bill Gates, in a 1993 Smithsonian interview, saw the PET 2001 as a landmark machine for the pricing as much as the technology. “The Commodore machine, the PET, was actually the most aggressively-priced machine,” he recalls. “It had some very innovative things.” One of the major questions of the period was what to name these new computer systems. They were definitely not kit-computers, nor were they hobbyist computers. The term microcomputer was no longer adequate to describe these easy-to-use, fully integrated computers. Some writers felt there was no distinction between the more expensive minicomputers from DEC and IBM, so they used the term minicomputer to describe the PET. Byte contributor Dan Fylstra tried the term television-typewriters, which he used to differentiate between computers with a picture tube and computers with simple lights. Another name was appliance computer, which gained some early acceptance due to a Popular Electronics cover of a PET 2001 sitting on a kitchen counter. Byte began using the term personal systems, which came close to describing the new computers. All these referred to the same thing, and eventually the term personal computer became accepted. Writers were also unsure what to call people who used these computers. Byte magazine frequently used the word computerist, a clunky word that never caught on. Other writers preferred the term computer hobbyist, enthusiast, or hacker. Eventually computer or simply “” came to signify anyone who used a computer. Technical publications were having their own problems understanding the technology. Popular Science, a magazine that should have known better, was sometimes comically wrong. The reviewing author did not understand the function of the cassette unit in the PET. He believed the tape player was for audio cassettes to teach people how to use the computer. He also thought the monochrome monitor would display photos, something that would only be possible years later on far more powerful computers.

* * * Fully assembled PET computers started coming together in the small factory on Reed Street in September 1977. By mid-October they had boxed 100 computers, enough for a full shipment. It had taken Peddle and his small team less than a year to conceive, design, and manufacture their computer. The PET was a hit with schools. “They went to schools and the kids beat on them,” says Seiler. “I think the early people really appreciated the ruggedness of the PET because everybody kind of carried them around and banged them on things. They were made really tough.” Despite some early problems, the PET was destined for success, and the stock market took notice. “Commodore’s stock was running because people were starting to catch on that this is going to be a big deal,” recalls Peddle. “Two or three months later, we’re having trouble shipping and we’re not shipping on time.” Commodore began defaulting on its 90-day delivery guarantee, and Jack Tramiel became worried people might demand their money back. To Peddle’s relief, “Nobody ever asked for their check back. Never.” However, some customers became impatient and aggressive. One day an investment analyst wandered into the company and told Peddle, “I’m going to go back to New York and tell the world you are lying about shipping computers. You are just a stock fraud like Viatron.”[4] Peddle replied, “Well gee, it’s really not just for show. Let me take you to the factory and I’ll show you I’m shipping machines.” The distraught analyst had ordered a PET and he gave Peddle his queue number. “His order was another two or three weeks away,” recalls Peddle, who had become nervous about bad publicity. “I figured it out. I said, ‘I’ll tell you what. I’ll put your number on this machine.’”

A row of PET computers awaiting final assembly.

The gesture caused a sudden reversal in the demeanor of the New York analyst. “Suddenly he was the sweetest guy in the world! He was just there to give me shit so he could get his machine.” The analyst was satisfied and returned to New York. “The guy ultimately gives us a great recommendation and the stock runs. And he goes and signs up as a dealer with us.” Commodore was able to produce less than 30 PET computers per day during this period; far below the demand. With frantic dealer calls coming in every day, it was an excruciating situation. In 1977, there was no need for a warehouse yet. “We weren’t really in very heavy production until the next year,” recalls Peddle. According to Dick Sanford, one of the main limiting factors on PET production was the lack of cash flow for components. “If you get a demand for a million pieces, how do you make a million if you only have the money for 300? What do you do? There was a finesse needed to move cash around the world and get the whole thing started. It was very complicated.” Commodore still had stacks of unusable circuit boards lying around, and the engineers decided to see if there was any way to salvage them. “We bought all these boards that had not been properly short tested,” he recalls. “I we spent Christmas sorting our way through these shorted boards with low resistance ohmmeters. It was me, Seiler and a couple of other guys.” “The vendors were just goofing us up,” recalls Seiler. “I finally sat down with one I knew had shorted. After I found 50 shorts on it, I gave up. We eventually chucked those out and bought good ones.” With the production line mostly figured out, Tramiel started two other factories in Japan and . Although much of Commodore Japan’s product would feed the European market, the PET also sold to the Asian market. “We also launched it in Japan because we built the product in Japan at the beginning,” says Peddle.

However, Tramiel used the Japanese operation mainly for obtaining cheap electronic components. “We didn’t have a large development operation in Japan at that time, more a sourcing operation,” says Spencer. Selling computers in Japan was an afterthought. * * * Due to the circuit board fiasco, longtime employee John Calton was about to receive a severe reprimand from Tramiel. Calton had worked for Tramiel since he opened his typewriter repair shop in New York in the 1950s. “He had been with Jack since he was an Irish kid and he was totally loyal to Jack,” says Peddle. Now, Calton was the corporate buyer for the company, just a few doors down from Tramiel’s office. With cost cutting so central to Tramiel’s strategy, Calton held a key position at Commodore. “John wasn’t the sharpest guy in town, so every so often he would screw up,” says Peddle. “Jack didn’t really run his business like a business,” says Yannes. “To him, business was war, and his employees were troops, and if you got close enough, you were like family.” For Tramiel, it was important to maintain discipline within his troops. “Maybe his style was beating on the table and embarrassing people, which is not the way you should do it according to whomever,” says Sanford. “But come on! With no money, you’ve got to surround yourself with talented people and selectively take some out.” Although inner family received reprimands, Tramiel rarely gave up on them. “If Jack would get pissed off at you for something and you were a loyal employee and you were going to be there forever, he would send you away on these long extended journeys around the world with some kind of an assignment,” explains Peddle, who refers to the sabbaticals as punishment tours. The concept was similar to a Japanese business practice called the “window seat.” When a Japanese employee made a mistake, the executives sent him to a less important role in the company until he learned from his mistake. The punishment tours were also similar to the work tours Tramiel witnessed in Polish work camps. “He could be pretty ruthless,” says Kit Spencer. “To

understand Jack, you need to understand where he had come from, and that was the concentration camps. He came to me once and said, ‘A written deal doesn’t mean that much. Lawyers can always find a way around that, but a handshake with somebody I trust means something. Laws didn’t help me survive in , but trusted people did.’” “Maybe people thought his style was incorrect, but there’s a guy who came out of the holocaust and was really a true survivor,” says Sanford. “He really was.” Whatever Tramiel’s inspiration for this concept, he used it regularly. “The idea was that they were getting punished for not doing what they were supposed to do,” says Peddle. “It was an interesting punishment. You weren’t fired; you weren’t sent off into the corner; you were actually sent away from your family and sent away from everybody. Sam [Tramiel] got some of that, too.” “That was a huge dichotomy and it was hard for people to understand that Jack could be this autocratic, Napoleonic, Machiavellian dictator on one hand, and yet on the other hand he could cultivate an infrastructure of trusted insiders who he allowed to follow their instincts to great advantage,” says Michael Tomczyk. Calton was having trouble keeping up with Commodore’s sudden changes. “Some people couldn’t deal with the growth that was coming all of a sudden,” says Dick Sanford. “You’re walking along in this $45 million company, and all of a sudden this thing starts to skyrocket and explode in a whole new industry, and people don’t know how to adapt. They think they can do it, but they can’t do it. So you have a lot of people get turned over in those kinds of environments. They think, ‘Woe is me. I was here when it was doing so well’, but you can’t handle the speed and the growth.” Calton received a punishment tour for the circuit board fiasco. “The next thing you know, you hear about John, and he would be here for three months; then he would be over there. Sometimes he would leave the United States and leave his family for a year. He was on these punishment tours for Jack,” says Peddle. By the end of 1977, Commodore had only managed to assemble a meager 500 machines. “That’s a funny volume compared to today’s world,” remarks Peddle. In their review, Byte magazine commented on the poor availability of the PET. Tramiel set about replacing Calton with a purchasing agent who could find reliable parts and ultimately increase PET production. It was a difficult position

to fill because he needed someone he could trust, and he trusted no one more than Calton. “Those guys are easily corrupted because there is a lot of money moving back and forth,” says Seiler. “If they approve a part from a certain vendor, and it’s a big company that’s buying millions of parts, that’s a lot of money. Vendors will hand them money under the table and say, ‘Take my part!’” Tramiel soon found an able replacement. “Dick Barton was our purchasing engineer. He was really good,” says Seiler. “He would find parts and things for us, vendors had second sources. I’d ask him for something and before the end of the day he had samples there. He had all the vendors so terrified they would come running over to give him parts for us engineers.” Soon, Commodore would begin turning out reliable PET computers in large quantities. * * * The jewel of Commodore International was the British division, known as Commodore UK. The headquarters were located in an industrial estate 35 kilometers East of London in Slough, Berkshire at 675 Ajax Avenue. The offices were old and worn, with walls painted a dull yellow. Many offices had windows that pointed not outward to the English countryside, but inward towards storage aisles and calculator assembly lines. Despite the drab workplace environment, Commodore UK was teeming with enthusiastic, motivated people. Their motivation came from the privilege of working for one of Commodore’s top executives, Kit Spencer. “He started in the sales and advertising area,” says Leonard Tramiel. Christopher Spencer, known as Kit to those around him, was the competitive and charismatic sales manager of Commodore UK. Born on October 10, 1945 during the close of World War II, he grew up in Sheffield, a large steel town he describes as “the Pittsburgh of England.” Spencer received a physics degree from the University of Sheffield. “Solid state physics was just coming in at that time, but only very little,” he says, in reference to transistors and semiconductors. With his rapid-fire speech and Michael Caine accent, he stamps out one thought after another like a factory machine, occasionally pausing to hyperventilate (his version of laughing). Even in retirement, his mind actively churns out marketing strategy. Ask him almost any question and before long, he is pondering the

marketing implications. Spencer’s slim, athletic build and abundance of energy made him a natural at sports. “He liked to play tennis,” says Leonard. Spencer played at a competitive level, holding trophies in several divisions and even competing in the Junior Wimbledon tournament. Spencer took his natural desire to compete into the business world, first working for Philips, the largest European manufacturer of electrical products at the time. When calculators became popular, he jumped over to Bowmar, a major competitor of Commodore in the early calculator industry. He spent two years as their UK marketing director before Bowmar gave up on calculators. Then Spencer ed Commodore, where he marketed its early calculators. With Commodore on the verge of introducing a revolutionary new product, Tramiel wanted Spencer to push his UK marketing to a new level. “When we went into the computer business after the first microcomputer, Jack said to me, ‘I think this is going to be a very big business. I’d like to experiment in the UK and find out what we can really do in it.’” “He asked me to start up a separate computer division when we launched PET in the UK,” says Spencer. Although Commodore UK already had a general manager, Tramiel created a new microcomputer division with Spencer as a second GM. “I took over that with profit responsibility for the Commodore PET. I was relatively free to make my own decisions.” Tramiel was not quite ready to give up on calculators, but he was beginning to see the potential of computers. He told Spencer, “I want you to keep an eye on all the things you’ve been doing well with the calculators and watches. But I want you to really work primarily as a General Manager getting this new microcomputer business running well.” According to Spencer, “Jack saw that this could be the main future of the company in the years ahead and it was an opportunity for me to create something new and potentially very large from the very beginning. A marketing man’s dream!” To familiarize Spencer with microcomputers, Tramiel sent Chuck Peddle to England for a visit. Tramiel instructed him, “Please make sure Chuck has a good time. He’s brilliant, but he’s a bit volatile and I need him.”

Spencer picked up Peddle and his wife Shirley from the airport. “He wanted to buy some English clothing so we took him to stores in London. He bought a Harris Tweed jacket and a long-neck cashmere sweater.” That evening, Spencer and his wife took Peddle and Shirley to an old-fashioned English restaurant called the French Horn. “A stuffy old English place,” says Spencer. Chuck was now wearing his upscale Harris Tweed. “We walked in there and they said, ‘You have to have a tie to come in here.’ It was bad of them because Chuck had on a jacket and all that.” Undaunted, Peddle said, “Give me a tie man,” despite not having a collared shirt. “So the waiter gives him a tie, which he ties around his roll neck sweater. Now he looks really silly, but we are going to have dinner.” The patrons sat down and began looking over the menu, ignoring that Peddle could not have looked more ridiculous had he been wearing a clown suit. “Chuck looked at me and said, ‘Is grouse on the menu? I’ve never had grouse. I should have grouse.’” Spencer replied, “You know, Chuck, grouse is very gamey.” Peddle stood firm, replying, “No, I want grouse.” Spencer was almost certain Peddle would not enjoy the meal. Already he feared he had failed Tramiel’s specific instructions to give Peddle a good time. “So here’s Chuck, who’s already made a little bit of a scene in the restaurant, saying he wants this grouse.” While waiting for the meal, conversation invariably turned toward the computer business. “We’re having interesting conversations and I’m learning about microprocessors and things,” recalls Spencer. Spencer remarked, “I better hire somebody who knows more than I do about this from the computer business.” Despite coming from the mainframe computer business himself, Peddle immediately replied, “Don’t ever do that. If they are from the traditional computer business, they won’t understand microcomputers. They have learnt certain rules and we are breaking all of them!”

The patrons continued their conversation, interspersed with anecdotes from Shirley about their marketing adventures with the 6502. Eventually the waiter delivered the grouse. Peddle tried a bite and immediately said, “This is off.” “Grouse just has that sort of taste,” says Spencer. “But this is Chuck.” Peddle sent the grouse back, much to the frustration of the restaurant staff, and they changed his meal. “Chuck, with his volatile self, escalated the situation rather than de-escalated it,” says Spencer. The French Horn was located in a beautiful location, so the couples decided to take in some evening air. “Eventually we finished the meal and we walked down to the grounds to the river,” recalls Spencer. “We got to the end and they switched the lights out on us.” The pitch dark forced them to grope their way back to the car. “Surely they did it deliberately,” he remarks. “I didn’t go back to the restaurant again because I didn’t think they handled it very well.” After the visit, Peddle returned to America. The two would next meet at the January 1978 CES in Las Vegas. * * * Following the departure of Bill Mensch, MOS Technology downsized the calculator chip department. Al Charpentier inherited Mensch’s previous title. “At the time, MOS Technology just had a big layoff,” says Yannes. “As a result of the layoff, Al ended up as the head of the chip department.” The design process for the VIC 6560 was similar to the process used by Peddle to design the 6502. “There was a team of people involved,” says Charpentier. “There was a logic designer, a circuit designer—generally, they would be the same person—and then a layout designer.” Much like the 6502 design process, the methods were primitive. “With today’s capabilities, it seems so crazy what we were doing 30 years ago,” he recalls. “There wasn’t the computing power, so it was all done by hand.” As usual, once the design was completed, Charpentier spent weeks measuring to make sure the different layers lined up. “After the layout was done, you had to

check it by hand and make sure all the interconnects were correct,” recalls Charpentier. Charpentier impressed his manager Charles Winterble with his dedication to the semiconductor craft. “It was amazing, the number of things he could carry,” recalls Winterble. “Back in those days, we slept on cots to get some of this done. These were very serious guys. Charpentier was probably close to a genius.” The 40-pin chip worked exactly as Charpentier hoped. “I had that chip operational in 1977,” he recalls. “We had the full palette of colors.” He even added a feature that allowed programmers to redefine the character set. On the PET, programmers could not change the fonts and character graphics, but the VIC chip allowed s to change all the characters. This feature would allow diverse graphics in video games. Charpentier hoped Atari would use the VIC chip in a game console, as they had with the 6502. “The original intent was that we were going to sell it as a standard piece of silicon and sell it to somebody who would want to make a product,” he says. “A videogame was the intent at the time.” However, Atari was not receptive to demonstrations provided by Charpentier. “MOS had developed a chipset to replace the 2600 chipset,” says Seiler. “It was supposed to be the next generation for Atari but Atari didn’t like it. It didn’t have quite enough of whatever they wanted.” MOS Technology would now attempt to sell the chip to anyone who was interested. “They were doing it for any market they could sell it to,” says Seiler. Commodore would unveil the chip at the 1978 January CES. * * * By the end of 1977, it looked as though Commodore might dominate the personal computer industry. The PET 2001 easily sured the first generation computers from MITS, IMSAI, Ohio Scientific, and Processor Technology. However, in the year Commodore released the PET, Apple released the Apple II and Tandy released the TRS-80. Byte magazine later dubbed the competitors the “1977 Trinity.” Engineer Bob Yannes witnessed the dawn of the personal computer while working in a computer store. “The big breakthrough, which came with the PET

and the TRS-80 (and to some extent the Apple II, although I won’t give them as much credit as they want), was the fully assembled computer.” Chris Crawford, an early game designer and PET owner, summarizes the 1977 Trinity. “A lot of people don’t realize that the Apple got off to a slow start. In the early days, the horse race was between the PET and the TRS-80,” he says. “The Apple was nowhere to be seen. It was too expensive to sell.” By the end of 1977, Apple had sales of $775,000 for the fiscal year, which included sales of the Apple I. This clearly put Apple in third place, after Radio Shack and Commodore. One of Apple’s earliest breakthroughs, and one of Commodore’s earliest missteps, occurred with the largest computer dealer in the United States, ComputerLand. ComputerLand stores initially sold mainly hobbyist computer systems, such as IMSAI, Proc Tech, Polymorphic, Southwest Tech, and Cromemco. The release of the 1977 Trinity should have brought both Commodore and Apple into its stores.[5] However, an early misstep by Tramiel prevented the PET from receiving wide distribution. ComputerLand ordered all computers for its franchises. Its business model required favored pricing to remain profitable, but Tramiel refused to give them a deal. ComputerLand was not happy. As a result, the chain decided to promote Apple II computers instead of the PET. In December 1977, ComputerLand d the Apple II computer with 4 kilobytes of RAM for $1,298 or $1,698 with 16 kilobytes of RAM. As the franchise spread across the continent, so did Apple. Though the chain could not hope to match Radio Shack’s proliferation, it gave Apple a solid market from which to build. Between Mike Markkula at Apple and Jack Tramiel at Commodore, it was clear Apple respected the computer distributors more. In a 1993 interview, Bill Gates identified this strength. “Apple … really went out to computer dealers and did a good job, far better than people like the MITS guys and the IMSAI guys. They really thought of this as a market where they had to develop the channel and do new things.”

* * * The struggles to produce an adequate supply of PET computers occupied most of the engineers time, but they also took some time to consider improvements to the PET 2001. “The product took off, except the universal response was, ‘We want a regular keyboard,’” says Peddle. Unfortunately, the chiclet keyboard was not as durable as a real keyboard. “They’d wear out in a couple of months,” says Seiler. “They were just anodized aluminum that was glued like a calculator would have been.” Peddle convinced Tramiel to replace the calculator keyboard with a full-sized keyboard. “He was reactive enough and we said ‘Okay, that didn’t work. We’ll put another one out.’” By late 1977, it was clear Bill Seiler was the primary hands-on hardware engineer on the PET team. “I was pretty much the main hardware guy we had, and Chuck was standing side-by-side with me,” he says. “There was immediately a turnaround of a redesign because of the way the video worked in the first PETs.” He began putting together an improved PET computer with a full-sized keyboard and a better (and larger) display. Although the video circuitry had already undergone one revision earlier in 1977 by Takamori, it was still lacking in many ways. “There were some major things wrong with the first video display on the PET,” says Seiler. If a attempted to scroll the text on the screen, the text disappeared. “We had to blank the screen in the early PETs when the guy was scrolling, otherwise the screen would flicker at you. We shipped thousands of them like that.” Seiler had previous video experience from the video card he made for his Digital Group computer, and he decided to refine the graphical circuitry to scroll the text smoothly. “The PET put out a pretty proprietary video signal that we actually designed our own displays for,” he says. “We sat down and brainstormed about it and I redesigned the display.” It was trivial to swap the new monitor and keyboard in, since they connected to the motherboard using standard connectors. The hardest part was creating a larger case for the components. The engineers would preview the prototype at the CES show in January, held in nearby Las Vegas for the first time.

Peddle made sure his employees received compensation for their contribution to the PET. “We had stock by that time,” says Seiler. “Chuck wheeled and dealed and got Jack to do that for us.” After the surprise success of the PET, it was clear Peddle had done much to reverse Commodore’s ailing revenues. He was justly walking confidently through the hallways of Commodore and people took notice. “The official inside Commodore joke is that [PET] stood for Peddle’s Ego Trip,” says Peddle. One of the oldest pieces of wisdom says, “Pride goeth before destruction, and a haughty spirit before a fall.” Chuck Peddle was about to learn the meaning of those words. At the time, Peddle was confident about the superiority of the PET over the Apple II, namely because, compared to the Commodore PET, loading programs with the Apple II was slow and unreliable. “Woz was bored with how slow his cassette loaded his program,” says Peddle. “In fact, he used to get errors because his coding for his cassette was not robust.” “Everybody is thinking we need to go to a better storage device,” recalls Peddle, referring to disk drives. An IBM engineer named Alan Shugart invented the first floppy disk drive in 1971, which used eight-inch disks. In 1976, Shugart invented the 5¼-inch floppy disk, which soon became standard. Peddle explains, “Steve [Jobs] knew that they needed to go to floppy because everybody was working with floppies at that time. So he told Woz, ‘You have got to get a floppy done by the Consumer Electronics Show.’”[6] After months of waiting for Wozniak to act, the Apple II disk drive failed to materialize. “They had committed that product for a show and Woz wasn’t doing it,” says Peddle. “I think he hadn’t figured out how to do it and he was just thinking about it, trying to make it happen.” “Finally everybody in the company puts pressure on him,” says Peddle. “Steve cajoles him, sits with him, and does everything he can.” Wozniak studied an IBM floppy disk drive to inspire his own design. Finally, just before CES, he came through. “Woz invented redundant coding for disks again—GCR,” says Peddle. “It was a brilliant piece of work. He made it work and got it ready for the show.”

When Apple representatives attended their first CES in Las Vegas in January 1978, they were able to demonstrate a crude prototype of the disk drive. Tramiel was not happy to learn his competition had developed the technology ahead of his own engineer. “At that time, Woz had gotten his little floppy drive going so they were showing that off,” says Seiler. “We were really impressed by Wozniak’s little floppy drive. We knew the Apple II was going to kill us if we didn’t get a floppy out.” Prior to the January 1978 CES, political maneuvers within Commodore conspired against Peddle. “The first problem happened in the fall of 1977,” recalls Peddle. “Jack wants to replace Chris Fish, the CFO (Chief Financial Officer) of the company, with Dick Sanford because he thought Chris wasn’t capable of doing his job.” For months, Fish had no real title within Commodore and worked at odd jobs, but Tramiel needed to assign him to another role. At the January 1978 CES show it became clear to Tramiel that Peddle was having a hard time focusing on the business side and his engineering duties. “Jack decides he wants me to focus on engineering,” explains Peddle. “In order to keep Chris off his high horse, Jack tells him Chuck’s going to be too busy and the only guy Chuck will work for is you. So I go to CES show in January, being told for the first time I’m not running the business.” Instead, Tramiel promoted Chris Fish to VP of Engineering.

Viewing the VIC demonstration at the 1978 Winter CES.

Many within Commodore agreed with the decision. “I never really saw Chuck as best suited for running a computer division. He was a technical genius,” says Kit Spencer. “I’m probably a bit like Chuck in some way. He’s great at deg computers but if you ask him to run a whole load of different engineering groups rather than focusing on particular product developments, it’s maybe not what Chuck’s best at.” Tramiel’s decision was a devastating blow to Peddle. “It’s the first time he and I are not buddies,” he says. “The personal computer business, which I set up, was supposed to be mine.” Peddle was already close friends with Chris Fish, an inner family member, and Fish convinced him to continue forward with his work.

Chuck Peddle demonstrating the PET at the 1978 Winter CES in front of a wall of PETs. Note the Starship Enterprise on each monitor.

While Peddle’s computer systems division showed off its revised PET computer, the components division, principally MOS Technology, had its own products to show. Al Charpentier brought his video chip to CES. To demonstrate the features of the chip, a programmer named Fred Bowen created a small, colorful demo that showed a ball bouncing around on the screen. They also handed out literature encouraging manufacturers to use the chip in low-cost CRT terminals, biomedical monitors, and control system displays. The two Steves from Apple were also walking the show floor checking out the new products, including the VIC chip. “Steve Jobs and Wozniak came by,” recalls Charpentier. “They were very impressed and really liked it a lot. Woz was wondering how we did the colors. Woz must have come by three different times and I wouldn’t tell him,” he laughs. Although attendees expressed interest in the chip, the biggest complaint was the character display. Charpentier designed the chip around video games, which is clearly visible in the thick, blocky characters produced by the VIC chip. The chip could only display 22 characters per line, compared to 40 characters for the PET. As things turned out, no one was interested. “We were trying to sell it into the merchant market as OEM (Original Equipment Manufacturer) and it never went anywhere,” recalls Charpentier. He feared his VIC chip was in danger of fading into oblivion. While the engineers were showing off the new PET models, John Feagans noticed Bill Gates step up to one of the demo computers. “He was looking over his shoulder,” recalls Seiler. “Gates walked up to our machine and played around with it.” What Feagans saw amazed him. Gates typed a simple command and the screen displayed the word MICROSOFT. “If you put WAIT 6502 and then a number, it

would print MICROSOFT that many times,” says Seiler. Gates cleared the screen and walked away, unaware Feagans had observed his deed.

John Feagans, Commodore’s premier systems programmer.

Feagans soon realized Gates had snuck an Easter egg into Commodore BASIC. “There was another guy who did the first work on BASIC but Gates was working on it near the end. That’s how small the company was,” says Seiler. Gates used the hidden code to fingerprint this particular version of BASIC. “He put in something in the PET because he was real suspicious of people like me stealing his BASIC on paper tape,” says Seiler. If someone attempted to incorporate BASIC into a machine using Gates’ code, it would be easy to check if the code originated with Micro-Soft. Feagans felt perturbed because he had previously scanned through the code searching for hidden messages and found no discernable words. “It wasn’t in ASCII where you could see it. Gates had fiddled with a couple of bits with the ASCII character set, so it didn’t show up when you just looked at a raw binary dump,” says Seiler. Feagans vowed to hunt down the hidden message and remove it from Commodore’s code. “He was just so perturbed that he had put that in there,” says Seiler. “Feagans is kind of a perfectionist like that.” [1] At least two other companies received the specification for the Radio Shack computer: Sphere Corporation and Microcomputer Associates. In later years, Michael D. Wise of Sphere claims Radio Shack lifted parts of his design. Ray Holt of Microcomputer Associates (who wrote the ROM software for the KIM1), also believes the prototype he offered was reverse engineered. [2] 1993 interview with the Smithsonian [3] The virtually unknown Apple II was missing from the article. [4] In 1968, a company called Viatron tried to market personal computing with a product called the Viatron Everything Terminal. Viatron offered consumers access to a shared computer system for $39 a month, and like Jack, it began taking preorders for its service. By 1970, the company had $100 million in back orders, yet it had failed to deliver a product. The company folded amid charges

of fraud. [5] ComputerLand could not sell Tandy machines because Radio Shack stores sold TRS-80 computers exclusively. [6] Most s of this story say Mike Markkula pushed Apple to develop a disk drive.

CHAPTER 10

Storming Europe 1978

After the January 1978 CES, Kit Spencer made the short trip to Silicon Valley to continue his education. “I went around in Silicon Valley with Chuck just listening to him,” he recalls. “He took me down to computer clubs and computer stores. I buying microcomputer books in a computer store before our PET ever came out.” Spencer focused intensely on gaining knowledge, since he would need to launch the PET in the UK soon. “Going around with Chuck, I learned his vision of where the industry was going. All I had to do was put them into play when it was relevant.” Spencer had an edge over many of his rivals due to his background. “As a marketing person, I was more technically oriented than most because I had done a physics degree and wanted to learn more.” The trip through Silicon Valley was sometimes overwhelming, and there was the possibility to misunderstand a lot about the new industry. Back home in the UK, Spencer had time to consolidate his thoughts. “I coming back and reading all those books about computers,” he says. He began to formulate a marketing plan. In early 1978, Spencer received a limited number of PET computers to use as demonstration and development machines. “The first Commodore PETs turned up in the UK with no manuals,” he recalls. “They were rushing to get the product out and people like Chuck never liked to sit down and write down what they were supposed to do. It took too much time!”

Chuck Peddle explains the reason for this omission. “We had a guy by the name of Gordon French who was supposed to write a better manual, but he didn’t do a good job,” he says, referring to the founder of the Homebrew Computer Club. Apple sured Commodore with their manual. “Jef Raskin wrote this absolutely great book for the Apple. That was one of the ways Apple got ahead.” The documentation included with the PET was scarcely a few photocopied pages with instructions for turning on the PET, some rudimentary BASIC commands and a game code listing. “We taught them how to write the code and bring up Lunar Lander,” says Peddle. Spencer was in an unusual situation. With no manual, there was very little he could do with the machines. It was time to hire someone who knew more than he did. Taking Peddle’s earlier advice of avoiding mainframe computer employees, he sought out hobbyists. “The first two hires were guys who had built their own 6502 microcomputers,” he says. “It was good advice from Chuck.” Spencer’s hobbyist had a formidable job to complete. “To the first guy I hired, I said, ‘I don’t want to see you. Just go ahead and get me some sort of manual. Anything you can write.’ He spent about two weeks writing a crude manual, which we then photocopied and included with every computer.” The company controller regularly sent reports back to the US, which Jack Tramiel inspected meticulously. Soon Spencer received a reprimand from Tramiel, who asked, “What’s this? Why are you spending money on a manual?” Spencer replied, “We haven’t got one, Jack, and we need one.” Tramiel replied, “That’s the USA’s job, not yours.” The next day, Spencer received a call from Commodore US asking, “Can you send a manual please? Jack says you have one.” To keep costs low, Spencer encouraged his employees to attempt to earn profits from even small endeavors. “In marketing with such things as software, magazines and programmers manuals, which are essential, it helps to make them profit centers,” he says. “If you don’t, things can get out of control.” Spencer allowed his employees to create and run their own businesses within

Commodore. “One of the best disciplines you can have, instead of having cost centers, you make profit centers, and you try and make everybody think that way,” he says. “Most of the workforce responded very well to that as it felt like they had their own business. They knew if they earned more they could do more and expand.” Spencer wondered how he could target potential customers. “The more you can market and communicate with s, the better,” he explains. “You have to put in a communication system so you can keep in with them.” However, it was not immediately clear how to do that at first. “I think the idea came to me when Leonard Tramiel was demonstrating an original PET prototype at an exhibition in the UK because none of us could work it at the time,” he explains. “At that exhibition, we saw the incredible interest, so I said, ‘We’ll print up a form to say we’re going to have a magazine to keep those interested informed ing the PET. The cost will be $20 a year. You can sign up for it now and we’ll send it to you.’” Spencer began taking subscriptions right at the show. “So we got some revenue in before we had even sold a PET!” Although the US had microcomputer magazines like Byte, the UK market was still barren. Spencer would help change that situation by introducing what would become Commodore magazine. The physical quality of the initial magazines was laughably primitive. “It was literally started off with 20 Xerox sheets and it eventually became a full magazine,” he says. For an annual subscription fee equivalent to $20, Commodore UK would send out twelve issues per year. Using one of the demo PET 2001 machines, a Commodore employee created content in anticipation of the release. “I hired a technical guy and he would write that magazine every month giving out news about the PET, tips about software, little free programs and things like that, including details about any PET computer clubs that were forming. This was the start of a full-blown magazine.” Tramiel knew demand for the PET was high and predicted that in the UK, demand would outstrip production. As a result, he raised the price for the European market from $795 to the equivalent of $1295. He could lower the price later if demand subsided. “When he first introduced the PET in the US, he

realized he set the price too low, because he got too many advanced orders,” says Michael Tomczyk. “So he took the money in advance for those orders in the US. He kept it for something like six months and then he used that money to sell computers to Europe at a higher price. And then he used the profits from that to come back to the US.” In order to move product, a manufacturer needs either retail stores or specialized dealerships. The PET was not suitable for retail stores at the time, so dealerships were the only viable solution. “I recognized at that time the PET wasn’t suitable for the consumer channels,” explains Spencer. Unfortunately, there were no distributors so Spencer helped British entrepreneurs start their own dealerships. “We recognized that at the price it was at, the it needed was quite different from calculators. We needed service and software so we made sure the dealers all had that in their operations.” Normally, entrepreneurs approached Commodore. “We helped quite a few people into business,” recalls Spencer. “Some software guy would come and say, ‘I’d love to be in this business.’ Then some guy with a sales operation, who didn’t know much technically, said, ‘I want to be a dealer.’ I would say, ‘If the two of you will get together, you can have a dealership provided you have some finance and a service operation.” * * * Commodore’s investors waited in anticipation of the PET release and eventual media coverage. It would tell them if their investment in a radical new product had been justified. “This was brand new and people didn’t understand the industry,” says Dick Sanford. “I mean, what do you do with these damn things? Look at today, you can’t live without them. But back then…” The gadgets section of Playboy magazine (February 1978) featured an entire page dedicated to the PET. It included seven PET photos, and one photo even included the Playboy bunny logo in PETSCII characters. Playboy titled their article, ‘Your Own PET Computer’. The friendly name by Andre Sousan was having the desired effect. Though reviewers universally praised the PET, they also noted the lack of a proper keyboard. It was obvious to the reviewer from Electronics Today that it would be difficult to type on the PET for sustained periods. The calculator keys

did not produce enough tactile , and s had to watch the screen carefully to input. The lackluster keyboard left an opening for competitors like Apple. Despite the keyboard, Commodore clearly had momentum over the competition. The PET boasted a built in monitor, a built in data storage device, lowercase characters and clear text. s received everything they needed right out of the box. The PET was the first totally assembled, totally integrated computer with all the necessary components for software distribution and development. Reviews for the PET and Apple II appeared in the March 1978 issue of Byte magazine. Future Personal Software founder Dan Fylstra wrote an in depth, nine-page review of the PET. Fylstra was impressed with the ease of use of the PET 2001 and gave it a favorable review. In contrast, editor Carl Helmers labeled the Apple II a “proof of concept” of the idea of a personal computer, but clearly thought the design needed to evolve further. In his ten-page review, he criticized the case, Steve Jobs only contribution to the Apple II, for its poor design. Helmers noted several mechanical problems that caused the fasteners of the top cover to break after a few uses. He also noted a problem keeping the keyboard moored to the case. Helmers also noted strong interference patterns on the television caused by radio frequencies inadvertently generated by the Apple II. However, the review hinted that high-resolution capabilities held potential, once s began purchasing more memory. With memory prices so steep in 1978, most s did not equip their Apple II with enough memory to take advantage of the high-resolution graphics. If memory prices dropped, the Apple II could become a promising video game machine. Commodore print advertising was conspicuously absent from many of the leading computer magazines of the time. MOS Technology had previously placed ments in all the leading computer magazines for the KIM-1 but following the acquisition by Commodore, advertising stopped. The lack of ments made Commodore look less substantial than Apple. The slight must have angered the publishers, who relied on advertising to remain profitable. Instead, Tramiel expected free publicity from reviews. The misstep would soon cause magazines to ignore Commodore. In the early stages,

however, Byte continued to cover Commodore products. * * * Soon after PET computers started appearing in Mr. Calculator stores, an unexpected phenomenon began to occur. Kids and teenagers instinctively drifted towards the new machines. At first, this irritated the store managers. One of these teenagers was Neil Harris, who would eventually work for Commodore. “In the early 70s I was in high school in Philadelphia,” says Harris. “The school district of Philadelphia had a computer downtown that you could to from a teletype in the high schools. And that computer was programmed in BASIC. It was an HP 2000 computer system. None of the teachers knew how to work it, but there were students who understood it, so you could go to a class and the older students would teach the younger students how to work the computer. So I became a fairly adept BASIC programmer in high school, and wrote a lot of games.” After high school, Harris pursued an English major at Cornell University but dropped out partway because he was unsure what he wanted to do with his life. “I figured I’d go out and work and figure it out later if I wanted to move into something else.” Harris programmed ing systems for a while but found it unsatisfying. “In 1978 I kind of wanted to make a change and saw that there was a Mr. Calculator store that had opened downtown selling personal computers,” he recalls. “So I answered an ad and talked to the manager of the store. I said, ‘I’m an experienced BASIC programmer, I’ve done a little bit of retail work in my younger days. This sounds interesting can I come by?’ He said, ‘Sure.’” In March 1978, Commodore was trying to get rid of some of their outdated calculator inventory. “They had run a full page ad in the newspaper with major discounts on Commodore calculators,” says Harris. “So I arrived for my interview at about two in the afternoon. The store was absolutely packed with people, and the poor store manager was the only guy working there so there was no way he could actually talk to me at that point.” Harris spotted a Commodore PET in the corner. “People were typing on it and it was coming back with syntax error messages,” he says.

Harris was in luck, because the PET 2001 version of BASIC was almost identical to the HP 2000 he had used in high school. “It spoke a dialect of BASIC that was pretty easy to figure out.” Harris entered a rudimentary two-line program that asked the to input their name, and then printed it repeatedly. The other patrons were amazed. “People were going like, ‘Wow you are some kind of magician.’” Harris stayed in the store demonstrating the computer until the store closed at six o’clock that evening, and the storeowner was finally free to give him an interview. “The next day the manager called me up and asked me to work there. So I spent the better part of that year programming and selling Commodore PET computers.” Harris spent the next few years attending groups, editing a newsletter, and doing occasional contract work. “The manual was fairly worthless at that point, so there were a lot of us in the community who were hacking away at the system trying to figure out how things worked, and sharing information through newsletters.” Newsletters soon appeared, with names like The PET Gazette, The Computerist, and The PET Paper. The newsletters focused on Commodore products, with news, hints on hardware, tutorials, and even full programs for s to type in and save. These early publications did not last long, but larger magazines later supplanted them. A proper manual appeared in 1978. “We finally wrote some decent manuals by bringing in outside people to do the writing,” says Peddle. Commodore eventually included Adam Osborne’s 430-page manual with every PET. Harris was typical of the PET community, and Tramiel was quick to recognize this phenomenon and exploit it to sell more computers. As word began to filter back to Commodore, the marketing people formed a new sales tactic: get to the parents through the kids. Now, ments read, “Until now, parents interested in expanding their children’s educational experience, turned to expensive encyclopedias. With the advent of the PET personal computer a new avenue of visual learning is now available for your youngsters for as low as $595.” The ment concluded, “Your youngster will think he’s playing, you’ll know he’s learning with the remarkable PET.”

Peddle had little doubt his computer appealed to ordinary people. “These were people wanting a machine so bad they could taste it,” he explains. * * * The story of VisiCalc starts with Peter Jennings, a genius programmer who created an intelligent chess game in one kilobyte of memory for the KIM-1. By the summer of 1978, Jennings had ported his game Microchess to the most popular machines of the day—namely the PET, TRS-80 and S-100 machines. The Apple II version would come later. The PET version was easy to port, since it used the same 6502 processor as the KIM-1. It was easier still because the PET was swimming in memory. Now, Jennings had access to four times the memory, giving him room to improve his chess-playing algorithm, as well as provide a graphical interface so the player could see the chessboard. Peddle noticed sales of software could help increase sales of hardware in a symbiotic relationship. “He was a very early developer and he wrote MicroChess, which was very important to us when we were selling the PET,” explains Peddle. “Jennings was one of my first partners. We did shows together. We had dinners together. He was a very important guy in our lives.” Jennings sold his software under the name Micro-Ware. Packaging was amateurish but standard for the time, with the cassette tape and printed instructions sealed in a plastic zip-lock bag. Jennings sold Microchess at computer shows, personal computer dealers, and through mail order. At one such computer show, PC-78 in Atlantic City, Micro-Ware went in a new direction. Jennings met with Carl Helmers, editor of Byte magazine, who wanted Jennings to meet Dan Fylstra, an MBA student at Harvard and a knowledgeable programmer of the 6800 microchip. Fylstra was also a writer for Byte and had previously written a review of Microchess. “He was a very Ivy League, business kind of a guy,” says Seiler. “He always wore a three-piece suit when I saw him.” Fylstra was just beginning a new company with the sole purpose of selling software. His plan was to sell third party software under the Personal Software banner. Personal Software was the first company to publish the legendary Zork adventure game, developed by Infocom.

Fylstra and Jennings eventually decided to merge the two companies under the name Personal Software. “[Peter] Jennings had a partner, a young guy from Harvard,” recalls Peddle, referring to Dan Fylstra. “I’ve got this memory of him at one of the shows, of him taking four machines and saying we’re going to go out and write a bunch of code for these, which they did.” Peddle helped Fylstra any way he could because he saw Personal Software as symbiotic with the PET computer. “[Fylstra] is hooked on Commodore,” says Peddle. “Jennings is hooked. We’re personal friends and we’re helping them sell product in Europe. It’s a really good relationship.” Personal Software owned at least four PET computers for software development but decided to try the Apple II market. “They had bought one Apple, but they decided they would back Commodore because we had other things going for us,” says Peddle.[1] Personal Software eventually became interested in distributing a software program from a fellow Harvard student, Dan Bricklin. Bricklin had an idea for an electronic calculating ledger that would allow real-time manipulation of numbers in rows and columns. “One day, a couple of guys from Harvard wander in and say, ‘We’ve got this idea for something we want to write,’” recalls Peddle. What happened next would change the face of computing. “[Fylstra] says, ‘Well, the PETs are all tied up. Why don’t you take that Apple there and see what you can do with it. It runs the same BASIC.’” The PET computer was the leading business oriented personal computer, with s perceiving the Apple II as an expensive game machine. Despite this, the Apple II would receive VisiCalc, a business oriented application. “It was just absolute blind-assed luck,” says Peddle. “These guys had the idea for the architecture and they didn’t care about the machine.” “It is true that Dan Bricklin prototyped the first spreadsheet on an Apple II because it was the available machine at that moment,” says Jennings. Fylstra and Jennings immediately liked the idea and agreed to pay for development costs of $4,000 to $5,000 per month, spent mostly for access to a timesharing system. Jennings agreed to defer his royalties on Microchess for the development of VisiCalc. In effect, profits from TRS-80 and PET software were about to fund an Apple II application.

Fylstra and Jennings brought a demonstration of VisiCalc to Commodore hoping to create a mutually beneficial business deal. “They showed it to us in the office at Commodore Scott Boulevard,” says Seiler. “They brought it in, and it was on an Apple II. The guy was showing you could put numbers in boxes and you could make this box equal that box plus that box times this box. We said, ‘Wow, that’s really cool. I don’t know what we would do with it, though.’ We didn’t think of the ing angle on it. They went off with it and Commodore didn’t do anything with it.” “It was one of the most creative software packages ever written and the only true piece of software that went on PC’s without ever having a parallel on mainframe computers,” says Peddle. “Word processing, we were doing on timesharing systems. Everything you did on a [personal] computer, we were doing on big computers, except for VisiCalc.” When Personal Software released VisiCalc on October 17, 1979, it became an instant hit. “The stores said ants would come in and say, ‘I need VisiCalc.’ The salesman said, ‘Well, do you have a computer?’ They said, ‘No.’ ‘Well, you’ll need to buy a computer to run it on.’ And they sold them an Apple II with a floppy. That really pushed them way up in sales, and that was a big success for them,” recalls Seiler. Sales of VisiCalc directly drove the sales of Apple II computers, even more so than the sales figures indicated due to software pirating of VisiCalc. “VisiCalc saved Apple,” says Peddle. “People were buying Apples and taking them into work to do VisiCalc work. The Apple II became the definitive VisiCalc machine.” In the first month alone, VisiCalc sold over one thousand copies at $100. According to Seiler, “ants saw it and said, ‘God, I need one of those.’” Soon, Personal Software raised the price to $150 but sales only increased. Personal Software became a software giant, with Microsoft a distant second place. * * * Games were rapidly becoming the killer-app for early microcomputers, and Microchess was one of the biggest sellers. Over the next ten years, Peter Jennings sold several million copies of Microchess to owners of home

computers. With the help of the KIM-1, Jennings helped pioneer the computer games industry. In 1978, Commodore realized the value of Microchess. At the time, electronic games were becoming popular in households. Several companies released handheld games such as Simon (Milton Bradley), Merlin (Parker Brothers), and later Speak & Spell (Texas Instruments). Each of these electronic games would go on to huge success, especially when E.T. the Extra Terrestrial featured Speak & Spell in 1982. Commodore wanted a piece of the electronic games market, so they ed Jennings about creating their own handheld game. “We did a thing called CHESSmate based around Jennings work,” says Peddle. Internally, CHESSmate was very similar to an ordinary Kim-1 computer, with a sibling of the 6502 microprocessor, the 6504, substituted. Commodore hired Jolt maker Microcomputer Associates to construct the CHESSmate prototype. “Of course, it didn’t work very well,” adds Peddle. At the time the CHESSmate was developed there was only one other electronic chess game, the Fidelity Chess Challenger. The contract with Commodore called for Jennings’ game to beat the Chess Challenger in tournament play. “One of the funniest moments was my lawyer going over the contract,” recalls Jennings. “He read the clause which required that my program beat the Chess Challenger and pondered it over for a while. He just couldn’t wrap his mind around the concept of the two machines playing against each other and one winning. ‘What do you mean?’ he kept asking. It’s hard to put yourself back to the seventies when most people had no with computers and had no concept of machines that played games against people or other computers.”

Developing an early prototype of CHESSmate.

Commodore engineers added more ROM memory to the game, and Jennings improved his Microchess code by adding 32 opening moves. He even provided eight different play levels to the game so novices could enjoy the game. CHESSmate was able to hold up Jennings’ end of the contract by consistently beating the rival Chess Challenger. In 1978, Jennings traveled to Pasadena where Bobby Fischer was in hiding from the world. He spent several days demonstrating the prototype CHESSmate software running on an expanded KIM-1. Fischer played many games and handily beat the program each time, but he had a strange fascination with the alien strategy devised by the machine. Fischer even challenged Jennings to a game, an offer that caught Jennings off guard. “It was a surreal experience being challenged by the reigning world champion to a friendly game at a time when very few people even knew where Bobby was and nobody had seen or played him in years,” says Jennings. Though Fischer predictably won, Jennings believes he gave Fischer a respectable challenge.

The production version of CHESSmate with manual and chessboard coordinate stickers.

A consummate businessman, Jennings offered a royalty to Fischer in return for calling the electronic game the “Bobby.” At the time, the strongest commercial game available was the “Boris” named after Boris Sky, who Bobby Fischer had defeated in Iceland for the title of World Champion. Even though the CHESSmate would play competitively against the Boris computer, it could not defeat it every time. Fischer declined and “Bobby” became CHESSmate. * * * Commodore UK relied on PET computers assembled in the US, , and Japan. In April 1978, the first machines began arriving on British soil for distribution. Over time, Commodore UK developed a dealership network. “We did that for the first couple of years of the Commodore PET,” recalls Spencer. Spencer understood the software aspect of computers. “I immediately set up a software department to start developing software and encourage software developers to do things.” If Spencer thought a software product showed potential, he occasionally acquired it for Commodore. “We bought several products from some of these people to market ourselves,” he recalls. “We also helped the independent developers by producing a catalogue of any software and equipment related to the PET microcomputer. We would distribute this catalogue for them to PET s, dealers, journalists and anyone who could help our sales and development.” The Commodore PET was an immediate success in the UK. Britons loved the computer, among them science fiction author Douglas Adams who later wrote, “I the first time I ever saw a personal computer. It was at Lasky’s, on the Tottenham Court Road, and it was called a Commodore PET. It was quite a large pyramid shape, with a screen at the top about the size of a chocolate bar. I prowled around it for a while, fascinated.”[2] Spencer’s enthusiasm for the work drove him on. He did everything, including

the creation of presentations, advertising slogans, press releases, interviews, and newsletters. He did it all and he did it fast, resulting in a steady increase in market share. He even created a training division. “We had a guy who gave courses on how to operate PETs and program in BASIC. We charged to come to our courses and learn about the computer,” says Spencer. “He ended up with four or five staff running training courses, and he covered all his costs with a small profit. If he made money we plowed it back in and it helped the product. A lot of people in US saw that and didn’t realize we weren’t spending money, we were making money.” There was a sense of youthful fun in the UK microcomputer division. The new employees derived a large degree of satisfaction from what they did because they were hobbyists. The young staff loved to come into work each day, since Commodore paid them for doing something they would be happy to do anyway. “The biggest problem was keeping them and my financial controller apart,” comments Spencer. “They came from different planets but both were needed in our business.” Commodore even hosted the aptly named PET Show in Piccadilly during the first year of release. “The goal was to invite the people supplying any sorts of peripherals and software,” explains Spencer. “We took a load of stands and staged an exhibition. It was to promote interaction between software developers and marketers and dealers and everybody. And it actually even made a little money too.” Because of the aggressive marketing campaign and well-oiled dealer relationships, Commodore was able to capture close to 80% of the microcomputer market in Britain. Spencer actually sold more computers in the United Kingdom for a higher price than Commodore could in the much larger US market. also managed to capture 80% of their microcomputer market using similar tactics. Forbes magazine pinpointed the key to success in Europe. “Commodore did better in Europe, where it had a ready-made distribution system from its adding machine days. Priced at $1,295 in Europe—vs. the $795 price in the U.S.—the PET had sufficient margins to be ed aggressively.”[3] Spencer sold over 30,000 machines in the first year. Since Peddle designed the PET 2001 to sell profitably at $795, the margins were astronomical.

Kit Spencer (left) and Bob Gleadow (right) arriving at the Commodore UK headquarters.

Curiously, Spencer’s product launch contained no advertising. “Contrary to what a lot of people thought, I did virtually no advertising in the first year of the PET in the UK,” he reveals. “The Americans used to think we did well in the UK because we d. That wasn’t totally true. I had a pretty aggressive marketing campaign but it wasn’t through big spending in advertising. In the first year there was no straight media advertising because quite frankly we couldn’t afford it.” Spencer’s secret to developing a market without a large budget was to make each division profitable, whether it was a computer show, the software division, customer training, or the magazine. “I used to say I could do almost anything as long as it was cash positive and profitable,” says Spencer. “We set up the computer division as almost an experiment in the UK. What I tried to do was put in those service operations but try not to make them a big cost. If you buy a car, you don’t get free driving lessons. If you sell a low cost computer, you can’t afford to spend a lot of money teaching people how to operate them for free too. You need to give them decent manuals and offer such things as training courses. But at the price you’re selling the PET at, you can’t expect them to all be included in the price.” Spencer had remarkable success in the business community with the PET. “From the very early days of the Commodore PET, people used to say, ‘IBM gives you all this service. What do you think about service warranties?’ I’d say, “Well you can buy 10 of our computers for the price of one of those. They charge a lot for a service contract. For their price you can have ten PET backups which I think is a pretty good service contract!’” * * * Although the immediate success in Europe benefited Commodore, it actually hurt Commodore US, since Tramiel decided to channel resources away from the US market. “In North America, Jack wasn’t pushing as hard, because we had limited ability to make product and we sold them for more in Europe,” says Peddle. “If you’ve only got limited resources, which we always had, you focus

on where you make the most money.” Even in the midst of success, Irving Gould was not willing to raise money on the stock market to reinvest in Commodore. “Commodore went public, the stock really just kept running, but Commodore never picked up any of that money,” says Peddle. Instead, Gould forced Tramiel rely on profits and occasional loans to run the company. According to Spencer, “I don’t think either of them wanted to dilute their shares very much. On the other hand, they could see the need for money. Could they do it other ways? That’s something I don’t think anybody really knows.” “In the US, it never had much good marketing from day one,” observes Spencer. “We were lucky in the UK. Jack left me incredibly alone. I think distance always helps. And I’d worked with him for a number of years so I think we’d already learned to understand each other. Provided you handled it right, Jack let you do things.” One of the biggest US successes turned out to be the stock analyst from New York. “He became one of our biggest dealers in the United States,” says Peddle, who noticed some extraordinary sales figures of this one dealership. He decided to investigate the secret of his success. “I just wanted to drop in on him in New York. He’s selling them in a basement right off Wall Street.” To his surprise, the first thing he saw in the basement was his competition. “They’ve got Apples stacked everywhere. People are coming in and buying Apples.” “I said, ‘You’re moving a lot of Commodore’s. Where are they?’ He said, ‘Well, I keep them out back,’ or something like that. It turns out he wasn’t selling any Commodores in the United States. He was getting them at a cheaper price than they could in Europe and he was bootlegging them into Europe.” As Peddle later found out, the dealer began selling PET computers exclusively in the United States but soon discovered a grey market in Europe. “He worked that scam out later,” says Peddle. The dealer began ordering extra PET computers just for the European market, and then stopped selling PET computers in the United States altogether. The situation hurt Kit Spencer, since dealers in the United Kingdom began ordering computers from American dealers rather than Commodore UK. The grey market was a natural byproduct of the dual pricing scheme.

Tramiel looked into expanding PET production in the US by purchasing an existing production facility. In early 1978, Tramiel spotted a CB radio manufacturer named EF Johnson, which had lost $19 million for the year. Cheaper CB radios from Japan had begun arriving and threatened to put the company out of business. Tramiel knew the underutilized factory would be desperate for work to keep the factory going. “It kind of dried up,” says Seiler. “They had a nice big plant and not much to do, and they were real eager to get into the new budding personal computer business and learn how to build computers.” Tramiel sent Bill Seiler to help transition their factory to PET production. “We decided to build PETs at EF Johnson,” says Seiler. “It was one of my crazy spinoff jobs they sent me off to do. I flew out there to help them bring the plant up.” Seiler re the perilous journey to Mason City, 2000 miles west of Commodore’s Santa Clara headquarters. “I used to have to fly into Des Moines,” he recalls. “Then I would either rent a car and drive all the way to Mason City, which was a couple of hours, or I could get this little puddle-jumper to fly me up there. I used to have to find the pilot in the bar. He was half-drunk when he’d take me up there. It was almost the same price as renting a car to get him to fly me up there.” To save time, Seiler usually went with the propeller plane. Seiler noticed an immediate difference between the factory back in Santa Clara and the one in Iowa. “At Commodore in California, most of the people were Filipino or Mexican,” he explains. “I’d always have to find a translator to talk to people on the line when there was a problem. Usually the head person spoke English. But in Iowa, everybody spoke English. It was amazing.” Tramiel was studying EF Johnson to see how cheaply it could produce the computer. “I don’t know if Jack was considering buying them or just hiring them as some kind of contract manufacturer,” says Seiler. “It was a really nice big factory and they were good production people. We started to make PETs there but I don’t know how much we built.” Tramiel was testing the company to see if they would be open to a below-value acquisition should they ever run into financial problems. [1] Peter Jennings says it was just salesmanship. “Of course we told Peddle that,” he says. “We told Steve Jobs that we ed Apple above all the others

and we told Tandy that Radio Shack was our favorite.” [2] Douglas Adams, Salmon of Doubt (Harmony, 2002) p. 91. [3]Forbes magazine, “Albatross” (January 17, 1983) p. 48.

CHAPTER 11

Race for the Disk Drive 1978

During development, the PET became a friendly, low cost personal computer for the home. “We started off making a $595 computer for the general masses,” says Chuck Peddle. After the response in Europe, the potential to reposition the computer for business use became clear. Peddle was determined to transform the PET into the serious business machine he had originally envisaged. Already the engineers had added a real keyboard and larger monitor. To make the machine ready for business tasks, it would need a printer and disk drive, which would connect to the IEEE-488 interface. Perhaps they could add more memory to allow it to run sophisticated software. The task of redeg the PET went to Bill Seiler. On top of the new keyboard and monitor, he would redesign the motherboard to make it more cost effective to manufacture. Along the way, he would add the ability to expand the computer with up to 32-kilobytes of RAM. Seiler, who previously created sound effects for Allied Leisure pinball games, would also add a simple speaker to the new PET design, giving it the same sound capabilities as an off-the-shelf Apple II. In the spring of 1978, Peddle began work on his disk drive. “I hired this team to come in and build a dual floppy disk controller that works on our IEEE bus.” As with his previous hires, Peddle turned to Roger Camp at Iowa State University. “Every year, Dr. Roger Camp fed me the brightest and best out of his class, who were weird,” says Peddle. “They were guys who were not going to fit in other places but were really sharp.”

“We immediately ran off to Ames, Iowa and got some more engineers and started them working on the dual floppy drive,” says Seiler. “They were Glen Stark and Scott Patterson. Those guys were working on the floppy.” Peddle hired Stark to build the electronics for the dual disk drive. According to Peddle, Stark did not start out at Commodore as a “weird” engineer. “Stark came west with his girlfriend. He was a much more of an open guy than Seiler or Feagans,” recalls Peddle. “Then his girlfriend left him and broke his heart. For about two years, he also became a recluse.” Luckily, Peddle preferred recluses on his team. “I had these three loners,” says Peddle. “The nice thing about having loners who are really smart working for you is they bust their ass working 24 hours a day.” The new hire impressed Peddle. “Glenn was brilliant when he got going in the right direction,” says Peddle. “You give him an assignment and he would get it done.” Peddle hired another Iowa State graduate named Scott Paterson to program the disk drive. “Scott Paterson was hired around the same time [as Glen],” recalls Peddle. Patterson was a fitting addition to the team due to his perfectionist attitude. Besides Peddle, Scott Paterson was the only family man on the team. “Scott has got a wife and kids,” says Peddle. “She was always important to him. They are the image of an Iowa couple—both really nice people, hardworking, industrious —all the things that made the Midwest work.” Paterson and Stark formed a close relationship while working on the disk drive project. “They were a team for years,” says Peddle. “It was kind of funny having this one happily married guy in this team of loners. Scott was exactly the opposite from Glen.”

Glen Stark working in the engineering labs.

With his team assembled, Peddle and his engineers began working on the dual floppy disk drive. He sketched out the basic design for the drive on paper. “Now I’m too smart,” Peddle states sarcastically. “I’ve done a bigger version of the PET and put a keyboard in it, and we’re ready to go into business computing. Now I know that if you want to do real business computing on a computer, you need two floppy disks.” It was a belief Peddle carried from his days at General Electric. “Chuck really believed you needed two floppies,” says Seiler. “That’s what he thought was wrong with Wozniak’s and Jobs’ single floppy for the Apple II. He had come from the old school where you’d work from one device and write to the new device, kind of like what they do on the big reel-to-reel tapes on the big mainframes. He had come from that back at General Electric computers.” Peddle believed that microcomputer software would have the same requirements as business software on mainframe computers. “He felt that you’d put a floppy in with your old records on it, you’d do your ing on it, and you’d write to a new floppy on the other drive,” says Seiler. “So you needed two drives.” Peddle planned to house the dual disk drives in a single unit. As a result, the design was bound to be costly, bulky, and over engineered. “We built the very first floppy with dual drives,” says Seiler. Ironically, the design of the IEEE-488 interface should have allowed multiple disk drives to connect to the PET. It would be simpler to sell individual disk drives and allow s to choose how many they wanted to purchase. “The bus wasn’t real fast because we were implementing the IEEE-488 interface in software,” says Seiler. “It was fast because it was an 8-bit parallel bus, but if you had to copy a whole floppy, it would be slow. It would have worked but he felt that you could copy one floppy to the other very fast if it happens locally instead of having to go over the bus. That was an advantage there.” * * * While John Feagans was adapting the ROM code for the new revisions of the

PET, he also fixed the bugs from the original PET 2001. One particular bug was difficult for the engineers to isolate. “Every so often the machine would go off into Never-Never Land because it would be doing a computation,” explains Peddle. Leonard Tramiel, who used a PET computer for his studies, finally solved the mystery. Peddle recalls, “Leonard found it by virtue of being bored while he was living at Columbia University.” Leonard diligently tried to reproduce the reported bug and finally cornered it. He explains, “I diagnosed it down to a sample program that [reproduced the bug.]” Surprisingly, the bug originated in code written by Bill Gates himself. “Bill wasn’t that familiar with the 6502 architecture,” says Peddle. Micro-Soft corrected the bug and Commodore released a new version of the ROM code, called Upgrade BASIC (later renamed BASIC 2.0). Feagans was still determined to find the Easter egg snuck in by Bill Gates. “It drove Feagans crazy for a while because we couldn’t find the word Microsoft anywhere in the code,” says Seiler. “He had hidden it.” After a fruitless search for the code, Feagans resorted to running the code one line at a time and monitoring it carefully. “Finally he single-stepped through the code and found out where he had hidden it,” says Seiler. Gates’ Easter egg used a minimal amount of memory by piggybacking on the code to create random numbers. “The random number generator needs a seed to start up with,” explains Seiler. “And he put it in the random number seed for the random number generator in BASIC. It’s as long as the word Microsoft, but it had a couple of bits twiddled in it.” Although the code was impressive, it technically interfered with the proper operation of the WAIT command. “Feagans found it, and he said, ‘This is a bug and I’m going to take it out.’ Because you couldn’t see location 6502,” says Seiler. “I don’t know if there is even a 6502 location in the machine that would be worth looking at.” At the next computer show, Bill Gates would discover he no longer had his fingerprint on Commodore BASIC.

* * * Peddle planned to maximize the use of the IEEE-488 interface. “I was trying to build a machine that could be used for business and I also knew people who needed word processing,” explains Peddle. “I wanted a cheap printer.” If Peddle could dominate the word processing market with the PET, he could take a dramatic lead over the competition. “Apple wasn’t playing in the printer market,” recalls Peddle. Since the PET 2001 had lower case letters (something the other two computers in the 1977 Trinity lacked) it would give the PET a clear advantage as a business computer. Peddle approached Centronics, a North American printer manufacturer. “There’s a Centronics port on your machine,” explains Peddle, referring to the parallel port which existed on IBM PC computers. “The printer port is called a Centronics port. It was defined by this company called Centronics, who was at that time the leading minicomputer printer supplier.” “I went to the guy who was in charge of the company. I sitting with him in New York and talking about the fact that personal computers are going to sell millions and they were not going to sell millions at the price he was charging for printers. If he wanted to make things that sold in the millions, he had to reduce his costs.” Peddle offered Centronics a deal. “We were willing to sit down with him and design a new printer, along with him, using our [6502] microprocessor,” explains Peddle. Microprocessor technology, especially using a 6502 chip, would drastically reduce the price of Centronics printers. “At that time, he may not have even had a microprocessor; he might have had discrete control. We could go in and make some serious cost reductions for him and get him a lot of volume.” The response was not what Peddle hoped. “He basically looked me in the eye and said, ‘I don’t need to do that. I’ve got the world, I’ve defined a standard, and I know what I’m doing. I don’t need to be any more competitive than I am now. Everybody will come to me and pay my price.’” Centronics missed a deal worth millions. “I specifically said to him, ‘What you just did was to force me to ruin you. I’m going to go find a supplier that wants to be a player in the millions market. He’s not one of the guys you are playing

against today but he’s going to eat you. Are you sure you don’t want to reconsider?’” Peddle’s dramatic prophesy had no effect. “He was clear. ‘I don’t have to, I don’t want to, and I’m not going to.’ It was an American company, an American product, and all the flag waving you want,” he says. Peddle was incensed that a company would turn down a lucrative offer. “I’m pretty ticked off because we really need a low cost printer if we want to make the market really take off,” he says. Peddle had no option but to align with the Japanese and teach Centronics a lesson in free market competition. Jack Tramiel called on his Japanese s to find an appropriate manufacturer. Commodore had a long history with Japanese partnerships. “We were using Japanese [cassette recorders] and we were using Japanese monitors so Japan was a real important part of our program,” explains Peddle. Yash Terakura escorted Peddle during his stay in Japan. “Chuck and I used to work on many things: printers, chips, etcetera,” says Terakura. “Mostly I helped him out when he was in Japan. I took him to many Japanese companies to cut deals for production and procurement.” While there, Peddle tried to find a Japanese company to help manufacture the video portion of the PET. “We visited Konami for possible subassembly of Commodore computers,” says Terakura. “They were a very small upright game machine maker in Osaka.” The two finally arrived at a company willing to assemble printers. “Jack knew the people at Epson-Seiko pretty well because they had been buying calculator stuff from them,” says Peddle. The two Japanese engineers in California once again proved their worth. “[Mr. Fujiyama] set up the meeting and we met the guy who designed the LCD watch —a really, really wonderful guy from Japan,” recalls Peddle. The two talked, and Peddle proposed the Japanese would supply the printer head mechanism while Peddle would design the electronics. “He basically ed me on the program in of getting a mechanism,” says Peddle. “They had a 20 or 30 column adding machine printer.” Twenty columns was fine for calculator output on rolls of paper, but Peddle needed

something to print to standard paper up to 8 ½ inches wide. “He agreed to make an 80-column carriage.” Peddle made sure his business dealings with Epson-Seiko would not create additional competition for Commodore. He requested a promise from Epson to stay out of the personal computer business. “They told me they weren’t going to go into the computer business,” explains Peddle. The Japanese were enthusiastic to strike a deal compared to the reluctant Centronics. “So I sat down and designed the electronics using my microprocessor,” says Peddle “I handed him the design and an order for a large number of the printers.” Peddle designed the electronics for the printer, using MOS Technology components. “The processor was a 6504, which had a 16 kilobyte address space,” says Feagans. “A 6520 PIA provided all the I/O for the printer, and timers for controlling the dot matrix process. The interface to the PET 2001 was IEEE-488.” The printer was similar in design to a regular computer, and required its own software on a ROM chip. “I wrote all the embedded software to control printers at Commodore,” says Feagans. Commodore allowed Epson freedom of access in Japan. “During this time they had been working with us, they had been going into our factories,” recalls Peddle. “They had understood what we were doing and the way we made the computers. We were meeting with these guys and they were charming, smart, and really ive.” It was too good. Months later, Peddle returned. “At this point, we want to start talking about how we are going to do some volume business,” he says. “It was an early product and there were some issues we wanted to deal with, so we scheduled a trip to the factory which was out in the mountains outside of Tokyo.” Peddle arrived in Tokyo to visit the factory. “You go out on this little train,” recalls Peddle. “We’re staying at this Japanese inn where you go in and your bath is a dip bath; there’s no shower. We’re eating Japanese food, and they’ve got the karaoke bars.” “Two things happened that day. One, in the railroad station I had bought a

handheld computer made by Seiko, after they told me they weren’t going to go into the computer business.” The discovery surprised Peddle, who believed Epson-Seiko had promised not to compete with Commodore for the personal computer. “So here they have a handheld computer running BASIC. It’s not a direct competitor, but you know, it’s kind of a pretty good indicator they were thinking about it. I said, ‘I want to sit down and discuss this.’” “I’m ready to go in and visit the factory,” continues Peddle. “Well, they wouldn’t let me into the factory building my printer.” Peddle found the refusal humiliating. “They said, ‘There’s company rules against anybody coming in and visiting our factory.’ I was very upset. I’m dealing with this guy who is one of the world’s biggest patent holders. He’s a very straight guy, but he just basically said, ‘I talked to my Japanese people and they had made the decision that the Americans weren’t going to steal from them, they were going to steal from the Americans.’” Although Peddle’s treatment in Japan left a bitter taste, Epson delivered exactly as promised. “The people at Epson did in fact build a low cost printer and did help us to develop the market,” says Peddle. “We shipped a lot of Epson printers.” Commodore used an OEM arrangement with Epson. This meant Commodore purchased the printers from Epson and branded the printers with the Commodore insignia. The first printer was the CBM 2020 printer, released in 1978, which was almost as tall and as heavy as the PET computer itself. Epson decided to expand their market for computers beyond the IEEE-488 interface. “They did a version with the Centronics interface,” recalls Peddle. The cheaper Epson printers dominated the costly Centronics printers. “We absolutely put Centronics out of business and we put Epson in the business because the guy at Centronics wouldn’t cooperate.”[1] Epson also released a personal computer system on the international market in 1982. “They went into the business of competing with us,” says Peddle. Were their actions a little duplicitous? “What do you mean a little duplicitous?” laughs Peddle. “Absolutely! That’s not a new story.” In the end, the Epson deal benefited Commodore more than it hurt them. “It was one of the keys that made Commodore work very well,” concludes Peddle.

“Commodore created the whole market for printers with personal computers. If you look at the Apple II, the printing and the business stuff was not as important.” With the printer, increased memory, speaker, and professional keyboard, Peddle felt his computer could compete for the business market better than Apple or Radio Shack. “We had better peripherals, we had a bigger machine, and it was stronger,” he says. * * * Jack Tramiel intended his components division to act as profit center, delivering semiconductor parts to the industry at large. However, with the meltdown and departure of the 6502 team, he had few people with the ability to advance the 6502 architecture. “We wanted Bill [Mensch] back in the fold,” says Peddle. “He talked to me about it and I told him about Mensch and what he could do.” Meanwhile, Bill Mensch was growing disenchanted with his current job at ICE. “In about January of 1978, I’m saying to myself, ‘I don’t want to be here,’” recalls Mensch. In the previous year, he had done consulting for an aircraft company in Kansas City and taught IC design to Kodak. However, none of it was as exciting as deg microprocessors. “I’m thinking, ‘Okay, how am I going to get out of this?’” Mensch went through a bout of depression and relied on visits to the local card shop to help him get through the day. “I read greeting cards to feel good about life and recharge my batteries and then I went back to work.” Luckily, Peddle wanted Mensch back at work for Commodore—unfortunately, it was to work on calculator chips, rather than evolving the 6502. “Within days Chuck called me and says, ‘Jack wants you to come back to work deg calculator chips.’ I go, ‘Really? Wow, I’ll be deg again. That sounds fun. Let me think about it.’” Weeks later, Peddle put on the pressure and said, “He wants to meet with you.” Mensch was still hesitant, having just quit Commodore the year before. He replied, “Okay, but let me do some thinking. I can’t just drop everything and fly to Las Vegas.”

While considering his options, Mensch spoke with his boss Glen Madland, the founder and CEO of ICE. Madland had done some calculator work for Commodore in the past. When Mensch told him he was considering working for Tramiel, Madland replied, “I will never do anything for Jack Tramiel. He owes me money, he ripped me off, and I’m never going to do anything for him again.” Mensch called Peddle and told him of the exchange. Peddle told him that Tramiel believed he had been the one who was ripped off in the deal. Figuring there might be a little bit of truth in both stories, Mensch told Peddle, “I’ll meet with Jack if that’s on the table.” On Saturday, May 13, 1978, Mensch flew into Las Vegas and met Tramiel at the Dunes Hotel. “He was at the blackjack table when Chuck and I arrived,” recalls Mensch. “I watched Chuck gamble a few chips and Jack gamble $500 hands and a couple of $1000 hands.” Tramiel was not having much luck with the cards when the two arrived. “He was down by between ten and twenty thousand dollars in the brief amount of time I was there,” says Mensch. However, his luck changed and soon Tramiel climbed back to zero, and kept climbing. Eventually, he collected his chips and told Mensch, “It looks like you brought me luck so this would be a good time to talk.” “When we left the table he was up by $17,000,” says Mensch. Tramiel had one of the executive suites in the hotel. “His room was complimentary because he gambled a lot,” says Mensch. The three made themselves comfortable before getting down to business. “I’m sitting on the sofa, and Chuck is sitting to my right, and Jack is across from me.” Peddle felt that his star was riding high at the time, due to the success of the PET, and he was justifiably confident. Tramiel asked Mensch, “So how are we going to do this?” Mensch suggested that Tramiel start. Tramiel told him, “Here’s what I’m prepared to offer you. You can manage a division within my company. I’ll offer you the position of Vice President if you work for me, and I’ll pay you more than I pay Chuck.”

The dig at Peddle took him off guard. “He didn’t look happy with that comment from Jack as I ,” recalls Mensch. Mensch replied, “Jack, I thought about this for a long time, and I want to have my own company.” Tramiel was intrigued and asked, “So how would that work?” “The way to do it is you write a contract that we can work with and put in the milestones and schedule for each project. Before the project is started, we’ll have an outline of what needs to be done. Just pay me milestone payments. And when the design is completed it will be basically break even. I’ll have to hire mask designers, probably an office manager, and a secretary.” For equipment, Mensch explained that he would rent a Prime Computer time-sharing system for $150 per month. To perform circuit design and logic design, he would use the software he used at ICE, which his boss Glen Madland gave him. He would rent a small office in the back of a local shopping center. Mensch assured Tramiel, “The budget will be low. I’ll keep the costs under tight control. After the design is finished and you introduce it into production, I’ll get a royalty off it and that will be my profit.” Tramiel liked the low development costs of his proposal, but he asked for assurances that he would get his chip designs. Mensch emphatically told him, “You know the only way for me to make a profit is if you’re successful first.” Tramiel was always ready to gamble on a new idea. He paused and said, “Well alright, that’s the deal.” Mensch helpfully asked, “So we’ll sign the contract?” Tramiel shook his head and offered his hand instead. “He never did sign any contract so that if he ever wants to bail on it, he just bails on it,” says Mensch. “The deal with Bill was we would set him up in a thing called the Western Design Center,” explains Peddle. “He would get rights to use the [6502] technology. In return for that, he would develop most of the semiconductors.” It was a remarkable deal for Mensch, who now had his own technology lab and a degree of autonomy.

Mensch knew he was potentially walking into a bad situation, given the lack of a written contract and Tramiel’s reputation. “With Jack Tramiel, who doesn’t sign s, I never really expected him to pay up,” he reveals. “I had my eyes completely open. I never counted on it because I was just happy being a design engineer again.” Later in the month, Mensch worked the last day as an employee for someone else. “My last day of work at Integrated Circuit Engineering was Friday May 26, 1978, which I also refer to my first day of freedom,” he says. “My first full day of work was Saturday May 27, 1978 with my desk set up in the master bedroom of our small three bedroom house.” For his first assignment, a Commodore executive named Bud Fry wanted Mensch to reverse engineer a 5-function calculator chip from Toshiba called the LC5K3. According to Mensch, banks often gave out the small calculators for opening new s. Tramiel originally wanted to buy the chip from Toshiba, but the company doubled the cost of their chips when selling to Americans. “He called it the Japanese price, which was one half the US price,” says Mensch. Mensch was delighted with the arrangement, which gave him a degree of freedom and financial security. “I needed to have my payroll covered twice a month, so every two weeks I would send MOS Technology an invoice,” explains Mensch. “I’d round it off to nearest hundred.” Mensch estimated he could have the design ready for Commodore by the end of the year. * * * While Peddle’s team struggled with the disk drives and Mensch’s team designed calculator chips, a palpable mood overtook Jack Tramiel. “Jack changed his character during that time,” recalls Peddle. “We were all having meetings and talking about business computers and everything else, and Jack wasn’t very interested.” According to Peddle, the change resulted from a traumatic experience for the Tramiel family. In mid-1978, cancer returned to Tramiel’s daughter in-law and put the Tramiels through more anguish. “All of a sudden, she reverses and dies painfully over

about a six-month period,” concludes Peddle. “Their personal life was just shit during that time. It was like God had given them a present and then took it away.” The loss of his daughter-in-law changed Tramiel. “Jack came back from that whole experience just never the same person. It really soured him on a whole lot of life. There was a piece of him that died with her, and it was a nice piece.” Tramiel seemed to through the normal stages of grieving. “For a long time he was just totally depressed basically,” says Peddle. “You didn’t walk into his office and find him weeping, but he was just an angry person. Everything made him angry.” Tramiel’s inability to out of the anger stage would make life a lot more difficult for his employees. “Jack never really recovered from that in my opinion,” says Peddle, who noticed Tramiel’s despair changing into something worse. “Maybe five years later or ten years later, but during that time he was much more ruthless.” * * * Peddle had promised Tramiel a disk drive by the end of the year, but as the year progressed it became apparent the release date might slip. Peddle believes the inexperience of his new hires slowed down the disk drive project. “The problem I had was I just barely had hired Scott Patterson and Glen Stark and I didn’t have any horses. They came out of school and they were supposed to do this job. They got tangled up on the hardware and I didn’t have time to go fix it.” The other factor slowing Peddle down was his impending surgery. Both Peddle and Seiler had become more serious about their running. “Chuck was a fairly slow runner but he could run a long time,” says Seiler. “He and I ran quite a few marathons.”

Bob Skyles (left) and Bill Seiler working on prototypes.

The two also lured in Commodore engineer Bob Skyles, who ran the manufacturing plant on Reed Street, into their runs. “Chuck and I would run all the time together with Bob,” says Seiler. “Bob was a skinny old guy, bald headed.” Although they ran together, Skyles and Peddle often had disagreements in the lab. “He ran engineering for a while but he had trouble getting along with Chuck,” says Seiler. “Chuck would just kind of wave his arms and say, ‘We can get it done.’ Bob would say, ‘No, you’ve got to solve the problems.’ Chuck would say, ‘Oh, we’ll get it done.’ And they would fight over how they would go about it.” In the middle of Peddle’s technology push, he found out he would require major surgery on his gallbladder. “I took up running seriously because of the [impending] operation,” recalls Peddle. “I started running marathons and lost 60 pounds. I thought I was cured because my body totally changed.” According to Peddle, when he returned to his doctor, he told him, “You don’t understand. You still have to have your gallbladder out.” Peddle had his operation. “They took six great big stones out of my gallbladder,” he recalls. “He said any one of them could have caused me just nothing but trouble.” During recovery, Peddle received visits from family and friends in the hospital. “I go into the hospital and Leonard came by,” he recalls. Though he was back in University, Leonard Tramiel continued his association with Commodore. “I came back and worked the summer of 1978, and did a fair amount of unofficial consulting for several years after that,” he recalls. Peddle’s closest rivals also paid him a visit in the hospital. “Some people that came by weren’t from my company—Steve and Steve,” he says, referring to the Apple founders. “We had always kept in touch and we had always stayed friendly during that period.”

After his release from hospital, Peddle continued pushing the disk drive project forward. “We signed up for a set of [disk drive heads] sold to us by Simon Connors, from Shugart Associates after Alan [Shugart] had been forced out,” recalls Peddle. The disk drive head was a crucial mechanism in the disk drive, allowing it to physically read and write to a floppy disk. Apple also relied on Shugart for their drives. In July 1978, Apple delivered the disk drive at the introductory preorder price of $495, and $595 after that. It was an incredible price breakthrough for a company that had the costliest microcomputer of the 1977 Trinity. “The single floppy made Apple take off,” says Peddle. The company Tramiel once declined to purchase and perceived as little or no threat to Commodore was becoming a force. Wozniak’s disk drive was much simpler than the forthcoming PET dual disk drive and it suited most s’ needs. “The [Apple] single floppy didn’t do anything like [handling complex] data, but it let you store and load programs much faster,” says Peddle. Tramiel was unhappy his competitor beat Commodore to the market. “Jack Tramiel almost killed me,” laughs Peddle. “He said, ‘You’ve been fucking around making this high-end dual machine and you let them eat your lunch.’” Peddle received his first Jack Attack. “The guy was just kicking my ass because I deserved it. There was no shouting and there was no hate. I fucked up and he was telling me I hurt the company really bad.” Tramiel was clear about his reasons for punishing Peddle. “His view was that I was spending too much time on other things.” Peddle had previously had his division handed over to Chris Fish, but what came next crushed Peddle. “Jack punished me and, to get me focused, he took me totally out of management of the computer business,” recalls Peddle. “He said, ‘You are not going to do management anymore. You are only going to do development.’” Peddle would no longer have a say in management of the computer systems division he had built for Commodore. * * * After his demotion, Peddle felt cheated. “We started out with an agreement that I

would run the systems business,” he says. If anything, it was amazing the two headstrong men were able to work together for as long as they did. “You can imagine, when you get personalities like Jack and Chuck together, it could get quite volatile,” observes Spencer. “They both had very strong plus points and they both had some minus points.” Unfortunately, with Tramiel’s personal tragedy eating at him and Peddle’s inability to focus on engineering, the relationship soured. “Our personal relationship got screwed up during that time,” recalls Peddle. “He was in a bad mood and he was playing with my life in ways in which he played with everybody’s lives.” Peddle also felt unhappy with some of Tramiel’s business dealings. One person caught off guard was Bob Schreiner of Synertek, who collaborated with MOS Technology on the 6502 project. “I went out and cut a deal to get Synertek to do something for me on memory,” recalls Peddle. “We promised we’d buy more memory from them.” Tramiel seemed to have no qualms about breaking agreements. “It was a signed deal,” recalls Peddle. “Jack then said, ‘I know it’s a signed deal, I know I agreed to do it, and I know these guys really helped me. But I don’t have to do it now and I’m not going to.’” “Jack’s attitude was, if he didn’t think it should be paid, he would just rip the check up and tell them to sue him,” says Peddle. Bob Yannes recalls Tramiel’s philosophy. “His motto was, ‘Business is war’. You didn’t win by competing with your competitors or outdoing them; you won by destroying your competitors and putting them out of business.” It was often difficult for Commodore’s business partners to survive. “He had suppliers make a bid on something and he would say, ‘That’s too much. This is what I want it for,’” recalls Yannes. “They would say, ‘But we’re going to lose money if we sell it to you for that.’ He would say, ‘Do you want the business or not.’” Surprisingly, many suppliers accepted Tramiel’s offer. “They would sell him stuff at a rate where they were going broke,” says Yannes. “They would go out of business and Jack would go find another one. That’s just how he did business.

Everyone knew he did business that way so it’s their fault if they got burned by him.” Tramiel’s aggressive policies might have helped Commodore in the short term, but they alienated beneficial allies in the electronics industry. “They survived it and we survived it,” says Peddle. “But [Bob] Schreiner said, ‘I will never trust Jack again.’” The broken agreements placed Peddle in an awkward position, since he was often responsible for making the deals in the first place. “It was just obvious Jack was not going to keep his word, and it was my word because I had cut the deal,” he says. “I quit cutting deals for Jack,” reveals Peddle. “I told him one day, ‘When I cut a deal, I mean it. You can cut the business deals and if people want to go along with you that’s fine but I will not cut a business deal.’ And I never did.” At first, Peddle tried to work through his problems at Commodore. “Some of the time when you were there you just put up with it,” he says. Gradually, Commodore was losing Peddle’s loyalty. At the same time, Apple was desperately seeking a new head of engineering because of their dissatisfaction with Steve Wozniak. After the difficulty in motivating Wozniak to develop the disk drive, Apple management wanted more stability. “They suddenly realized that their entire company was dependant on this one petulant 26 year old,” explains Peddle. “Markkula, Jobs, and particularly Mike [Scott] said, ‘We’re not going to be in that situation again. We’re going to go hire technical people that are better than Woz.’” In August 1978, Apple came to Peddle in search of a head engineer. “Jack fucked me up and Steve [Jobs] came to me with a great offer,” recalls Peddle. And what was the job offer? “To replace Woz and be the technical guru for the company.” Apple used everything in their power to lure Peddle. “I met with [Mike] Scott and Steve [Jobs],” says Peddle. “Basically the guys at Apple offered me a deal that would have made me one of the top five stockholders at Apple.” The offer was too attractive to decline. “I guess he’d just had it,” says Seiler. “He was always fighting with Jack

Tramiel. There were always differences about how the business should be run. Jack was cheap. He was trying to keep the whole thing floating. He was wheeling and dealing to get enough money to buy enough parts. He ran out all the vendors out to maximum. Back then, you could go to 60 or 90 days before you had to pay. He was doing that. He got TI so mad at us they almost wouldn’t sell us any parts.” “Chuck and him finally had it out and Chuck left and went over to Apple,” recalls Seiler. “That was kind of scary.” Seiler recounts the final meeting with Peddle. “He came in and got us all there: Feagans, me, Patterson, and Stark. He said, ‘I’m leaving.’” Peddle’s engineers wanted to continue working with him. “We all wanted to jump ship with him and go to Apple,” recalls Seiler. “He said, ‘You guys stay here because you’re working on important stuff. Finish it.’” In the summer of 1978, as the Commodore’s Flying High hit the top 40 on the music charts, Peddle walked out on Tramiel. Leonard Tramiel recalls Peddle leaving. “I assume it was because he thought he could do better there or had more flexibility,” he says. Incredibly, Peddle walked away from a fortune in Commodore stock. “Terrible decision on my part,” he says. “I left Commodore without sorting out my life with them from a financial standpoint.” Seiler recalls the deals Peddle had made with Tramiel. “He had a deal at one time that he was supposed to get a dollar for every PET that they sold. That would have been millions of dollars. Jack was able to somehow claim he had reneged on their agreement.” As might be expected, the sales pitch made by Steve Jobs was better than reality. “Going to Apple was a terrible decision,” says Peddle. The Apple executives had no real plan for the senior engineer. It soon became apparent he would not enjoy the same freedom he previously held at Commodore. “It turns out they had, through a totally independent channel, hired this HP guy that Woz had worked for,” recalls Peddle. The rival engineer was John Couch. “Woz loved him and they thought he was the greatest guy in the world.”

Peddle was still new to Apple and trying to concentrate on engineering, so it took him a while to see the true picture. “I walked through the door, and it was obvious they didn’t expect both of us to accept,” he says. “They basically hired both of us for the same job.” Peddle found himself in competition with Couch. “The problem was this guy was very good politically. He was a typical HP builder type who wants a good desk and all of the perks. I don’t have fancy offices and stuff like that. It’s not what I do.” Peddle’s first task at Apple was to help Steve Jobs on the new Lisa project. At this point, Jobs had not yet visited Xerox and raided their best ideas. “I worked closely with Steve through the time of the Lisa. We were close. We used to go to lunch together and go for walks in the woods.” [1] When IBM released their personal computer in 1982, Epson benefited enormously. “For a long time, Epson was the leading supplier to the PC market,” says Peddle.

CHAPTER 12

New Headquarters 1978-1979

After Chuck Peddle’s sudden departure, Jack Tramiel replaced him with Peddle’s colleague and occasional jogging partner, Bob Skyles. In late 1978, Commodore began churning out upgraded PET computers. When showing the computer to potential customers at the previous CES, they had received that the graphical PETSCII characters on the keys made it difficult for professional typists. In response, Commodore released two models: a regular version with PETSCII characters on the keys and a business model without. Commodore dubbed them the PET 2001-N and PET 2001-B. Because the larger keyboard took up more space, Bill Seiler had to remove the built-in cassette unit. Instead, Commodore introduced a separate product, the CBM C2N Cassette Drive, for a retail price of $95. Commodore also released their first printer in 1978, the CBM 2020, which Chuck Peddle had instigated before leaving for Apple. This was also the first Epson printer on the North American market. Eventually, the Japanese manufacturer would become the leading supplier of printers in North America for much of the 80s and 90s. Commodore made a name change before releasing the revised computers into the European market. Improbably, another company reserved the PET name. Electronics behemoth Philips, based in the Netherlands, was just about to release a professional computer, which they called the Programm Entwicklungs Terminal (Program Development Terminal). Though it would not pose a threat to the Commodore PET, Philips legally held the rights to the acronym PET. Jack Tramiel picked his fights carefully and he was not about to tangle with a massive

company like Philips. Commodore changed PET to CBM, hoping to elicit comparisons to IBM. The PET and CBM line became the most business-oriented computers of the 1977 Trinity. “The focus was on business computers as Dennis Barnhart had ed as VP of marketing,” says John Feagans. The only part missing from the business system was a disk drive. Chris Fish, a member of Tramiel’s inner family, was the VP of engineering at the time. Fish had been running the personal computer business after Tramiel demoted Peddle in 1978. “Chris got to do that for about six to nine months, and the next thing you know, Jack said, ‘You’re not doing a good enough job at this so I’m sending you to set up distributorships.’” “Gary Summers became our new engineering VP,” says Feagans. In the autumn of 1978, Tramiel sent Fish on a punishment tour. “Chris had been assigned to effectively fly around the world and set up distributorships,” says Peddle. “There were all these stories of being literally ostracized from his family for a long period of time.” Immediately afterwards, Peddle’s replacement Bob Skyles departed Commodore. “Bob Skyles left to start his Skyles Electric Works doing products for the PET,” says Feagans. His first products included memory expansions and full-sized keyboard replacements for the PET 2001. Tramiel replaced him with an engineering manager named James D. Kennedy. The new manager started several new projects. “In the fall of 1978, Jim made a trip to MOS in Valley Forge to meet the management and familiarize himself with the chips available from our internal sources,” says Feagans. “It was there he learned of the VIC chip.” The VIC chip was a relatively new product and so far, Commodore had been unsuccessful in enticing outside product development. Kennedy thought it made sense for the systems group to develop a product using the chip. “Kennedy started talking about doing a color computer using the VIC,” says Feagans. “Returning to Palo Alto, he immediately pitched to Bill and me about doing a color computer. Jim committed the design to paper and several memos circulated.”

However, the focus was on enhancing the PET computers with features for the business market. According to Feagans, “This included a dynamic RAM-based PET motherboard, 80-column display, printers based on the C. Itoh and Epson mechanisms, and the dual floppy disk drive.” The large number of new projects overwhelmed the engineers, and the VIC color computer faded in importance. “We liked the idea, but it was back burner to the projects for winter 1979 CES that Jim had undertaken,” says Feagans. Kennedy wanted a series of new models that were expandable to 32 kilobytes using dynamic memory.[1] “We eventually changed the board so it would take DRAMs,” recalls Seiler. “We started building 16 and 32K machines.” Commodore hired an engineer with expertise in computer systems with DRAM. “We hired a cowboy to do the first dynamic RAM PET design,” says Seiler. “He drove a big old truck and lived on a ranch with his girlfriend, and they had horses.”

Jim Kennedy planning Commodore’s next PET products.

The design worked, but needed some refinements to make it more cost effective to produce. “We put his in production and it worked okay,” says Seiler. “He did the first DRAM design and I kind of redid it again. We did some optimizing on it.” Unfortunately, they were not able to complete the disk drive by the end of 1978 as planned. “Jim Kennedy was not real pushy and at that time I think we needed to be pushed,” says Seiler. “He was kind of a big happy guy, not too driven or anything but he kind of massaged over problems and got things going.” * * * In late 1978, Mensch and his team completed the reverse engineered 5-function calculator chip. “Toshiba had some interesting, tricky ways of doing some things to camouflage what was really going on,” remarks Mensch. “But we got it to work and we got the LC5K3 chip out and running.” Commodore would still have to spend a significant amount of money to finish off the project. “Mensch got it working, but they were going to have to write the test program for the testers,” explains Seiler. “That was going to cost another $100,000.” When Tramiel realized the project would cost more, he made an astute move. “So Jack Tramiel leaked it to Toshiba,” says Seiler. Tramiel just wanted to pay the “Japanese price” for the chip. Mensch was counting on Tramiel to supply royalties to WDC on the LC5K3 chips. Instead, Tramiel used the reverse engineered chip to negotiate with Toshiba. “Sam Tramiel did the negotiating for the chip prices,” recalls Mensch. Toshiba reacted to the news exactly as Jack Tramiel hoped. “Toshiba found out and lowered their price,” says Seiler. Knowing that MOS Technology could potentially sell the same product on the market, Toshiba gave Commodore a fair price for the chips. “Jack got the

Japanese price on the calculator chips so they never built them,” says Mensch. The astute business move had the engineers wondering if Tramiel had them working all year with no intention of using the chip. “We contemplated that Jack did the whole thing just to get the price lower from Toshiba,” speculates Seiler. “The whole project was just part of a chess move in his business plan. Jack was like that.” Although Commodore paid Mensch for his office costs, he had no profits to show for his work. “That’s when I realized the LC5K3 didn’t work out for me, but it worked out for Commodore.” * * * By late 1978, MOS Technology had found no interest from manufacturers in the VIC 6560 chip. “There were samples being given to places like Japan,” recalls Charles Winterble. Unfortunately, the company was not able to find a customer to use the chip in mass quantities. It was beginning to look like Al Charpentier had wasted valuable resources developing a product no one wanted. Charpentier began looking for rookie engineers to develop new projects. “I was looking to hire some people because when Commodore took over MOS Technology, there was an exodus that happened over a year or so,” he recalls, referring to the departure of the 6502 team and the calculator engineers. “There were only a few people left in the semiconductor design department, so I wanted to hire some people.” Charpentier visited the local universities in the Philadelphia area in November 1978. “Bob Yannes was a senior at Villanova, so I interviewed him. When he came in for the interview in our office, he saw the VIC-I chip that I was working with and he got really interested in it. He wanted to know if he could do his senior project on the chip.” Yannes re the incident. “When Al told me about the VIC chip he had worked on, I was like, ‘Great! Tell me about it.’ I got him to send me some samples.” Charpentier was glad to receive some overdue enthusiasm for the color chip he designed. The student would use the remainder of his senior year to piece together a project using the VIC chip. * * *

According to Byte magazine, by the end of 1978, Commodore had managed to sell approximately 4,000 PET computers in North America. Total sales for Commodore International reached $50.1 million, mainly from calculator sales. The year ended with profits of $3.4 million, a welcome change from losses in the previous year. The newfound success of the PET allowed Commodore to pay off debts and solidify some of their holdings. MOS Technology had retained their headquarters at 950 Rittenhouse Road since the 1976 acquisition, but had been leasing the building from Allen-Bradley. In 1978, Commodore finally had enough cash to purchase the entire building. By the end of 1978 Apple was not faring well compared to the rest of the 1977 Trinity. “Apple was getting into serious trouble,” says Peddle, an Apple employee at the time. “Their sales were dropping off, they didn’t have market share, and [Commodore was] kicking them. Radio Shack had come out with two or three models and was doing pretty well and the /M machines were starting to come along. Apple almost went away.” Apple was selling approximately five computers per day. In contrast, Radio Shack sold over 55,000 TRS-80 computers in their first full year of production. Despite poor sales, Apple began to claim they were number one. “That was the kind of stuff they did,” says Peddle. “Markkula turned McKenna loose and said, ‘Any lie McKenna tells, as long as I don’t tell it, I don’t give a fuck.’” In December 1978, Apple computer ads began calling the Apple II, “The World’s Best Selling Personal Computer.” Despite the lofty claim, both the PET and TRS-80 outsold the Apple II. “They kept building on that image and just kept telling lies,” says Peddle. “McKenna has always been good at that. He can make you believe any lie he wants. Clearly, during that time, that whole revisionist thing started.” Despite their third-place standing, Regis McKenna was determined to promote Apple as the clear winner. “Apple did a great job of promoting the success of their product, whether it existed or not,” says Leonard Tramiel. “I know that through that entire time, Apple had ads claiming that they were number one, where in fact they weren’t. Every month, they had an ad asking the question, ‘Why is Apple number one in the world of Personal Computers?’ And the

answer of course was because they were number three, behind Radio Shack and Commodore. They just lied.” Peddle was at Apple in early 1979 when they first began thinking of using a graphical interface. “Jobs and I spent all this time doing architecture work on what was then the Lisa. I clearly Jobs went to see these guys [at Xerox PARC]. He and I had lunch and he was just totally in love. He said, ‘This is what we have to do’, and he put it in the Lisa.” “The Lisa was going to be the Xerox Star,” says Peddle, referring to a minicomputer in development at Xerox and later introduced in 1981. “The Xerox Star was a tremendous product. It had a very high-resolution screen and a tremendous word processor. The Lisa turned out to be the Star but it wasn’t hard disk-drive oriented.” While Peddle concentrated on engineering, his rival John Couch played politics, according to Peddle. “This guy effectively went in and started playing power games. He consolidated his position, ran his position up, hired more people, and became head of engineering. I wound up working for him, which was not the deal at all.” Peddle predicted Lisa would be a marketing disaster. “I basically walked out on Lisa when it was obvious that they were going to make it a ten thousand dollar box,” explains Peddle. “My words were, ‘Apple doesn’t have the ability to sell a ten thousand dollar box, nor does the market need one.’ They got all huffy about that so they reassigned me to the Apple III.” After management removed Peddle from the Lisa project (they also later removed Steve Jobs), it fell to Peddle’s rival John Couch to complete the project. “He’s the guy that ultimately led to the Lisa fiasco,” says Peddle. In the meantime, Peddle concentrated on the Apple III architecture. “I sat down and defined it as a bridge product, which wasn’t the product they came out with. I wound up hiring two or three guys and we kind of set off in a corner and started doing some work.” As the months went by, Peddle realized it had been a mistake to leave Commodore. “The most important thing that was in my heart was that I felt I had left the thing I created,” says Peddle, choking up slightly. “What we were doing [at Apple] was interesting and all of that but my heart just wasn’t in it.”

* * * At the January 1979 CES in Las Vegas, Commodore’s engineering manager Jim Kennedy brought the recently released PET 2001-N and 2001-B computers, along with the Epson-designed CBM 2020 printer, and a prototype of his dual disk drives. The drives were not really working yet, but it was important to let distributers know they were on the way. At the same CES, Commodore released the CBM 2022 and 2023 printers, which sold for $795.00 and $695.00 respectively. These were true dot matrix printers, allowing text and graphics output, much like inkjet printers. The 2020 printed upper case, lower case, and PETSCII character graphics. Commodore also previewed two other printers behind closed doors at CES, the 4021 and 4022 printers. “I wrote [the ROM code] and brought up the 4021 and 4022 in December 1978 and both were shown at the 1979 Winter CES,” says John Feagans. During the show, Bill Gates inspected the new PET computers. He tried typing WAIT 6502, 1 to check if his Easter egg still worked. John Feagans and Bill Seiler watched him from a distance. “I Gates walked up to one of our machines at one of our shows and typed that and it didn’t do it anymore,” says Seiler. “He was kind of tweaked about it because he had put that in and spent some time hiding it.”

Bill Gates in front of the Microsoft booth.

Tramiel eagerly anticipated CES because it gave him a chance to check on the progress of the competition. This year, a new competitor met him on the battlefield. Tramiel was friendly with the biggest customer of his components division, Atari. This was the first year Atari founder Nolan Bushnell was not present, however. The new owners of Atari, Warner Communications, forced him out a few months earlier. Since Peddle’s deal with Atari in 1976, Commodore continued to sell the 6507 microprocessor to Atari. “MOS shipped a lot to Atari,” recalls Al Charpentier. “Apple wouldn’t deal with MOS for some reason, but we shipped a ton of those 6502s.” Atari was also hungry for ROM chips, since each game cartridge contained a four-kilobyte game program stored in ROM. MOS Technology produced some of the best ROM technology at the time. “We were their biggest supplier of ROM for their games,” says Bob Yannes. Atari even used the 6502 chip in many of their coin-operated arcade games, including Super Breakout (1978). They also developed several groundbreaking vector games such as Asteroids (1979), Lunar Lander (1979) and BattleZone (1980), all using the 6502. Tramiel was curious to see the new computers that Atari had recently announced. At the Winter CES, Atari showed off the Atari 400 and Atari 800 home computers, code-named Candy and Colleen. These were legitimate computers with keyboards and the ability to display powerful color graphics on a television. It was immediately clear to him that Atari could become a powerful competitor in the computer industry. Not only did they have a good reputation with consumers, but they also had a huge retail distribution channel for the Atari 2600. All they had to do was introduce the 400 and 800 into this channel and they would almost certainly have success.

* * * The PET and CBM computers had made Commodore exceedingly profitable. After the madness of the January CES subsided, Tramiel decided to amalgamate Commodore’s operations, including the manufacturing facility on Reed Street, into one large building. “It was pretty crazy there at Reed Street,” says Bill Seiler. “They moved us quickly to something better at Scott Boulevard.” Tramiel closed down his modest headquarters at 901 South California Avenue in Palo Alto and moved Commodore into a high-tech building located at 3330 Scott Boulevard (next door to AMD). “They started building the offices on Scott Boulevard and Commodore moved into it brand new,” says Seiler. The single story concrete building looked a bit like a bunker, with angled windows along the front and large solar s on the roof. “That was a real bizarre building with a real slanty front,” says Seiler. “The corporate offices were up in the front and the back half of the building was production. It had a real production line, a real cafeteria. It was pretty nice.” The state-of-the-art building could also deal with seismic activity. Giant metal springs and steel cables helped dampen the effects of tremors.

Commodore’s 1979 headquarters.

Commodore executives took the prime windowed offices along the front. “Jack’s office was way over in the front corner,” says Robert Russell. The corner office gave Tramiel more window space than anyone else, but it sat under the large angled glass windows, making it the most dangerous place to be in case of an earthquake. Beside him was the conference room where executives and managers laid out their battle plans. Tramiel placed the istrative people near him, in a large area crammed with desks and cubicles. These were the Commodore foot soldiers: ants, the purchasing department, and sales people. Their area was the most chaotic in the entire building. Behind the istrative and technical areas was the factory, which produced PET computers. Tom Mitchell, one of Tramiel’s trusted executives, oversaw manufacturing. Mitchell kept the facility more organized compared to the haphazardly constructed assembly line at the old Reed Street facility. “Scott Boulevard had real palette rolling and people sitting down like you would see on production lines,” says Seiler. “They were building PETs like crazy there.” At the new facility, the engineers realized they would have to abandon part of the vertical integration strategy with the PET computers when they gave up on the unreliable MOS Technology RAM. “We finally designed those out and started buying 2114 industry standard parts,” says Seiler. “It took two of them to make a byte. I think there were 8 on a board for the 4K PET, and 16 on the board for the 8K PET. Everything started working a lot better in production.” The engineers also disliked the MOS Technology ROM chips. “They were funny parts that were made for embedded games and stuff like that,” says Seiler. “We threw their parts out and we just started buying the standard 2716 off the shelf from all the ROM vendors out there. Everything worked great after that. They started working good on the production line and yields started going up.” The warehouse section was at the back of the facility. It had a high roof to

accommodate boxes that were often stacked to the ceiling. Here, a surly group of workers loaded shipments into delivery vans that pulled up to the large doors. Sometimes they ran out of space. “When product backed up it piled up in the hallways in engineering,” recalls Russell. The building also housed the engineering labs, where Commodore engineers devised new products. “Engineering had a hardware lab in one corner and the drafting people were up in the front under the windows,” says Russell. In 1979, the engineering projects were in trouble. * * * Tramiel had previously sold LED (Light Emitting Diode) calculators and watches on the consumer market. However, technology never remains static, and calculators were making the transition from clunky models to lightweight pocket devices powered by miniature batteries. In order to create low-power calculators, Commodore would need to either purchase or build their own low-power calculator chips. “Their calculators were all low-power PMOS, but not really low-power like CMOS,” explains Mensch. “They wanted CMOS calculator chips so they could compete low-power-wise with Toshiba, who was the leader at the time.” Andre Sousan had previously scouted out the MOS Technology acquisition for Tramiel, but now Tramiel relied on a man named Bud Fry to make the scientific calculator happen. “Jack had hired a marketing guy who knew semiconductors pretty well,” explains Peddle. “Bud Fry had this idea he wanted to do something in calculators.” MOS Technology had both the PMOS and NMOS process, but they were not capable of producing CMOS. In late 1978, Tramiel found a semiconductor house to manufacture CMOS. “Commodore acquired Frontier Semiconductor,” recalls Mensch. The new acquisition was located over six and a half hours south of Commodore’s headquarters in Santa Clara. “Frontier made the watch chips in Costa Mesa, California,” says Feagans. It was a natural transition to begin using the chips in low-power calculators. After the acquisition, MOS Technology and Frontier fell under a subdivision of

Commodore called Commodore Semiconductor Group. “Commodore acquired other companies and put them all under that umbrella,” explains Al Charpentier. To extend battery life, calculators also needed a more energy efficient display than the power-hungry glowing red LED displays. The answer was in Liquid Crystal Display (LCD) displays. Although LCD was a purely Western invention, Japanese calculator makers began undermining Commodore’s prices with LCD. “Commodore was making LED calculators and LED watches, and they got killed by the LCD,” recalls Peddle. Tramiel’s first reaction was to attempt to purchase LCD components from Japan. “That was when they first discovered how clever the Japanese were,” says Peddle. “The Japanese basically told them that there was too much internal demand and that’s why they wouldn’t sell them the LCDs or the parts for it. Then they hammered into the market, Seiko being the primary driver.” While Peddle was away at Apple, Tramiel progressed with his vertical integration strategy by looking for two more companies that would allow Commodore to produce their own LCD electronics. “That’s when Jack went out and bought himself an LCD company and a CMOS company,” says Peddle.[2] Tramiel found a potential source of advanced LCD technology with Dallas manufacturer Micro Display Systems (MDS). An engineer named Tom Hyltin founded the company and sold his LCD product to Seiko to use in their digital watches, and thereafter MDS became a subsidiary of Seiko. In 1979, Seiko was willing to part with their manufacturing plant. Tramiel acquired MDS by giving 100,000 shares of Commodore stock to Seiko. At the time, it looked like a good deal for Commodore since their stock had climbed quickly in the last three years and was likely to stabilize. Tramiel amalgamated his existing LCD line in California with MDS. “The LCD line moved to Dallas, Texas through [the MDS] LCD acquisition,” says Feagans. He renamed the merged operation Commodore Optoelectronics. An executive named Elton Southard managed the division. Bud Fry’s goal was to create a line of lightweight scientific calculators for Commodore that used low-power CMOS chips and low-power LCD displays. Now all he needed was an engineer to design a scientific calculator chip in CMOS.

Now that Mensch was done with the Toshiba clone chip, Fry suggested using him and WDC to design the new scientific calculator chips. Mensch knew Peddle wanted back in Commodore, so he suggested bringing Peddle onto the project. Fry went to Tramiel with the request. “He basically got Jack to give him permission to come and pick me up to go do this thing,” says Peddle. “They put me and Bill into a deal at the same time.” With the door open to return to Commodore, Peddle decided to leave Apple. “When we parted, Steve [Jobs] said, ‘You never looked happy here. You just never looked like you liked it.’ And he was right.” With almost no hesitation, Peddle gave up a major stake in Apple that would have been worth millions. “I wanted to go to Commodore. If I had been money oriented, I would have sucked it up, skated through all this bullshit, and been a very rich guy.” The decision seems to trouble Peddle. “Leaving and coming back from Apple without getting a major stock position was another stupid decision, but I really wanted to come back.” The Apple III project turned into one of the biggest disappointments for Apple, largely because of poor production design that caused the chips to seat improperly in the motherboard (something that had nothing to do with Peddle’s contribution to the project). “It turns out that the guy who did it wound up making some pretty bad mistakes,” says Peddle. Steve Jobs realized Peddle was no longer any use to him and the friendship ended—a characteristic often noted by people who have worked with Jobs. “We parted ways after I left Apple,” says Peddle. “We never looked back, either one of us at that point. He wouldn’t give me the time of day now because Steve is Steve.” “He turned into something that he wasn’t, and maybe it was always under there,” muses Peddle. “Steve was a friend at one time. Steve lost most of his redeeming characteristics. He got rich and ugly and arrogant and a whole bunch of other things.” Ray Hirt, a member of the 6502 team, worked with Mensch as a WDC employee. A fourth member of the original 6502 team, Will Mathys, was also set to the project. “Will was going to be part of the design team. He was going

to be a consultant to WDC for the chips,” says Mensch. “I said, ‘Alright, we’re going to work together on this. Great!’” The team had made a breakthrough for microcontroller devices with the 6502 chipset and now they hoped to create something similar for the LCD market. Commodore hoped to sell the semiconductor devices to manufacturers of calculators, meters, and anything that used an LCD screen. “The idea was to have a product that could be used to sell intelligent LCD products,” says Peddle. “It would be highly programmable and easy to adapt to a bunch of applications.” Their chip design was not limited to calculators, and could conceivably power an LCD computer. “We were making a small microcomputer that was LCD oriented,” explains Peddle. Mensch was excited to be back at work deg semiconductors. “Now we’re going to do a scientific calculator, it’s going to be a brand new architecture,” recalls Mensch. “So then I built up for that.” The 6500 series of chips did everything from microprocessor functions (the 6502 to 6510 line), to memory, to video. Mensch named his new chips the 4500 series. “It was like the 6500 only 4500,” he says. Chuck Peddle focused on the LCD chipset, while Bill Mensch focused on the scientific calculator chip. “I did the design of a 4-bit data bus processor that could work with a scientific calculator,” says Mensch. “It was a reduced instruction set 4-bit microprocessor that could do multiple 4-bit operations ranging from 4- to 64-bit operations.” Bill Seiler was impressed with the unique design. “It was a step back to a 4-bit microprocessor but it was pretty cleverly done. The instruction set was variable length, so you could get a lot of utility out of it,” he says. “Mensch was really good at that.” However, Peddle would not get very far on his part of the design. His true focus was the PET engineering group. * * * In early 1979, Peddle discovered a way to return to his beloved PET. While Bill Mensch was working on the LCD project at the Western Design Center, Chuck

Peddle visited the new Commodore headquarters at 3330 Scott Boulevard to check out the progress on his beloved PET. He was looking for opportunities to get back into the computer division. “While I was away at Apple, this engineering manager, Jim Kennedy, was brought in to replace me,” he says. “I was only gone for a short time and this one guy totally fucked things up.” Commodore was supposed to ship the disk drives in late 1978, but after Peddle left for Apple, the project fell into development hell. “Management had let these guys get totally off track on what the disk drive was supposed to do,” says Peddle. “I discovered that the machine was going to be a serial read/serial write device—effectively two tape drives. Well, that’s not what you use floppies for.” Peddle used Kennedy’s failure with the disk drive to ingratiate himself back into the systems division. “Chuck was supposed to work on a processor for calculators with Bill Mensch—and I got involved in calculator software—but soon Chuck advised all aspects of PET development,” recalls Feagans. “Jim resigned when Chuck Peddle returned in February 1979.” The takeover was easy, since he had the full of the engineering team he had personally hired. “I jumped in on the disk drive and then effectively forced Kennedy out and took over all engineering,” says Peddle. “I managed to convince them to cancel it.” Peddle negotiated with Tramiel to reinstate his previous financial agreements. “They promised me that they would finish my stock option deal,” he says. However, Peddle could not get every agreement reestablished. “I had a deal that I would get a dollar for every computer they made and I lost that deal, which is okay. I never expected him to honor it anyway.” Although Peddle was able to continue his original stock plan, it would take more time to earn back Tramiel’s trust. “Jack didn’t forgive me for leaving for a while —maybe ever,” recalls Peddle. “We developed a personal relationship again but there probably was not the degree of trust between us that would have been there if I had stayed.” The engineers started the project over from the beginning. “We sat down and redefined the whole product after I came back.” The redesign meant the disk drive would not appear for several more months, but Peddle was determined to do it his way.

The Apple disk drive could only store 150 kilobytes on a 5.25-inch disk. Peddle’s 2040 disk drive could store 170 kilobytes by taking advantage of the outer rings of a disk. A disk is circular like a vinyl record, and the disk drive head writes data in concentric circles around the disk. Because the concentric circles are larger at the outer edges of the disk compared to the smaller rings near the center, Peddle realized it could write more data on the outer rings by simply changing the speed of the motor. “We did things with densities on floppies at Commodore that put us way ahead,” recalls Peddle. The 2040 was also useful for more than just loading and saving programs. “It was a real honest-to-goodness file management device,” says Peddle. “You could read and write records, you could go in and sort records, and you could do all the things with that product that you could only do with computer systems dramatically more expensive.” The 2040 was a massive unit, almost as large as the PET itself. It even had a hinged lid for easy access, much like the PET, with a arm to hold the lid open. The 2040 drive was suited for the low processor speed of the PET computer. The 2040 dual drive had two processors, a 6502 and a 6504. “A 6502 was used to run the disk operating system and handle the traffic on the IEEE-488 bus,” says Feagans. “The supervisor communicated with a 6504 through shared memory to perform the disk controller functions.” Although Peddle had perhaps over-designed the dual drive specs, Bill Seiler believes they did an efficient job with the implementation. “It was a pretty elegant design,” he says. “The neat thing is they shared memory. They had 4K of shared memory. It had two 6502-type processors, and the 6502 only used the memory at half a clock. We ran the clocks out of phase to the two processors, so either one could jump on the memory and use it at any time. So they shared the memory in a really neat little elegant design. It was pretty fast too.” The 4K of memory and internal code resided on the 2040’s RAM and ROM chips. In effect, the 2040 drive was a standalone computer. “We went to distributed intelligence peripherals, so there really was no operating system level of activity,” says Peddle. In contrast, the Apple II stored the disk operating system (DOS) in the Apple II

RAM, which taxed the memory and U cycles. “The [Apple] disk drive wasn’t really extensible and never did get extended,” says Peddle. With U speeds of one megahertz and very little available memory, computing power was at a . Peddle’s design philosophy left memory and U cycles to the computer. In the spring of 1979, the basic design was ready. “We were a little late on it,” says Seiler. Peddle had come through for Tramiel, and now he allowed him more control over the computer systems division. Determined never to fall behind the competition again, he began to consider a proposal for Tramiel that would formalize research into new products, similar to Hewlett-Packard. [1] The semiconductor company Mostek introduced dynamic memory commercially in 1973. [2] According to Peddle, Commodore also wanted to use the LCD technology in the ChessMate product. “It was supposed to be an LCD also, but apparently the LCD never got started,” says Peddle. Instead, ChessMate used red LED digits.

CHAPTER 13

Business is War 1979

Commodore brought the new business-oriented PET computers to the April 1979 Hanover Fair. With an improved keyboard, a larger monitor, and more memory, the CBM computers (as Commodore branded them in Europe) were legitimate business machines. “The Europeans were basically putting us into competition against Hewlett-Packard and all the others making that level of computing device,” says Chuck Peddle. The new PET computers received acclaim at the Hanover Fair. When it came time to vote on the Computer of the Year for 1978, the PET won. “At the Hanover show in , we had really incredible booths in the early days,” says engineer Robert Russell. Commodore ’s general manager, Harald Speyer, ran the division from his headquarters in the city of Braunschweig. “Harald Speyer was an exceptionally good manager and did a tremendous job in ,” says Kit Spencer. “The general managers in Europe got along very well. We didn’t fight, we learned from each other. We all benefited from our own profit and loss responsibilities and we all essentially reported to Jack.” Speyer ultimately brought success to Commodore with the PET computer. According to Peddle, “The Germans always pay more for their machines than anybody.” Speyer increased the price and marketed the PET to the industry rather than home consumer s. Jack Tramiel did not attempt to curtail this strategy, since in 1978 he was still feeling out the real market for microcomputers. Germans accepted the PET as a serious business machine.

Tramiel soon realized some European retailers were making modifications to the PET computers that he did not approve of. “They found that the overseas dealers would buy a bunch of the cheaper 16K machines and put in more memory themselves,” explains Bill Seiler. Commodore was losing potential profits from the more expensive 32-kilobyte machines. “So Jack had somebody pretty clever figure out where he could drill holes right where those chips went.” The holes made it impossible for the dealers to upgrade their memory, forcing them to buy the 32-kilobyte models from Commodore.

Commodore ’s headquarters in Braunschweig.

Commodore took a clear lead over the 1977 Trinity in total worldwide sales, mainly due to the European operations. “Commodore was just destroying Radio Shack and Apple,” recalls Peddle. Commodore introduced the PET into other European markets, but not all. “The Spanish weren’t a big market at that point, although later on they became an enormous market,” says Peddle. However, Commodore was able to penetrate into most of the smaller Western European nations. “The Scandinavians absolutely paid good prices.” One of the most interesting battles in Europe was in between Commodore and its former ally, Andre Sousan, who now worked at Apple. The Apple brand was unknown in Europe and it would take serious resources to compete against the already established Commodore name. French laws at the time restricted American companies from opening divisions in . It was up to Tramiel’s friend, Elie Kenan, to open the market for Commodore products. Kenan used his company Procep to market Commodore computers in . Meanwhile, Sousan established a company called Euroapple to distribute the Apple II. Under his agreement with Apple, Sousan funded the European operation himself. According to Peddle, “One of the reasons they didn’t do better against us is that Andre wasn’t playing with the kind of money we were willing to play with.” As a result, the Apple II competed on technology and price alone, without the advantage of skilled marketing, and it did not fare well. * * * In May 1979, Peddle received two more programmers from Iowa State University, Bob Fairbairn and Robert Russell. By the time Russell graduated, he was well aware of Commodore technology. “I was in charge of the labs where the original PETs were installed in the University,” he recalls. “We used the 6502 emulators and programming devices that existed at that time to actually design classes around them.”

The two recent graduates already knew most of the Commodore engineers, including Glen Stark, Scott Paterson, and John Feagans. “Bob Fairbairn was my friend from Iowa State,” says Feagans. “Bob and I did many microcomputer projects and scavenger hunts at surplus stores.” According to Russell, the classmates did not agree when it came to microprocessors. “I actually argued against them in college in the computer labs about some of the processors they were pushing,” he says. “I was an Intel man and they were Motorola and MOS Technology pushers at that time.” At Iowa State, Russell purchased a Sol computer kit from Processor Technology. “I put one of those together on my dorm desk in my first year of college. It basically took over my desk.” The new technology was fraught with peril. “I had to build my own power supply, which caught on fire. I was typing away and it was burning right in front of me, but I was looking at the TV screen. Luckily my roommate noticed.” When it was time for Russell to graduate, Dr. Roger Camp felt he had another weird engineer for Chuck Peddle. “[Camp] got the other guys into Commodore and was well hooked up to the 6502 community,” says Russell. “He really pushed me through.” The young engineer decided the time was right to leave Iowa. “What do you do in the Midwest? You go someplace else to get a job,” says Russell. “I had good job offers but they were cold places. All of a sudden, I had an offer from Commodore to go work in Santa Clara with my old friends. California definitely looked a lot more appealing than my other options at that time, which was to go work in Minneapolis.” Compared to the other reclusive engineers, Russell was strikingly normal. “If you’ve ever seen a picture of Russell, he was pretty round,” says Peddle. “He looks like a corn fed Iowa farmer.” Russell enjoyed working for his new boss. “I really liked Chuck Peddle personally. He treated the engineers well. Honestly, they were part of his family, though obviously the other guys were more favored than me,” he says, referring to the veteran engineers. “I going over to his house and looking at the car he had taken apart in his garage. It was a Triumph or Stag or some sort of import British car that was in pieces. I always wondered if he ever put it

together.” * * * The scientific calculator project was a reunion of many of the original 6502 team, but when Peddle dropped out, Wilbur Mathys decided to do the same. “After we agreed to do the design of the 4-bit scientific calculator chip, Will [Mathys] without warning, just stopped communicating,” recalls Mensch sadly. “I don’t know why.” Unknown to Mensch at the time, Mathys wanted to get back into microprocessor design, and began consulting with Commodore on a 16-bit sequel to the 6502. Although MOS Technology was very competitive with Motorola and Intel with the 6502, they were weak in the key area of microprocessor research. Engineer Bill Mensch had contemplated a 16-bit chip as they were finishing off the 8-bit 6502. “When it came to next-generation, we were thinking 16-bit in 1976,” says Mensch.[1] Robert Russell discovered evidence of the 16-bit project developed in partnership with Synertek. “I found a September 1978 description of the SY6516, which was a 16-bit 6502,” says Russell. “Synertek was the company that was supposed to be deg it. I have a specification sheet on it with timing and temperature settings. Reading this, it looks absolutely real.” Unfortunately, Synertek became a victim of the competitive semiconductor industry. “I think Synertek got into financial problems right around that point in time,” says Russell. “There were some rough years.” MOS Technology’s financial problems also hampered development of a 16-bit version even before the Commodore acquisition. “Nothing happened for the longest time,” says Charles Winterble. “I think part of it was the financial resources of MOS Technology. They just simply couldn’t afford it. I have no doubt that if they had the resources they would have done a spectacular job because the 6502 was a phenomenal architecture.” * * * At the June 1979 CES in Chicago, Peddle displayed working 2040 floppy disk drives and began taking orders for them. At $1,295.00, Commodore had priced

the product for the high-end business market, and not for the retail market. It had been six months since Jack Tramiel discovered Atari was entering the computer business, and he was curious about how they would price their new products. He sent one of his new engineers, Robert Russell, to scout out the Atari booth. Russell discovered that Atari would sell the lower-end Atari 400 for $549.95, while the more full-featured Atari 800 would sell for $999.95. Neither came with a monitor, and the PET had the 4- and 8-kilobyte computers beat on memory too. Tramiel was still competing in the calculator market, and he was always anxious to check on his biggest rival, Texas Instruments. The company had almost put Commodore out of the calculator business years earlier. To his surprise, the Texas Instruments booth displayed a new product they called the TI-99/4. It was a full feature color computer. Tramiel was particularly wary of Texas Instruments, since they were his only competitor with semiconductor manufacturing capability. This gave them the same edge with vertical integration that Commodore enjoyed. Texas Instruments used their own microprocessor, called the TMS9900, running at 3 megahertz. The TI-99/4 also came with a 13” color monitor. Tramiel wondered how low his competition could go with their prices, and suspected Texas Instruments would be his greatest threat. They priced the TI99/4 at $1150, an impressive price considering it came with a color monitor. It was clear to Tramiel that both Atari and Texas Instruments had superior sound and graphics compared to the rest of the 1977 Trinity. It was unnerving to think that the company that had once almost destroyed Commodore in the calculator market was now going after the computer market. Atari and Texas Instruments committed to releasing their computers in late 1978 and began taking retail orders at the show. However, Tramiel knew that if they would encounter the same manufacturing problems Commodore had with the PET. It would be at least a year before they could produce their products in sufficient quantities to become truly competitive. Tramiel knew he would have to motivate his engineers to deliver something superior in the meantime. * * *

Robert Russell had been with the company for only a month, but he soon noticed the lack of Commodore presence in the North American market. “We were in California and there was no place you could buy a Commodore computer,” explains Russell. “Early on, there was lots of demand [in the United States]. We asked him, ‘Why aren’t there more machines available?’ Jack said, ‘Listen, we’re getting twice as much money for them in Europe.’” The lack of North American success frustrated Commodore’s board of directors. In England and , Commodore enjoyed an 80% market share, despite higher prices than the United States. Robert Russell explains, “You see these price lists in Europe and you say, ‘My God! No wonder Jack doesn’t want to send any product to the distributors in the States.’” However, Tramiel began to feel pressure from Irving Gould to conquer the US marketplace, and that would require an increase in computer production and a focus on marketing. CES was a good time for Gould and Tramiel to meet and discuss strategy. At this particular show, Tramiel became frustrated with Gould’s refusal to raise money using Commodore stock. “Irving was fucking with his head. I when [Jack] was really pissed about that,” says Peddle. “One of the things that frustrated Jack more than anything else in the world was Irving wouldn’t let Jack go out and raise money.” Gould spent much of his effort in New York promoting Commodore. “Irving is a businessman,” says Peddle. “He went out and sold the idea of how smart we were to pump up the stock.” Since introducing the PET, Commodore had doubled and sometimes tripled their revenues each year at a time when most companies were happy to increase revenues by ten percent. The stock market reacted accordingly. By June 1979, Commodore stock climbed to $41.63 per share—a tenfold increase over the 1976 price. The recent acquisitions of Frontier Semiconductor and Micro Display Systems using Commodore stock were beginning to look like bad deals to Gould. With the stock riding so high, Commodore had overpaid for the acquisitions. Tramiel tried to convince Gould to raise money for Commodore to expand production, and promote the PET in the United States. “We’ve got the company

starting to move and Jack doesn’t have the money,” recalls Peddle. Unfortunately, Gould still wanted Tramiel to grow Commodore using profits from the European market. After CES, Tramiel and Peddle visited Gould to plead their case. “We went back to New York,” recalls Peddle. “Jack goes back to the board of directors and says, ‘You guys are just fucking with my life. Go raise me twenty million dollars. The stock’s moving so you can borrow against it.’” The demand was reasonable. “This was stock that was worth thirty times more than it was when I ed the company,” explains Peddle. “They could have easily raised a whole bunch of money and put legs under the company.” Tramiel’s prostrations had no effect on Gould. While Apple used their stock offering to grow the company, Gould refused. “Irving Gould was such a tight son of a bitch that he told him no,” says Peddle. Even more incredible was Gould’s reason for not selling off some Commodore stock. According to Peddle, “Irving said, ‘The stock is going to keep running and I don’t want to give away the cheap stock.’ He just kept doing that.” Gould had forgotten the reason for the stock market. “Jack came out of that meeting and he was so fucking mad,” recalls Peddle. “Irving had cost him market position in the US because he couldn’t do it with very little money. He was so pissed and he had every right to be.” The lost opportunity to dominate North America consumed Tramiel. “It was a tremendous pressure on a guy who was starting to see things work for him,” explains Peddle. According to Peddle, the board told him, “Jack, you’ll find a way to make it happen.” To survive without proper financing, Tramiel studiously monitored costs. “I used to ask him why he did it, because it was pretty tedious,” recalls Peddle. Tramiel replied, “It gives you a pulse of what is going on. Anytime a company is a little bit successful, a whole lot of people start empire building. It lets you figure out who is spending your money when they shouldn’t be.” The unorthodox method allowed Tramiel to keep a tight reign on his company, but it also resulted in unhappy visits to Tramiel’s office. “About once every two

days he would find somebody doing something either fraudulent or stupid,” recalls Peddle. “Then he would go jump all over them.” * * * In the spring of 1979, Villanova student Bob Yannes completed his engineering project using MOS Technology’s VIC. “I built my senior project for school around that,” he says. “I made a color video card that plugged into the S-100 that would give it graphics and sound. That was the Altair 8800 bus that eventually got adopted as an industry standard.” While other people his age played video games at the local arcade, Yannes intently studied the video game chip. “I worked on my senior project in my bedroom and at the lab at school. It wasn’t that complicated a project because the VIC chip did so much of the work. All I really needed to do was add memory to it and then design a bus interface.” Yannes was pleased with the results, but the project was too advanced for his instructor’s level of expertise. “A lot of the stuff was pretty much beyond them,” he says. “This was state of the art back then, and I was hand-assembling code.” Yannes graduated from Villanova University in May 1979. In July, Commodore hired him as a full-time engineer. Yannes recalls, “I it was not long after the movie Alien came out, as I went to see it with some friends and ended up in the front row, which gave me a sore neck that lasted through my first week of work at MOS.” Yannes became a protégé of Al Charpentier and quickly learned semiconductor design. “Once he ed the company, he was working on various projects on calculator stuff with me,” recalls Charpentier. Charpentier gave the young engineer tasks to familiarize him with MOS Technology’s many activities. “I started working at MOS as a junior engineer on all kinds of things like handheld game chip testing, documentation, VIC chip testing, peripheral chip design and customer development .” Yannes fell in love with the VIC chip but was surprised that hardly anyone was interested in using it. “MOS Technology was a merchant semiconductor house,” explains Yannes. “They were selling to the outside world, and here they had a chip they spent a lot of money developing. They were trying to get people

interested in it.” Unfortunately, most companies turned down the VIC chip. “There was one Japanese company that was using it in an arcade game,” says Yannes, referring to an arcade clone of Space Invaders. “It was really frustrating. I thought this was a great chip and hardly anyone seemed interested in using it. My job was to try to figure out how to get people interested in using the VIC chip.” Yannes created a video card for developers by adapting his school project. “I designed a board which I called the ACE board that would attach to either a KIM-1 or an S-100 bus to give you a video interface. That was going to be sold as kind of a development system to let people see what this thing was capable of.” “We were just looking for things to do and trying to get ways to sell the VIC chip to the OEM market,” says Charpentier. Yannes made it his personal mission to get the VIC chip into a marketable product. * * * In August 1979, approximately one year behind the Apple disk drive, Commodore started shipping the 2040 disk drive for $1,295.00. Cables cost another $80. As with the early PET computers, Tramiel preferred selling the costly disk drives on the European market. “We used to turn out tons of PETs and floppy disk drives, but they were all going to Europe,” says Russell. In the initial release, Commodore experienced a high failure rate from some of the early 2040 drives. “There were lots of times we had to solve drive problems, whether you were a drive guy or not,” explains Russell. Russell says the factory workers dumped non-working disk drives onto the engineers, hoping for a solution. “They wanted us to give them software that would actually make the drives so they wouldn’t fail,” says Russell. “At Scott Boulevard, we used to pile the 2040 dual drives up in the hallways until they reached the ceiling, and then you were like, ‘Okay, we better find a solution for whatever is wrong.’” Although the engineers were supposed to be developing new technology for Commodore, they often spent a large portion of their time ing the manufacturing facility at the back of the building. While the PET computers

underwent a burn-in, they ran a test program to all components were working according to their specifications. “We had a test that would run a burnin and do memory tests and a whole bunch of other stuff,” says Seiler. “There was something happening on the screen so the guys knew it was still alive and still working.” However, it took several minutes to load the test program into memory using the cassette. Seiler came up with a solution. “I built this thing I called it my no-op clip,” he says. “The 6502 was in a 40-pin dip package, and it had 20 pins on each side. This thing was like a big clothespin that would clip and grab all 40 pins at once. It had 40 wires that went into a little plastic box.” Technicians attached Seiler’s no-op clip to each PET motherboard that came off the assembly line. “Instead of coming up in BASIC, it would come up in my ROM and my ROM would copy the test program into RAM, and then you could take the clip off and it would run,” explains Seiler. “That’s how they loaded the burn-in test program.” If a PET computer failed the initial burn-in period, it was up to the technicians to fix the problem. “We had fairly good technicians. I really liked those guys,” says Seiler. “They were wave soldering these boards and there would be little solder shorts and it wouldn’t come up. They were fixing boards that didn’t work when they’d build them.” Although the technicians were slow and deliberate when they first started working at Commodore, they became remarkably proficient after troubleshooting a few hundred PETs. “There was a whole row of address busses on the bottom of the board,” explains Seiler. “I can one guy would take the scope probe and drag it across, ‘Click-click-click-click.’ He would find the bad one just by looking at how they danced across the scope.” Chuck Peddle noticed his engineers spending a lot of time helping with production problems. “I think Jack was happy for that, but Chuck wasn’t because nothing new was happening,” says Seiler. * * * Tramiel’s ruthless style earned him a reputation akin to a high-yield nuclear weapon in a business suit. A visit from Tramiel to a division of the company could easily leave behind a trail of devastation, often in the form of sudden

layoffs. As these visits became common, Commodore employees began calling them Jack Attacks. “The Jack Attack thing got to be routine,” says Peddle. “When Jack spotted people fucking up, he came at them hard, and he usually found people fucking up most of the time.” Despite Peddle’s disapproval of some of Tramiel’s methods, he ired his ability to unearth incompetence within Commodore. “Jack, for a long, long time, signed every check the company wrote,” explains Peddle. “The Jack Attack was instigated by Jack seeing something in a check he was supposed to sign that would piss him off and he would call people in and beat the shit out of them.” Peddle blames the frequent Jack Attacks on Tramiel’s policy of giving almost anyone a six week trial. “Jack would hire these truly incompetent guys and he would give them a chance to do something, and then they would go off and really screw something up,” explains Peddle. According to Kit Spencer, Tramiel tolerated mistakes as long as the employee did not repeat them. “Jack expected you to make mistakes. You had to because the business was moving so fast. As long as you acknowledged any mistakes, corrected them and made enough good decisions, it was okay.” Tramiel had a pragmatic view of his approach to management. “In business if people love you, you cannot be successful,” he says. “They’re paid to do things right. I’m paid to tell them what they’re doing wrong. But people don’t like to hear that. A lot of managers don’t like to confront what they’re doing wrong.”[2] Tramiel himself often made questionable decisions in Spencer’s eyes, such as antagonizing important distributors like ComputerLand. “I won’t go into those, but he could be a difficult personality at times,” he says. Spencer was one of the few executives who felt comfortable returning straight talk to Tramiel. “I saying to Jack when I was running UK operations, ‘Half of what I earn is for what I can do and implementing your good decisions, the other half is for not letting any of your bad decisions happen here.’ He laughed at that,” says Spencer. “Too many people were afraid to argue with Jack and he respected you if you did, as long as it was sensible and for the good of the company.”

In the US, few would be so open with Tramiel. “I don’t think there was so much of that type of interplay with Jack in the USA,” says Spencer. “People were more afraid of him and their jobs.” Jack Attacks occurred when managers made mistakes and attempted to disguise them. “It was fatal to try and hide any mistakes from him because that could lead to a loss of trust and that was important to him,” says Spencer. In many ways, it was better for Tramiel to berate an employee than to ignore him. “…my people know that when I stop telling them what they’re doing wrong it means they won’t be around much longer. I don’t waste my time on them,” he says.[3] Mike Canning was the head of MOS Technology in 1979. However, their latest product, the VIC chip, was not doing very well. Furthermore, MOS Technology was not keeping up with advancements in microprocessors with competitors like Motorola and Intel, and attempts to create their own RAM memory were unsuccessful. Canning was in a weak position. A manager named Jerone Guinn saw an opportunity to take over a more prestigious division of Commodore. “[Jerone] came in from Texas, originally to help Jack with the LCD business,” recalls Peddle. Unfortunately, Commodore’s calculator business was performing poorly. “He discovered that wasn’t a good place to be, so he retreated into taking over MOS Technology from [Mike Canning]. He was a great politician.” After a Jack Attack, Canning was out and Guinn was in. Robert Russell was amazed by the chaos of his first year at Commodore. “I was the mouse in the corner for a lot of interesting shit early on,” he recalls. “I came out in a time when there got to be a lot of turmoil as far as managers.” The constant rotating door of management at Commodore sometimes made things confusing for the engineers. “I don’t know how many managers I had,” says Russell. “I would go home on vacation and come back to have a different manager. I think there were almost ten managers in the very first year I was there that were supposedly my manager.” Perhaps most incredible to Russell was the lack of a chain of command at Commodore. With very little continuity from management, Tramiel micromanaged the engineers himself. “It just so happened that Jack came

directly to the engineers and impressed us with, ‘I want you guys to get the job done no matter who your boss is,’” says Russell. Russell has fits of laughter at the thought of the top person in the company giving orders to a junior engineer. “Yes sir! We knew who the boss was.” * * * Jack Tramiel had known wealth during his typewriter days, but his personal fortune had never been so high as it was at the end of the decade. When Commodore stock hit $48 per share (ignoring stock splits), his paper worth reached over $25 million. “You have to really hand it to Jack for taking something that was a watch and calculator company into a whole new damn industry,” says Dick Sanford. Although Tramiel lived a regular lifestyle, he indulged his love of expensive automobiles. “He had quite a car collection,” recalls Russell. “He had a flashy red 308-style Ferrari. Jack had some weird cars. He had things with expired Canadian plates sitting around and a junker that was probably a nice car at one point in time.” The pride of Tramiel’s collection was his V-12 Jaguar, which required expert maintenance. “He had the Jag that he had special mechanics to keep in tune,” explains Russell. Despite his wealth, Tramiel drove himself to work every day. During one visit to the Tramiel ranch, he offered Russell an unusual deal. “I was out iring his Ferrari one time and he said, ‘Don’t take salary for the next few years, and it’s yours,’” recalls Russell, who politely declined the offer. In contrast to Tramiel, Gould lived an expensive lifestyle. “Everybody was complimenting [Irving] on what a great job he was doing,” says Peddle. “He was having a lot of fun. He was doing it the way he wanted to do it. He engaged in a lot of vices all his life; smoking, drinking, women. He lived life to the fullest.” Peddle respected the company chairman. “Irving was an important part of my life and he always treated me fine,” he recalls. “He was always willing to sit down and talk to me. I learned a lot from Irving.” Irving split his time between New York, Toronto, and the Bahamas, where he owned a massive 70-foot yacht. “We’re talking about a Trimaran; a big, fucking

wonderful yacht,” says Peddle. “If he wanted to go somewhere in the yacht, he would tell the captain, ‘Go sail it into here and I’ll fly in and meet you.’” The frequent trips using commercial airlines became tiring for Irving. “Irving had an office in Toronto that he worked in part of the time and an office in New York. For a long time, he would travel back and forth with commercial airplanes,” says Peddle. With his reams of valuable Commodore shares, Gould could easily afford his own jet. However, he was not interested in selling his shares. Gould, the man who would not allocate the funds to expand the PET market into North America, insisted Commodore purchase a jet for his own use. Corporate benefits and status symbols were not something in Tramiel’s character. When he used commercial airlines to get around, he insisted on flying coach. He felt first-class tickets were a waste of company money. Gould had a different view. “Irving basically said, ‘My lifestyle needs a private jet,’” recalls Peddle. “Jack was actually looking at a turbo-prop, but Irving wanted a real jet.” One consolation for Tramiel was that a corporate jet would essentially become part of Commodore’s vertical integration strategy. “Jack flittered around the world continuously,” says Dick Sanford. Employee travel was as necessary to Commodore as silicon chips. With a corporate jet, it would be like owning an airline service and exchanging money internally, rather than handing resources over to airlines. Eventually, Commodore purchased a seven-seat Westwind IAI-1121, making them the first microcomputer company to own a corporate jet.[4] In recognition of the product that made it possible, Gould and Tramiel dubbed it the PET Jet, and even had the name painted on the fuselage. A Commodore “chicken-head” logo adorned the tail, above aircraft number N404. The company’s blue and red colors ran the length of the aircraft.

Commodore president Jim Finke boarding the PET Jet.

Gould made sure IAI ostentatiously decorated the interior of the jet. “That was Jack’s private jet and it was the gaudiest,” says Seiler. “We had gold plated faucets in the bathroom and real plush seats. It was really nice, but kind of a little bit too much.” As Tramiel would find out, the jet was also useful for holding meetings while on the move. In addition, it impressed potential new hires with the promise of occasional flights, and it helped retain old employees. “He flew all the engineers to Vegas, to the CES show on that,” says Seiler, referring to the 500-mile trip. “He shuttled them all over there and back.” “Jack flew whenever he could,” recalls Peddle, although Gould dominated its use. “It used to piss Jack off, because Jack knew he built the company; he made all this money, he bought this jet and everything else, and ninety percent of the time Irving was using it to go on this lifestyle that he had.” * * * At 48 years of age, Jack Tramiel was no longer the compact, muscular scrapper of his youth. He was now a rotund, balding, middle-aged man with wild grey hair. However, the ion that drove him showed no signs of diminishing. Tramiel had spent a lot of time away from his three boys, mostly because of the frequent travels he endured. “Jack comes out of the camps, comes to the United States, gets something going, gets the company started, loses the company, then builds it back up again,” recounts Peddle. “His whole life was that company. He kind of sacrificed his relationship with his kids. Other than gambling and some other things, Jack worked.” His sons were now adults and Tramiel felt a strong need to bring them into the Commodore business. He even shared his dream with some of his most trusted executives in the early eighties. He felt so strongly about his sons, he even considered placing a clause in his will to make it more profitable for his sons to work together.

Gary, the baby of the family, was the moneyman. In the early eighties, Gary Tramiel was an undergraduate studying for a career in finance. Incredibly, at this early stage in his life, he managed the Tramiel family’s multi-million dollar investment portfolio. After graduating, Gary became a stockbroker for Merrill Lynch. Leonard, the middle child, was the scientist and intellectual of the family. “Leonard was different,” says Charles Winterble, an engineering manager at Commodore. “Leonard and Sam were like night and day. Leonard was a software guy and didn’t really have a lot of interest in running the company.”

Leonard Tramiel demonstrating the PET at a computer show.

For the moment, Leonard was out of the picture completing his PhD at Columbia University in New York. “He would disappear and go off back to college, and come back during the summers and hang around with us guys,” says Seiler. “He didn’t like the business part of things.” The oldest child in the family, Sam, was now 32. While living in Canada, Sam received economics training at York University. He was the businessman of the family, who apprenticed under his father. Of Tramiel’s three sons, only Sam was involved in the day-to-day operations at Commodore. Al Charpentier liked the Tramiel sons, but felt they might not have the same potential as Tramiel. “None of the Tramiel sons were like their father,” he says. “They all had reasonable skills and they weren’t bad guys but Jack had that drive and desire to succeed.” For the time being, Tramiel continued mentoring his sons and guiding them towards a career at Commodore. Perhaps someday, they would learn enough from their father to run the company. In the late seventies, Sam incensed Jack and Helen with his actions following the death of his young wife. “Sam started fooling around quicker than they thought he should,” recalls Peddle. “He was a healthy young man, and he had been effectively cooped up with a wife who was dying for some time, so he went out and started partying.” His parents felt Sam had not allowed enough time before moving on with his life. “They got pissed and threw him away for a while,” reveals Peddle. “Jack got mad at Sam and sent him off to Hong Kong.” The engineers found Hong Kong impressive. “Hong Kong is an amazing spot where lots of capitalists and free market stuff just rages,” says Bill Seiler. “I like the Hong Kong Asians a lot. They’re pretty hip and they understand American humor a lot better than the mainland Chinese. They don’t get our jokes.” In 1978, the Chinese government began a transition from a planned economy to

a market economy. Commodore began to look into the Special Economic Zones set up by People’s Republic of China. “They’d do some goofy stuff,” recalls Seiler. “I being at one plant somewhere in Hainan, they turned off the power for three hours one day! The government just turned it off.” The poor infrastructure in China amazed Seiler. “If that was a private power company, they are quick to get the power back on because they are losing money if it’s not pouring into somebody’s homes,” he says. “Whereas the government, they don’t care. They just turn it off to fix something anywhere and shut down the whole economic sector for three hours.” The unreliable nature of business in China made it too risky for Commodore to enter into deals with mainland companies at the time, although the country would gradually embrace free markets over time. * * * By 1979, computer dealers were becoming a major force in the industry. CES was mainly for retailers like Kmart and Sears, and the computer dealers needed their own show to focus solely on computers. In November 1979, the very first COMDEX (Computer Dealer’s Exhibition) opened in Las Vegas at the MGM Grand Hotel and Casino. In total, 167 exhibitors showed up, including Apple, Atari, Hewlett-Packard, Texas Instruments, and Commodore. Almost 4000 people attended the first show. Apple was expanding, and used the show to woo employees from other companies. “I a discussion when Apple was starting to be fairly successful and they were starting to work on what became the Lisa,” recalls Spencer. “They were hiring every technical guru around, from top university professors and from leading companies like Hewlett-Packard.” Spencer was talking with Tramiel and Peddle about the Lisa computer. Peddle had insider knowledge of Apple from his brief stint working for them, and he told the group, “They are hiring everybody. You know what’s going to happen? Everybody will argue to have every possible feature in there because they all have their own hobbyhorse. They’re going to build something fantastic, but will be so costly it will never work in the market.” It was a prescient remark. Software vendors also attended COMDEX. One particular application, VisiCalc by Personal Software, stole the show. Although it would be a few more years

until Apple could legitimately claim to have the number one selling computer, VisiCalc would begin raising Apple to new heights. Unsurprising to Tramiel, Atari and Texas Instruments had missed most of the Christmas season with their new computers. In November 1979, Sears began selling the Atari 400 and 800 home computers. Texas Instruments also released the TI-99/4. As Tramiel predicted, TI could only produce them in limited quantities. Radio Shack was the undisputed leader in sales throughout the seventies. They refined their design and delivered the TRS-80 Model II on May 30, 1979, which was state of the art for the time. By the end of 1979, Radio Shack had sold over 150,000 TRS-80 computers, giving them approximately 35% of the personal computer market. With new peripherals like disk drives and printers, s could spend as much as $4,000 for a complete system. In 1979, its computer equipment sales were over $100 million according to Tandy, or approximately 12.7 percent of Radio Shack’s North American sales. It seemed like Radio Shack was unstoppable. Commodore also continued their remarkable climb. At the Comdex Show, Tramiel announced the company’s projected financial results. Curiously, he claimed the PET Jet expenses under research and development. “I the year that he bought that Jet. At COMDEX, he said, ‘I spent $4 million on R & D.’ Yeah, he bought himself a jet,” laughs Seiler. Dick Sanford spent time in New York promoting Commodore to investors and ensuring the company’s long-range plans were fully articulated. “I dealt with all the banking issues, ing the Wall Street issues, and ultimately serving Mr. Gould and Jack,” says Sanford. “The PET was taking off and the organization needed financing and so my role was developing a corporate finance team.” To promote Commodore as a company and ensure the share price continued its upward climb, Gould organized financial analyst meetings in New York. “When we had analyst meetings, which were quite common, you would be doing presentations to a number of financial analysts,” says Spencer. “Irving would do those and he would ask me to go along as the marketing guy at the meeting. He’d usually have a technical guy, Chuck or Charlie [Winterble] or someone like that, and a financial guy so you could cover all aspects of the company. I probably spent more time with Irving than most on that.”

The year was a resounding financial success for Commodore. They had total sales of $71.1 million, with all of the growth coming from the computer division. Profits were $6.5 million. However, the financial world knew it was on a rapidly growing field, and the stock price rose from 5 1/2 to 48 7/8 before a 3-for-2 split. Tramiel had conquered Europe, temporarily lost ground to Radio Shack, and matched Apple. As the new decade began, he would look for ways to dominate the North American personal computer market. [1]Compute! magazine, “New Life For The 6502?” (February 1986), p. 26. [2] The Sydney Morning Herald, March 20, 1989. [3] Ibid. [4] Israel Aircraft Industries manufactured the Westwind IAI-1121 jet. Coincidentally, Aero Commander dubbed it the Commodore Jet in 1963 when it began work on the project.

CHAPTER 14

Color Computers 1979 - 1980

Throughout late 1979, Bill Seiler continued improving the PET line of computers. “I worked on the 80-column display,” he says. “That was the business version of the PET with the large screen.” Commodore branded the computer the 8032, which contained 32-kilobytes of memory. MOS Technology manufactured the video chip, called the 6545. “By that time, there was a display processor out that we could use and we didn’t have to do it all with discrete-logic chips,” says Seiler. “It was based on a [Motorola] 6845 chip.” Commodore would display the computers at the January 1980 CES in Las Vegas, years before Apple released an 80-column computer. Although the PET had the Apple II beat in many areas, it conspicuously lacked color. “We had pretty much drawn a barrier for Apple at the higher end stuff in Europe,” recalls Chuck Peddle. “We were doing very well, but the Americans were buying the color story. Apple had color and we didn’t. We were getting ready to do color.” Peddle and his team began focusing on a next generation product to leapfrog Apple, Atari, Tandy and Texas Instruments. “As soon as we had the printer and other stuff done for the Commodore PET, we started focusing on what’s next,” says Peddle. Bill Seiler, the loner Peddle befriended in Florida, took the development lead on the new project. “Seiler had some interesting, clever thoughts in him,” says Peddle. “He was really good with kids. He was really good with games. He had a good understanding of video.”

Much of Seiler’s knowledge of computer graphics came from his attendance at computer graphics conferences. “Seiler was into this show called Siggraph, which is about graphics,” says Peddle. Siggraph, a gathering of computer s searching for ways to improve computer graphics, held the first conference in 1973. “Seiler had been going to Siggraph since the time it started. He was a true pioneer. He walked the walk and talked the talk.” It was hard to ignore the growing influence of computer games. “Computing was going a couple of places,” says Peddle. “It was going to good games, because at this point in time Atari is a very successful gaming company. It was obvious graphics were going to go someplace they hadn’t been. Color was going to be very important in that market.” The Apple II, along with VisiCalc, began gaining momentum in late 1979, and Jack Tramiel took note. “The instructions from Jack were that he wanted a product to compete with Apple,” recalls Peddle. “It was intended to be an Apple killer.” Bill Seiler describes the project goals. “The basic thing was something we could throw out there that would deal with all the issues about the Apple II. The Apple had color, we didn’t really have that. It had bitmapped graphics, we didn’t have that. It had better sound than we could do.” The engineers were free to experiment in many directions. “It was a pretty loose project as far as I ,” says Seiler. The new computer would require a color video chip. The 6560 VIC chip was not good enough for the TOI because it could only produce 22-columns of text. “We wanted to be competitive against the Apple II, because Apple had a 40-column color screen,” says Seiler. The 6560 VIC chip was the first color chip by Commodore. The next chip to reach the public was the 6566 VIC-II chip, which appeared in 1982. In between the 6560 and the 6566 were some curious gaps. Unknown to the public, Commodore made two failed attempts at more advanced color chips before a usable video chip finally emerged. Tramiel wanted Peddle to show his Apple killer at the January 1980 CES. “Jack really wanted to do a 40-character color computer,” explains Bob Yannes. “We began working on a 40-column version of the VIC chip, the 6562, in late 1979

when Jack told his group he wanted to show a color PET at the January 1980 CES. He was always coming to people a few weeks before a show and telling them he wanted something new!” Seiler dubbed his game machine the TOI. “I called it TOI because it sounded Japanese,” he says. “That was the internal project name.” Seiler intended the name to convey a sense of playfulness, like a toy. Meanwhile, Peddle gave meaning to the TOI acronym, calling it “The Other Intellect.” “The TOI became focused around a low-end machine,” says Peddle. “The little machine was all focused around the best graphics game-playing computer you could build.” Seiler began working with the MOS Technology engineers, who resided on the east coast, to develop an improved color graphics chip. “We wanted a good, lowcost color chip,” says Peddle. “Effectively MOS Technology were being sponsored by us to get their chips done. One group had the role of defining a new next generation computer system, and the other guys had the role of building the chips for it.” Al Charpentier wanted to use the work he had done on the VIC chip to fast track development. “The first approach was pretty much just a fancy VIC chip,” says Seiler. “We just took the VIC and tried to stack up two of them together some way inside the chips so they could do 40-columns.” Charpentier recalls the origins of the 40-column computer system, which he refers to as the VIC-40. “The VIC-40 was originally proposed to Jack as new video game project that would require the 6562 chip,” says Charpentier. “Jack read the proposal in ‘79 and blessed the project.” With the deadline less than a month away, Charpentier decided to modify his 6560 VIC chip to produce 40-columns. “The only thing we had time to do was take the VIC-I chip and double its speed so we could do twice as many characters,” explains Yannes. Doubling the speed of the VIC-chip was not easy, especially with the slow memory that existed in 1979. “That created some real timing problems because —this is kind of a laugh today when you think about it—the fastest memory chips you could get were 450-nanosecond access times, and this video chip would have required 200-nanosecond parts,” explains Yannes.

Charpentier gave Yannes the task of deg the 6562 circuitry. Though new, he was a quick study. “The 6562 was the first chip design work I had ever done,” he recalls. “I ed Commodore right out of college in June of 1979, and although trained as an electrical engineer, I had no training or experience in chip design. I thinking it wasn’t such a great idea to put me on a project that needed to be done quickly and correctly the first time.” MOS Technology was Commodore’s ultimate weapon. “We had to redesign it, simulate it, get it laid out, and get it through the front-end processing to get functional chips out in time for the show,” recalls Yannes. “If Commodore had not had MOS Technology as an in-house resource, there was no way [to get it done]. There is no chip manufacturer that would jump through hoops like that to make prototypes for a show.” Normally, it takes over a year from the initial conception of a new chip to the time it exists in silicon. Since Charpentier and Yannes started with the original VIC design, it only took them a week to produce the new layout and another week for the MOS technicians to deliver a working chip. “We had a turnaround time on a custom chip of two weeks, which is absolutely unheard of,” says Yannes. With less than a week before the CES show, Bill Seiler inserted the new chip into a modified PET motherboard. Meanwhile, Robert Russell changed the ROM code to allow the PET to work with the new chip. Externally, the prototype resembled a regular PET computer. “It was actually in a PET case with a color screen,” says Russell. Despite the rushed effort, Yannes made surprisingly few errors. “It turns out we had only one layout bug in the chip on the first , and we corrected it and got new wafers processed through the MOS fabrication in one week,” he says. Seiler received a new working chip just prior to the show. “I was trying to get this chip to work and they were giving me new prototype versions of it,” he says. * * * Just after the start of the new decade, companies gathered in Las Vegas to display their latest technology at the January 1980 CES Show. “You always do your R & D stuff for the January show,” explains Peddle. “The January CES show was the place where everybody goes to get ideas for what might happen

that year. Then we go to the June CES show to sign orders for what they are going to buy that year.” Commodore unveiled their prototype TOI system using the 6562 chip. “I believe what was shown at CES was literally a PET with a color display,” recalls Yannes. “I recall helping John Feagans with some of the music stuff for a demo.” The TOI demonstrated the sound and 40-column color text of the 6562 chip. Although the VIC-based computer was working, Peddle views the demonstration as something of a deception on their part. “The thing in January was just a demo. We were trying to make something that was going to be potentially real, but it was strictly all fake. You can do a lot of things if you put stuff on top of tables, and under the bench you have a lot of crap. That’s part of the business.” Yash Terakura sensed the TOI did not impress Commodore management. “At that point there wasn’t even a discussion whether they wanted to develop the computer in the US or not. It was too early at that stage.” Along with prototypes of the 6562-based TOI, the engineers released brochures to describe the specs. Byte magazine described their impressions of the CES demonstration:

Heading the list was the prototype of the TOI (“The Other Intellect”) color computer. Aimed at the low-end market, the TOI is designed to interface with your home color television set. The displayed image will feature 16 colors, 160 by 192 resolution (with three colors in the high-resolution mode), Microsoft BASIC, and a standard keyboard. The price could be under $700.[1]

Although the 6562 chip was good enough to demonstrate a color PET at CES, it was unsuitable for a mass-market personal computer. “We got through the show, but when Jack found out that it required 200-nanosecond RAM, he freaked out because that’s very expensive,” recalls Yannes. “He said we’ve got to change that.” The MOS Technology engineers tried to make something work, but it would be impossible to make the 6562 operate without costly static RAM (SRAM). MOS

Technology effectively gave up on Tramiel’s request for the 6562. “It was just too hard to get anything working well,” says Yannes.

Dennis Barnhart (Commodore’s VP of marketing) at the January 1980 CES.

Tramiel wanted a second opinion, so he packed up one of the TOI prototypes and brought it to Japan. He wanted the Japanese engineers to review the design and perhaps find out if they had an alternate solution to avoid using the costly SRAM it relied on. Tramiel’s actions rankled some of the American engineers. “You actually had competing engineering groups in Japan, Pennsylvania, and California,” explains Charpentier. “Literally, Jack would play us off of each other. He would send my work to Japan to have them critique it and have me critique their work. That’s not a bad thing to do, but he would do it in a way that set us up as adversaries.” When the Japanese engineers confirmed the SRAM problem, Tramiel allowed the 6562 project to die. “I think the product just kind of went away, because it was clear it needed to be designed from scratch and not pieced together from existing junk,” says Yannes. * * * Despite Tramiel’s rejection of the 6562-based TOI, due to the costly 200nanosecond RAM, he was still firmly on a mission to release a computer that would dominate North America. He began aggressively pursuing the goal, no matter what obstacles stood in his way. “In about 1980, Jack reached the decision that he was just about ready to come back and attack the US market,” recalls his assistant, Michael Tomczyk. Tramiel wanted a working TOI computer to market at the June 1980 CES show. Peddle began pushing the MOS Technology engineers harder for a working video chip. Failure was not an option around Tramiel, and Peddle knew it. He tried his best to make something happen quickly. “We started putting some serious pressure on the guys to get a better chip,” says Peddle.

Charpentier and Yannes began work on a new chip, which did not rely on highspeed RAM. Once again, Bill Seiler was involved in the specifications for the new chip, while MOS Technology handled the circuit design. “They were doing all the design of it, how it would work inside,” says Seiler. Seiler pushed for dynamic RAM (DRAM) rather than the more expensive SRAM. “Chuck Peddle’s group wanted to use DRAM to get more memory in less chips,” recalls Yannes. “We had to come up with a video chip that would do 40-character lines but would work with slower memory and refresh the memory.” Since they were starting again, the engineers gave the chip a new designation. “That’s what necessitated the 6564/65 chip, which was a 6562 redesigned to work with slower DRAM memory,” explains Yannes. The 6564 was a moderate improvement on the VIC chip. “Resolution was 320 x 200 with the same colors as the VIC 6560. In all cases, we were talking about a 40-character line.” As before, Charpentier created candidates and sent them to Seiler to use in his prototype computer. “He’d send me some chip and we’d try it and there was something wrong, and we’d go back and forth,” says Seiler. Seiler and the engineers worked up a number of prototypes of the TOI with different specifications. Some were Apple-like, others Atari-like. “We did all kinds of prototypes,” recalls Peddle. Robert Russell saw early plans for the computer. “It was like the Apple II and Atari 800, which were out at that time and doing quite well,” he says. “The whole idea was to give more options and expansion, and a more integrated package. It was a bigger box with slots.” Tramiel, hoping to spur the project forward, had his Japanese engineers create a prototype case for his Apple killer. “I had a box that somebody built that looked like an Apple II box and I built something into it,” says Seiler. “The Apple II had a keyboard up front and a door in the top. It was very similar to that. I think it was intended to be vacuum molded plastic.” Rather than a built-in monitor, like the first TOI, the new TOI prototypes used a household television. “The TOI was taking the concept of a color PET and [building it] around using a color television,” says Russell. “It was more a forerunner to the Commodore 64 as far as its ambitions.”

The west coast engineers were limited in what they could work on while MOS Technology developed the 6564 chip. “Once we concluded we needed a highend graphics controller, we were at the mercy of MOS Technology,” says Peddle. In case Charpentier failed to come through with a chip, Peddle’s engineers explored alternate plans. “We also started talking about how we could come to market without the full implementation,” says Peddle. “In other words, there were some things we could do to effectively work around the fact that we didn’t have a chip.” Bill Seiler began deg a video card, mostly at home, using full-sized transistors and other electronic components. “It was a 40-column color PET done with discrete logic rather than a devoted graphics chip,” says Feagans, who worked on the TOI software. * * * After completing printers and disk drives for Commodore, Peddle had an opportunity to take a small step towards changing the simple computer systems division at Commodore into a technology company similar to Hewlett-Packard. “We had some good gains and Commodore was doing pretty well,” he recalls. “We were trying to expand the company at that point.” Unfortunately, his new engineering projects within Commodore were progressing unusually slowly. Peddle began to feel pressure from Tramiel. “He was getting beat up by Jack because all these TOI projects weren’t getting done,” says Bill Seiler. Seiler had strong memories of the drudgery involved with fixing production problems. “There were so many problems showing up on the production line. People would come right up to the engineering and say, ‘This isn’t working.’ So we’d have to go back to production and figure out what was going on and help them. We weren’t getting any projects done because we were doing so much production . We were being too distracted.” Although getting the first production line up for the original PET 2001 had been exciting, it no longer held the same novelty. “The humdrum of normal engineering is when you’re doing paperwork and all that junk, and it can get boring,” says Seiler. “You get down to pushing it into production and you’re fighting with bills of materials and vendors.”

Peddle surmised that the reason his disk drive development had progressed so slowly was that his engineers were constantly being interrupted by manufacturing problems on the assembly line. He attempted to convince Tramiel to hire dedicated engineers for the job. “We realized Jack wouldn’t bite the bullet and hire engineers for production,” says Seiler. “The idea behind Moorpark was to get away from the day to day factory stuff,” says Peddle. “During that time we came up with the idea of splitting the group up and leaving guys like Russell to hand-hold the factory technicians while we were off deg what was going to happen next.” Peddle decided to increase the distance between the production line and his product development engineers. “Chuck had this kind of a coup,” says Seiler. “He moved a bunch of guys over to Moorpark and the production guys would have to fend for themselves more because we were further away. I think that was kind of a political move by Chuck to get the new projects going again.” By allowing his engineers to concentrate on future products, companies like Apple would have less chance of beating them to market with new products. Peddle planned to emulate Hewlett-Packard by creating several research and development labs, all operating out of separate buildings far from Commodore headquarters. Surprisingly, he met with no resistance from Tramiel. “These were nothing expenses,” says Peddle. “If anything, Jack used to get shit from Irving because we didn’t show up good on analysts’ screens. They said we weren’t spending enough on engineering.” Commodore even resorted to artificially pumping up the number of engineers on the stock reports. “We used to throw everybody and their dog into engineering just to try to get their numbers up,” reveals Peddle. By separating the research centers, Peddle felt each could focus on a specialty. “Chuck was out spawning these little things because he knew he needed a floppy, he needed a hard drive, he needed printers,” says Seiler. “He was juggling all these things up in the air, trying to keep all these programs going with almost no budget from Jack.” Peddle also developed a research center using former acquaintances from his GE days. “I knew a bunch of guys in Phoenix who were really smart guys, working

out of this company called [Monolithic Systems],” explains Peddle. “They had spun out of Courier because Courier had been bought and sold, and they were not happy. So a friend of mine [Larry Hittle] offered to set up a design center, similar to the one Bill had [WDC], which would be owned by Commodore.” Peddle wanted his Phoenix team to focus on high-end business systems. “It would be an advanced technology center for Commodore,” explains Peddle. The Phoenix team would focus on improving Commodore disk drives, with Peddle consulting on their projects. After the poorly received 2040 dual-floppy disk drive, the engineers began work on a single-floppy drive, which they initially called the 2030 and then later released as the 2031. The Phoenix lab even developed a dual floppy drive using larger eight-inch diskettes, which stored more data. “The 4060 had two eight inch floppy drives,” says Feagans. “The 4060 was the most difficult thing I ever worked on because the DRAM refresh was done in software. If that wasn’t working, there wasn’t much of a clue as to what was going on.” Feagans often resorted to monitoring the serial line using a logic analyzer to determine what was going on with the drive. To gain a lead over the Apple disk drive technology, Peddle introduced them to an innovation he developed back in 1961. “We took my patent, which had run out, and used it to double up the code,” says Peddle, referring to the amount of data that could be stored on a single disk. The team also began looking at data storage devices to handle mainframe data from the business world. Peddle wanted an eight inch floppy disk drive that would be compatible with the IBM 3740 disk format, as well as Peddle’s efficient GCR format. Peddle also set up a California lab that would become the crown jewel of Commodore’s research program. Here, engineers would develop secret weapons that would give Commodore long-term technological superiority over their rivals. He was creating his own playground for electronics. Peddle loved the vibrant microcomputer scene in Silicon Valley, so he wanted a research center to take advantage of the synergy with other companies. He set up the research center at 2996 Moorpark in San Jose. Commodore rented the second

floor from a non-profit organization. “It was owned by the American Heart Association, who occupied the lower floor and we were on the second floor,” says Feagans. Bill Seiler changed apartments to minimize his commute to work. “I moved out of Palo Alto to San Jose about the time we moved to Moorpark,” he says.

Creating prototypes at the Moorpark research lab.

The Moorpark lab was just over ten minutes away from Commodore headquarters in Santa Clara, California, and across the street from the bizarre 160-room Winchester Mystery House. “It was a funky building! We had shag carpet in an engineering building,” laughs Seiler, referring to the deep-pile carpet that was popular in the 1970s and notoriously difficult to clean. Engineers often dropped small electronic components, which the shag carpet immediately swallowed up. The research center was modest by most standards. “It was a second story office building and not a big office building,” says Russell. “It was all broken into offices or labs. They had a hardware lab, a software lab, and they had a conference room.” The engineers at the Moorpark facility were all young, recent hires who had proven their abilities with the PET. “They took away all the first rank engineers,” says Russell. “It was kind of a way to give all the senior guys their own office, because we couldn’t really do that in the other building.” The team included Bill Seiler, John Feagans, Glen Stark, Scott Patterson, and recruits Peddle would hire in the following months. Robert Russell was only recently hired and too inexperienced to . Moorpark quickly grew as more engineers started moving into the premises in order to concentrate on engineering projects. Shiraz Shivji moved into Moorpark to focus on Commodore’s scientific calculator. “Shiraz was doing something with that weird 4-bit calculator thing with Mensch,” says Seiler. Working with Peddle, Shivji programmed the ROM chip for the calculator. “The software was created by Shiraz Shivji, who worked for Chuck Peddle at the time,” says Bill Mensch. The engineers remaining at 3330 Scott Boulevard no longer fell under Peddle’s command. “Chuck was kind of head of Moorpark, and he may have had the title of head of engineering, but he never came and told us what to do in our building,” explains Russell. “It was much more likely that Jack came in and said these things needed to be done.”

According to Peddle, the research cost Commodore very little. “We didn’t spend a lot of money on things. We had a discretionary budget where we could get toys. We had to read books from that time and we had to buy some competitors’ equipment. It was a well-funded development center by Commodore standards, but certainly nothing very astounding; ten people or something like that. We were using people’s minds rather than Jack’s money.” * * * With the new goal of making a business computer with disk drives and printers, the team began to look harder at how the PET would those types of devices. Scott Patterson, a perfectionist, argued for a logically consistent architecture. “He was the guy who was trying to drive us into doing pure things, like the kernel and stuff like that,” says Peddle. It was up to John Feagans, Moorpark’s software engineer, to improve the operating system. “By the time we got to Moorpark, we were all finally fed up with the way we had been writing code,” says Seiler. “Feagans was trying to write some kind of kernel.” “He wanted to get out of BASIC because the original code inside the PET was just all spaghetti,” says Russell. “John said, ‘Let’s break up all this stuff that deals with the hardware and get that into its own section of the code,’ rather than here’s a routine that runs the cassette and here’s a routine that runs the keyboard and here’s a routine that runs the display.” Raymond John Feagans, who preferred using his middle name, was born on the west coast in Oregon. Since an early age, he had an affinity for electronics and creativity. At the age of eleven, Feagans tried to build his own electronic organ. While still in high school, Feagans published a pictorial book on trains called, The Railroad that Ran by the Tide. He also built his own high-quality furniture and brewed his own beer and wine. Now he was one of Commodore’s key developers. “John is schoolmasterly,” says Peddle. “He’s the sort of guy you would think of as a high school science teacher. He’s kind of tall, kind of slim, very mildmannered, and wears glasses. He has a studied, quiet voice.” Although the hardware engineers are more often glorified, software engineers like Feagans were equally important in making the system work properly. “He

was as close to a software engineering manager as we had,” recalls Peddle. “He really formalized that position.” Compared to Peddle, Feagans was introverted and quiet. “He was very private. He didn’t socialize very heavily. Generally, you could trust him to get done whatever he said he could commit to do. He was a very serious guy.” Bil Herd, one of Commodore’s most extroverted engineers, concurs. “He was regarded as something of a kook, but I think he always got his part of the job done.” One of Feagans’ closest friends at Commodore was his fellow programmer, Robert Russell. “He went to College at Iowa State University,” recalls Russell. “That’s where I ran across him because he used to argue for the 6502 and I would tell him he was full of shit.” When Russell was new to Commodore, he shared a house with Feagans and the two became a sort of odd couple. “John was kind of tough to get along with at points in time,” says Russell. “He was kind of a perfectionist—a clean guy. When I lived with him, I cooked sauerkraut in his pans. He freaked out because it boiled over, and it was going to etch the copper. We ed each other and remained friends, even though we had our differences about how things were to be done.” At Commodore, Feagans pushed for better architecture in software and wrote much of the early code for the PET computers. By the time Russell ed Commodore, Feagans ed much of the low-level coding to him. “He was an architectural-type of guy, not a down and dirty guy,” says Russell. Although a quiet engineer, Feagans actually worked his way into Tramiel’s inner circle. “Feagans had some kind of relationship that was around Jack and his family,” says Seiler. “Feagans came off as a big puffball to Jack because he never really fought him on anything. He had different politics of how to deal with the Tramiels. He worked his way around them by just being kind of nice and putting up a front with them. But he usually got done what he wanted to get done.” Feagans was also instrumental in creating tools for his computer programmers. “He wrote the compiler we used,” says Bil Herd. “You have to , during this time, they’re writing the code on PETs, and it would take overnight to

compile. There were no Vaxes at Commodore at that time.” Feagans discussions with Patterson led them to implement something called a kernel. “John Feagans was the architect of what became the kernel concept, which was a separate operating kernel versus the BASIC interpreter,” explains Russell. The new PET computer would differ from the HP 2000 style of BASIC interpreter on the PET 2001. The word kernel is a synonym for nucleus, as in “the kernel of truth.” In computers, the kernel is essentially the nucleus of a computer operating system that provides basic services for all the other parts of the computer. “It was similar to the BIOS (Basic Input Output System) that is in all the PCs now,” says Seiler. “It would have entry points that were standardized and we wouldn’t have to keep telling everybody how to recompile when we changed the BIOS. The BIOS allowed you to have entry points for if you want to print a character or you want to read a character or if you want to write a character.” These services include things like getting a character from the keyboard when the hits a key, loading and saving memory from external devices, and moving the cursor position on screen. Before Patterson, the hardware engineers gave the operating system very little thought. They tended to combine the kernel, the shell (the actual commands to access the kernel, such as LOAD, SAVE, and PRINT) and the programming language into one mixed-up creation. “In the early PETs, there wasn’t anything like an operating system,” says Seiler. “It was kind of piled around the BASIC that came from Micro-Soft.” Feagans was on a quest to separate these parts of the operating system into modular pieces. * * * At first, Tramiel appreciated the relative isolation of the research centers. He believed marketing people slowed down the design process and exasperated engineers, and he even forbade his own marketing people from visiting the Moorpark labs. This provided an ideal, free setting for Commodore engineers to develop elegant new technologies. The charter for Moorpark was clear. “The idea behind Moorpark was to be on

top of every one of the technologies that were starting to explode,” explains Peddle. “To understand them and decide how we were going to integrate them into the next products we were doing.” Peddle expected Moorpark to deliver consumer products for the coming year, while the Phoenix Research Center developed technologies with no definite goal. “[Moorpark] was both a technology center and an advanced development center, whereas in Phoenix they were just an advanced development center,” he says. Without the distractions caused by production problems, development on projects speeded up dramatically. “Those sorts of guys, once they get involved with a project, that’s it—they’re going crazy. The rest of the world stops,” says Spencer. Moorpark had up to a dozen different projects developing at the same time. After the Phoenix lab completed a prototype of the single disk drive, they sent it to Moorpark. “Glen Stark was working on the single floppy, with Scott Patterson,” says Seiler. “It was being worked on pretty hard at Moorpark.” John Feagans programmed the disk drive ROMs. Commodore needed to release a cheaper single floppy drive quickly due to customer demand. Previous floppy drives from Commodore used a drive mechanism from Seagate, the company formerly known as Shugart Technology. Unfortunately, the die casting parts in the mechanism made the drive very costly. “Seagate was making a floppy drive but it was too expensive because the floppy drives at that time were cast metal,” says Seiler. Commodore needed a cheaper drive mechanism if they wanted to release a lowcost disk drive alongside the TOI. They turned to a company called Alps, who used a cheaper method of production. “Some company came up with a real inexpensive drive mechanism,” says Seiler. “We found somebody that made stamped metal pieces.” When the engineers tried the mechanism, it did not perform as well as the die casting drive mechanism. “That crazy stamped floppy mechanism, the way the head moved, it didn’t have a motor,” says Seiler. “It had a ratchet stepper thing and it was really crude. So it made a lot of noise when it stepped from track to track. Eventually the ratchet would leave little metal filings all over your

floppy.” A difficult decision between quality and cost faced the engineers. Eventually they decided to go back to Seagate parts until Alps perfected the stamped metal mechanism. “We tried to use that initially but the first ones eventually used a more expensive drive,” says Seiler. Peddle knew many of the projects under development had no immediate potential as consumer products. “Moorpark had somebody working on everything a computer could do,” he recalls. “We were doing voice in/voice out. We were doing voice recognition. We were doing all kinds of communications stuff. We actually did some primitive networking.” Some of the technology seemed exotic for the early eighties, including a working prototype of a touch screen. “We were using LEDs along the sides [of the monitor],” says Peddle. Photoresistors on the other side of the monitor detected light levels, allowing the software to identify where someone touched the screen with their finger. Incredibly, Moorpark developed digital camera technology similar to a webcam. “We were trying to find what you do with computers,” says Peddle. “One of the things we wanted to do was have a computer where you can call up people and talk to them, and they can see you and you can see them.” In the late seventies, the digital camera revolution had not yet begun, so Moorpark had to pioneer the work using a technology called CCD (ChargeCoupled Device). A CCD is an array of light receptors in a grid. When a photon of light hits one of the receptors, it releases an electron. The CCD device has a grid of pixels, almost like a monitor, except the CCD pixels receive light instead of emitting light. The CCD technology was a holdover from Peddle’s tenure at Apple. “Chuck was not around in the fall of 1978 and didn’t come back until February 1979,” says Feagans. While he was away, a newly founded DRAM maker from Boise, Idaho dropped by Commodore and showed off CCD. “In that period, Micron Technology visited us with a DRAM-without-a-top and a demo system that hooked up to a PET. They could take a picture with their demo system.” Feagans calls the CCD a “DRAM-without-a-top” because the CCD was essentially a matrix of photosensitive RAM, with every byte in the matrix

representing a pixel. By focusing an image onto the DRAM surface using a lens, it was possible to read the image into a computer one byte at a time. “I immediately wanted to make a digital camera, but it was just too far ahead of its time,” says Feagans. Commodore’s management did not pursue any products using the technology. “We had some unimaginative managers at that time who could not see the potential.” “Basically we had one guy who was just totally in love with CCD and what you could do with vision, and that was his charter,” recalls Peddle. “He was a very weird guy but he was making that happen. I think they were the first people to show a CCD camera on a personal computer.” Peddle wanted to use the CCD camera to revolutionize personal communications. In the pre-Internet era, phone lines were the only publicly available infrastructure capable of transmitting data across the world. The team researched modems for data transmission. In many ways, the research team was too early with the technology. “Chuck was usually aware of what could potentially happen but not always practical on the timing of when it was commercially possible,” says Spencer. “Chuck was brilliant but he tended to be a bit ahead of his time.” John Feagans received one of the 300-baud modems to experiment with. “I had one for the PET. Modems of that time were 300 baud and there were no protocols for error correction,” he says. “1200 and 2400 was on the horizon.” “We had some pretty good communications guys,” says Peddle. “We had a clear understanding that we would get nominal speeds. I don’t think we projected 56 [kilobaud] but we had certainly projected 20. It isn’t screaming fast, but in the kingdom of the blind, the one eyed man is king.” Peddle knew the future of computing far before it was actually implemented. “Chuck was telling me earlier where microcomputers were going to be, and he said you wouldn’t have game cartridges and cassettes, it would all be ed over the modem in the future,” says Spencer. “At the time, nobody was doing it. Technically it couldn’t happen then because you didn’t have the bandwidth speed, but people like Chuck knew it was going to happen. He thought it was happening tomorrow rather than five years ahead.” The vast array of projects led Peddle and his team to wonder how a single

computer could use multiple peripherals elegantly. They began to think of the operating system. “The problem was, if you are going to do all this shit, you had to have some kind of an operating system, and you had to have some way of tying it all together,” explains Peddle. “Johnny [Feagans] and some of the software guys were trying to keep it all under one umbrella. We had some interesting questions about that.” Unfortunately, Tramiel did not like what was happening at Moorpark. “Jack definitely felt that we were spending too much time screwing around with all these different things and not getting done what he told us to do,” says Peddle. “His impression was correct; we weren’t getting it done, but we weren’t getting it done because the product that we had chosen to do was dependent on a chip that we didn’t have yet.” “I think he resented all us engineers,” says Seiler. “He didn’t really trust us. Engineers play about 80% of the time and then they do real work 20% of the time. You just need to do it as Chuck did and channel the play so we did useful work.” * * * While Al Charpentier struggled with the 6564 memory timing, he decided to improve the number of colors that the chip could display in high-resolution mode. “Al redesigned the color generator to produce better colors, and I redesigned the video output to provide the multicolor mode that would allow us to produce a bit-mapped display with more than two colors,” says Yannes. Yannes even found time to experiment with his sound synthesis, his true obsession. He improved the sound generator in the 6564 chip, which was initially identical to the 6560 VIC. “I also redesigned the sound generator to add new waveforms. The 6560 had step-approximations of sine waves, which sounded pretty bad. Al didn’t know much about sound synthesis.” Instead of merely generating the same types of sounds using sine waves, Yannes added several different types of sound waves. “Sine waves are pure tones with no harmonics, which makes for pretty uninteresting sounds, and the steps in the waveform made it sound pretty noisy,” he explains. “I changed the waveform generators to provide a saw tooth wave, a square wave, and either a triangle wave or a pulse wave.”

Charpentier and Yannes ran into problems due to the slow DRAM chips. “The system design became a timing nightmare as we tried to accommodate the VIC timing to a slower, multiplexed memory bus, which also needed to be refreshed periodically,” says Yannes. The time limit facing the engineers was demoralizing. “We had between January and June to try to design a chip that can do a 40-character display and they had to try to design a system that would use it,” says Yannes. “That’s what went into the TOI, but it was a nightmare, because we were trying to shoehorn stuff onto the original VIC chip that it just wasn’t meant to do. It was just—eww.” The unrealistic six-month deadline for the 6564 ended up hobbling the engineers rather than spurring them on. “There wasn’t enough time for a proper redesign, where the video chip would have generated all of the DRAM timing strobes, which is why the TOI prototypes weren’t reliable,” says Yannes. “As the parts heated up, the critical timing parameters would drift apart.” Robert Russell watched the second attempt at a color chip crumble. “The big problem [with the TOI] was they just couldn’t make it work,” he says. “The chips and stuff they were using just didn’t hang together.” It was clear to Peddle and his team they would have no video chip by the June CES, and they would fail the six-month deadline given by Jack Tramiel. The engineers now put all their efforts into a video card solution, rather than a video chip. [1]Byte magazine, April 1980.

CHAPTER 15

Turning Japanese 1980

On the surface, Commodore computers look like friendly, innocuous little machines. Behind the scenes, however, the first consumer level computer from Commodore caused a civil war within the company. Jack Tramiel orchestrated the war intentionally, playing one faction off the other. “Jack pitted people against each other,” says Chuck Peddle. “He played those games with everybody all the time.” Al Charpentier witnessed Tramiel breeding animosity between Commodore employees. “He created an environment of competition within his organization, and I’m not sure that is good or bad,” he says. “At the time I felt it wasn’t.” Tramiel was using intense competition to bring out the best in his people. The biggest rivalry occurred between the systems group at MOS Technology and Moorpark in Silicon Valley. Now, he brought Commodore Japan into the competition. With his American engineers failing to come out with a proper Apple Killer, Tramiel turned his gaze to the rest of the Commodore Empire. Up until recently, Commodore Japan had focused on calculator components and keyboard design. They also handled much of the PET manufacturing since 1977, mainly to supply Europe. Now, with Commodore winding down their calculator business, Commodore Japan had to fight to stay relevant. Tramiel met with his general manager, Tony Tokai and the head engineer of Commodore Japan, Yash Terakura. “My name used to be Yasuharu, but it was legally changed to Yash,” he says. According to Commodore engineers, he was a

hardworking, hands-on engineer with an amusing personality. Terakura was popular with all the North American engineers. In many ways, Terakura was more similar to the west coast engineers than he was to his Japanese contemporaries. “He wore his hair long, which was not fashionable in Japan at the time,” recalls Michael Tomczyk. “He was kind of an odd, easygoing, out there, cutting edge type wild-eyed engineer. He was a wonderful character and a genius.” Born in 1948, Terakura was 31 years old at the time and a father of three children, his latest born just months earlier. When Terakura was a teenager, his father, an entrepreneurial businessman with his own electronics company, sent Terakura to the United States to finish his education. “I went to high school in the US and then went to Oregon State University and majored in nuclear engineering,” he recalls. “I didn’t go anywhere on the nuclear stuff because it wasn’t that big at that time. So then, I decided to go into electronics. I ended up going to the RCA Institute of Technology, a kind of technical school over there.” After finishing his education in the early 1970s, Terakura returned to Japan to work for his father’s company. “I worked as electronics engineer.” Commodore’s constant search for cheaper components put them into with the Terakura business. “My father got a contract with Commodore when they were making calculators,” he recalls. “Since I could speak English, they said, ‘Why don’t you help us?’ I said, ‘Okay, fine.’” Terakura performed so much consulting work for Commodore that they hired him. “Somehow I ended up working for Commodore rather than my father’s company,” he says. In the mid 1970s, Terakura worked in Commodore Japan’s offices, located in Osaka, a massive city of over two million people. The GM of Commodore Japan was Mr. Fakagi. Unfortunately, the US educated engineer did not do well under Fakagi’s traditional style of Japanese management. “The way they operate is a little weird because it’s the typical Japanese way. I just didn’t like the way they managed people. I had my education in the United States, so I was kind of leaning toward the US way of doing things.” With the traditional management style, all decisions came from the top. Fakagi did not expect nor encourage his engineers to offer their own opinions, but the

rebellious engineer found that difficult. “When the boss says something, you just do whatever the boss says, but I didn’t do that,” he recalls. “I’m pretty much an independent person, so if somebody tells me what to do from the back seat, I just don’t like that idea.” The clash of cultures between the US educated engineer and his Japanese educated manager was bound to erupt. “Every time they said, ‘Do this or do that,’ then I got in a fight because when I think something is not right, I always tell them it’s not right. Normally the Japanese don’t like that.” In 1975, it was apparent the situation could not last. “I kind of got into a little fight with them, and then I quit the company.” He pauses for a moment, then adds, “Well actually, they fired me from Commodore Japan.” After Tramiel restructured Commodore Japan in 1976, amid falling calculator profits, he moved their offices to Tokyo with Tony Tokai as the new GM. “At that time they called me and wanted me back in Commodore Japan heading the engineering office,” says Terakura. With Commodore revving up production on the PET 2001, Tokai needed skilled engineers. “So I said, ‘Okay fine.’ I went back to Commodore and that’s when I actually started deg the computers.” Although Terakura remained in Japan, he frequently traveled to the United States offices for meetings. As head of engineering, he received his instructions directly from Tramiel. “He said, ‘First just stay in Japan and do some liaison, and from time to time come to the United States and help us.’ So I was doing a lot of liaison work, but mostly I stayed in Japan. I worked with Chuck Peddle and all those people developing the calculator chips. Then after that, Commodore kind of moved into computers.” Chuck Peddle knew Terakura from his earlier visits to Japan when he put together the Epson printer deal. He believes Terakura’s interactions in the US westernized his personality. “Yash at one time was a very Japanese guy,” he says. “He was involved with the work we did with Epson. He taught me a lot about Japanese culture. I learned to stop and ire my plate of food from him. He’s just a straight guy; a serious, intelligent, hard-working, very focused guy.” Terakura was largely unaware of the political battles between the east coast and west coast when Tramiel asked him to design a color PET computer. “I don’t

recall him ever getting caught up with anybody’s politics,” says Peddle. Tramiel asked Terakura to bring his prototype to the Hanover Fair in April 1980. To add color to the PET, Terakura replaced the 6845 video chip with a standard color video chip similar to the Texas Instruments TMS9918 chip. This was not a plug-in replacement, forcing Terakura to rewire the video section of the board. “It was a mockup with experimental stuff,” says Terakura. The engineer received help from a group of young Japanese PET 2001 owners, who frequently stopped by the Commodore Japan office. “All those kids from high school and college, they all owned a PET,” he says. “It was pretty expensive, like a $3000 machine, but everybody had one.” At home, the kids used their primitive PET 2001 computers with only eight kilobytes of memory and a rubber calculator-style keyboard. When Commodore released the more expensive PET computers with more memory and real keyboards, the kids yearned to use the better machines. “They had smaller machines and wanted to have bigger and better machines, so they came over to my office to use the better machine we had.” One of the young engineers was 21-year-old Satoru Iwata, the future CEO of Nintendo.[1] “Iwata was one of these kind of groupies when I was in Commodore who used to come to my office all the time,” says Terakura. “He tried to get all the information from me about the new type of machines.”

Yash Terakura (right) reuniting with Nintendo CEO Satoru Iwata in 2002.

Iwata had been a student at the Tokyo Institute of Technology since 1978. “I would have loved to study videogame programming, but nobody was teaching it then,” recalls Iwata. “So I went to classes on engineering and early computer science.”[2] Although curiosity brought the young programmers to Commodore Japan, they eventually became unpaid interns for the company. “His job was to come to my office and kind of clean everything up, making backup files of the different software,” laughs Terakura. “Those days we had a bunch of floppies, so he had to check everything and make sure everything was updated. He was pretty much my own secretary.” Terakura was something of a mentor to the young Iwata. “He was going to school at that time in college. So on the way home he always stopped by our office, almost every day. There were a few of them like that. That was before the Nintendo days and everybody wanted to work on a PET.” When Tramiel gave Terakura the assignment to design a new color computer, Iwata and the other groupies were understandably excited. “Being a software engineer, he wanted to get all the inside information I guess,” he says. Iwata’s intense interest in the machine was apparent, and soon he was helping to code the ROM software in the prototype. “He was a very good software engineer and he was helping me develop some of the test programs,” says Terakura. “Actually he helped me design the software side of the Commodore ColorPET, the one I built.” Another groupie at Commodore Japan was Kazuhiko Nishi, known to North Americans as Kay Nishi. The engineer was indispensible to Microsoft and Bill Gates in licensing Microsoft BASIC in Japan. “Nishi too was one of the groupies,” says Terakura. “He used to come to the Commodore Japan office to gather information on the PET 2001.” The Japanese engineer continued development on the prototype until late April, when it would be time to take their machine to Hanover, and present it

to Tramiel for his approval. * * * Tramiel sent his west coast engineers to the fifth West Coast Computer Faire in nearby San Francisco for three days, starting Friday, March 14th. “Commodore had a booth there but we only showed standard PET systems,” says John Feagans. The engineers demonstrated the new 80-column PET at the show. In early April, with the Hanover Fair fast approaching, Tramiel decided he wanted his MOS Technology engineers to make another attempt at a color computer. “I think it was around April of 1980 that Jack told us he wanted to show a PET with a color display at a big European show, which was only a few weeks away, of course,” recalls Bob Yannes. Bill Seiler and the west coast engineers had helped with the previous attempt at a color PET, which they showed in January, but this time Tramiel left them out of the project completely. “I think basically he was getting kind of unhappy with the west coast group,” says Yannes. “Either they couldn’t make the products that he wanted or they were taking too long.” Tramiel was looking for a redundant systems manager, since it looked like Peddle was incapable of getting what he wanted done. “There’s a whole bunch of management theories that say you shouldn’t have anybody who is irreplaceable,” explains Peddle. “Jack was always looking to do that. At Commodore if you would ask the stockholders, they would say the same thing: ‘Get this guy backed up. You should always have two of them.’” Yannes feels the request for a color PET was merely an exercise to test the east coast group. “I don’t think Jack had any real interest in the product. He simply wanted to see if the engineering group at MOS Technology, Charlie [Winterble], Al [Charpentier], and I, could deliver what he demanded when he wanted it.” “He said, ‘I want you to do this, and you’ve got two weeks to do it in,’” recalls Yannes. “At the time I was thinking, ‘He’s got to be kidding. Does anyone really want this product? When are they planning on producing it?’ I really think that all Jack intended to do was to see if we would jump through hoops for him and give him what he asked for, regardless of what he asked for.” With only weeks to go before the show, the team could not even contemplate

creating a new silicon chip. Instead, they created a circuit board. “There was just no time to do anything,” recalls Yannes. “I hacked together a simple color board that connected to the existing PET, and we put together a few units for the show. It was just a quick slap-together of stuff to get a color screen on the PET.” The result was a PET computer that could display bitmapped color graphics, but not color text. “There were no changes to the system at all,” explains Yannes. “All you had to do was, instead of writing one bit into memory to turn a white pixel on, you would write red, green, or blue. It didn’t have intensities or anything. It was just either red, green, blue, yellow, cyan, magenta, black, or white.” Yannes also programmed a simple demonstration for the show. “There was no software for it,” he recalls. “I think we scanned a color picture and had it throw it up on screen.” Although Yannes questioned the usefulness of the product, the team built it anyway. “I don’t think Jack had any intention of producing it. He just wanted to see if he could come in, disrupt our schedule, say, ‘Drop whatever you’re doing. You’ve got two weeks to do this—do it.’ And we did it. I don’t think Jack really trusted anyone except his family. So you really had to prove yourself with him.” Charles Winterble also believes Tramiel did not intend to sell the product. “It was nothing,” he recalls. “Part of the problem is that it wasn’t really well integrated. It was obviously an add-on. I’m sure Chuck’s ideas would have been the right way to do it.” Tramiel was impressed with the three east coast engineers. “That gave him the confidence that we would perhaps be able to pick up some of the design aspects of their products if the west coast group either wasn’t doing it or left,” says Yannes. “I don’t know what kind of politics were going on over there, but it was pretty clear Jack was trying to find some other group of people in the company to design products.” * * * While the 6564 video chip floundered, Chuck Peddle continued managing the projects under development at Moorpark. Peddle and Feagans wanted to allow these devices to connect seamlessly to the PET, but they began to recognize a problem. How could they allow these devices to work with different software

applications, such as word processors and spreadsheets? Each device needed a small piece of code, called a device driver, to run on the PET. Previously they had put the device drivers in the ROM, but if a developer released a new device, such as a printer, how could they get the new device to work with numerous different applications on the PET? “The higher-end CBM machines were absolutely running out of gas with the 6502,” explains Peddle. “We couldn’t develop the languages and we really couldn’t put a proper diskoperating system on it.” They needed to be able to add device drivers to the operating system. Unfortunately, the operating system in ROM was unchangeable once it left the factory. It was a difficult problem. “We had concluded that we had to get a totally different operating system, with totally different memory capabilities than we had in the 6502,” says Peddle. “It wasn’t a contender at that point.” As Feagans and Peddle discussed the problem, they realized they needed a changeable operating system, able to adapt as new hardware came along. The operating system would need to be available at all times, even if the removed the floppy disk. “Our job was to guess where things were going,” says Peddle. He believed computer architecture would center around permanent data storage devices. “I tried to convince Commodore that the future of computing was hard disks.” Peddle’s team began deg a hard drive for Commodore. “We had done a lot of really pioneering work on how we were going to do the disk drive and what the disk drive meant to us,” says Peddle. “We were considering putting a bunch of disk management stuff out in the hard drive controller, but you start getting too divorced from the central processor.” Peddle assigned the development of the hard drive to one of his R & D teams. “The hard drive came out of Phoenix,” says Seiler. “Bob Taylor was in Phoenix deg the hard drive.” Peddle’s analysis of the problem led him to some difficult realizations. “The hard disk effectively forced us to believe the analysis we had done. We concluded that we had the wrong microprocessor. It’s a big tough conclusion when you’ve built your whole business around a microprocessor.” Peddle abandoned the 6502, and with it, the TOI. “It started out as the Apple

killer,” says Peddle. “Then we changed its direction. What we needed was a totally different machine for the high-end—higher disk space, lots of memory, and a much better processor.” Peddle developed plans for a more advanced business computer. “We had basically decided it had to have much better graphics but not necessarily color,” reveals Peddle. “We thought there was a tradeoff to be made there. You could make an 80-column screen relatively easily if you didn’t have color. It had to have bigger floppy disks, and it had to be extensible into a hard disk.” The computer had steep memory requirements, and Peddle needed a microprocessor able to handle large amounts of memory. Unfortunately, MOS Technology had failed to develop a 16-bit version of the 6502. “We had to make a decision on what was going to be a contender,” says Peddle. The Intel 8088 microprocessor, released the previous year, seemed like the best candidate because it could address up to a megabyte of memory. To Tramiel’s dismay, work on his Apple killer virtually stopped while Peddle and his engineers regrouped. “We kind of abandoned the idea of doing what would be an Apple II look alike,” says Peddle. “I suspect it is one of the things that Jack got mad about.” The differing goals created confrontations with the west coast engineers. Most of the arguments with Tramiel revolved around cost. “Jack was tight,” says Seiler. “Most of the time, it was about saving money and doing it the most economical way. Jack was maximizing what he could get out of what he had to put in.” Seiler continued development on the TOI project, and often had discussions with Tramiel about the product. “I being in a meeting with Jack and we were arguing about something fairly technical that I know Jack didn’t know much about,” recalls Seiler. “But he won the argument. He was good at arguing.” After coming out on the losing end of a few arguments, Seiler tried his best to avoid confrontations. “I just really didn’t have anything to do with him as much as I could unless I had to,” he says. Peddle developed the technical specification for the next generation computer. He also developed a business plan, which he would unveil at the general managers meeting in London.

* * * In April 1980, Tramiel held an international meeting in London. “Jack brought a bunch of us to Europe to set up a dual planning session,” recalls Peddle. Prior to the meeting, Tramiel surveyed the computer scene in Britain and was surprised to see his old calculator competitor, Sinclair, in the computer market. Tramiel knew low-cost computers were beginning to appear on the computer scene and he was familiar with the Atari 400, released the previous year for under $600. What he saw in Britain was a revelation. The remarkable Clive Sinclair filled the roles of both Tramiel and Peddle in one person. He ran the company and designed their products, while carrying on highprofile relationships with an ever-changing line of beautiful women. Sinclair previously developed a computer similar to the KIM-1 called the MK-14, with modest success. In February 1980, he released the ZX80, which he designed for economy. It lacked color and even sound, but the sub-£100 price made it a success. Spencer was familiar with Sinclair. “I’d run into him several times and quite often played tennis with his brother,” he recalls. “Sinclair was very much involved in the same business as us. In calculators, he was a very early innovator.” When Tramiel saw the ultra-low-cost product sold in UK stores, he knew he wanted a Sinclair Killer of his own. “The Sinclair was directly responsible for Commodore doing the VIC-20,” explains Peddle. “He saw that it was taking off. Fundamentally, Jack always was into the market, and the market was buying the Sinclair machine.” On his first day in London, Tramiel held several private meetings with top Commodore executives. “He and Irving had a meeting in downtown London,” recalls Peddle. “Dick [Sanford] and I went to dinner with Irving and Jack, and we had this nice conversation, and everything went fine. It was a very pleasant meeting.” After the dinner meeting, Tramiel had a small run-in with the London police. “We had a little too much to drink and he got busted on the way home,” recalls Peddle. “He went the wrong way around a roundabout, which is the reason the cops busted him, but the cops were nice enough to let him go home.”

Tramiel rented a beautiful English country club called Burnham Beeches to host the conference, complete with waiters in tuxedoes. “We drive back to Burnham Beeches, which is out in the woods somewhere,” says Peddle. The next day, the major players within Commodore convened in a meeting room at Burnham Beeches. Sitting around the table were Harald Speyer of Commodore , Kit Spencer and Bob Gleadow of Commodore UK, Tony Tokai and Yash Terakura of Commodore Japan, Jim Dionne of Commodore Canada, and Ernst DeMuth of Commodore Switzerland. The top executives of Commodore US also attended, including Chris Fish, Dick Sanford, Dick Powers, Chuck Peddle, Jerone Guinn, and Tramiel’s personal assistant, Michael Tomczyk. According to Tomczyk, Tramiel was ready for a new initiative. “He said to me in our first meeting in 1980 that he wasn’t sure if it was time to come back to the US market and compete head to head yet, but he was getting the sense that maybe it was almost time, and he could use people like me to help him do that,” he recalls. “So basically he was mentally preparing to come back and compete in the US.” “Jack’s late for the meeting the next day, so we start the session off,” recalls Peddle. The executives felt comfortable starting the meeting without Tramiel. After all, the computer systems division operated almost independently of Tramiel. Until recently, Tramiel focused on calculator products. Chuck Peddle ran the computer business. Peddle wanted to convince Commodore management to phase out the PET and his next generation computer. “We were trying to convince Jack and the rest of the company that there was a serious business opportunity in going up market—a new machine, a new architecture, everything for the business computer market.” It was a shocking suggestion, considering the success of the PET in Europe, but he met with almost unanimous agreement that Commodore should pursue the upscale market. “All of the team had agreed that we needed to have a really good business product as a follow-on product [to the PET],” says Peddle. Later in the morning, Tramiel finally appeared. “Jack walks into the room after having studied the Sinclair thing,” says Peddle. “In that moment, he decided that

he was going to take back control of the company from those of us that were off doing the business stuff. Here was his chance to go back into the consumer electronics business, which he felt he could do better than everybody in the room.” Sometime after the January CES, Tramiel correctly surmised that his head engineer was no longer listening to him. When calculators fizzled, all the power at Commodore was with computer products. Tramiel was in danger of losing relevance in his own company. When he asked for an Apple killer, Peddle did not deliver it. It seemed like Peddle was making his own decisions. Even worse, the rebellion was spreading among his own executives. Tramiel knew he would have to do something to get his rogue engineer under control. Tramiel curtly dismissed Peddle’s next generation proposal. “We had a business plan and the whole management team had put it together,” says Peddle. “Jack walks in and says, ‘Forget it. This is where the company is going.’” The new plan was the low-end market. “He blocked out that we were going to do something and compete against the Sinclair,” says Peddle. “He expected us to then have the follow-on Apple product as soon as we could get it.” The executives had mixed reactions to his announcement. “It was a gut level, almost violent response to his suggestion to do a small color computer instead of an Apple-type color computer,” says Tomczyk. “Everyone from the engineering staff to the general managers felt that we should keep making $1000 computers and it wasn’t time to undercut ourselves and make cheaper computers.” Others questioned whether it would be possible to make a profit on something that sold for so little. Tramiel was not worried. He saw the same thing happen with calculators and believed the same would inevitably happen with computers. Commodore now had impressive vertical integration and he was confident it would give it an advantage over the competition. Tramiel gave a compelling justification for low-cost computers. “Jack had this overriding, almost religious philosophy that we should make computers for the masses, not the classes,” says Tomczyk. “He realized that computers were not going to catch on until they were affordable by all levels of the economic strata, including schools from kindergarten to high school and college.” Some attendees who initially ed Peddle’s business computer now

abandoned him as soon as they realized Tramiel did not it. “Of course, all of the local managers jumped in and told him what a great idea it was,” says Peddle. A handful of inner family ed Tramiel’s vision: Tony Tokai, Yash Terakura, and Kit Spencer. Jerone Guinn of MOS Technology also ed the proposal, since he believed his engineers could produce the computer. Tramiel’s personal assistant, Michael Tomczyk, was also a believer. Peddle was disappointed Spencer did not fight more strongly for his business computer proposal. “Spencer was the consummate politician, so he was pushing it until Jack walked in and said something else, then he was immediately on Jack’s side.” Spencer claims he ed the business proposal as strongly as the consumer product. “I basically thought both computers were a good idea. It wasn’t necessary to be for one or the other,” he explains. “I think Chuck was pushing in one direction and Jack was pushing in another. We needed to be pushing in both directions. The prime question was about which the first priority should be.” Tramiel instructed his managers to discuss the idea further and departed. He would return the next day to review their plans. Peddle was devastated. He was a purist when it came to computers and he wanted elegance and power in computing. Now he was supposed to get behind a computer less powerful than what Atari, Apple, and NEC currently produced. Once Tramiel left the room, Peddle began formulating a counter-proposal. “I had the whole European sales staff on my side, because they had built their reputation on business computers,” says Peddle. “Our response, which was put together that same day, was, ‘Jack, we accept that that’s where you want to go, but the company is big enough to handle both. You go off and drive that and let us go build the business computer business.’” That night, as Tramiel slept, his idea for a low-cost computer crystallized. When he awoke, he had a clear goal for Commodore. He returned to the meeting and asked the executives to discuss the strategies they came up with the previous day. Instead, Dick Sanford suggested the idea to create a new division, and Peddle championed the idea. “We proposed that we split the company,” says Peddle. “Jack would set up a separate division and let us go off and pursue the

business market. We would help him get his consumer business going until he could bring in a new team.” Tramiel was not convinced. “He had taken a position that, ‘Okay, we’ve had fun in the business computer stuff in Europe but we’re going to go focus on what we know best, which is consumer electronics,’” says Peddle. Tramiel felt the suggestion was open rebellion. “There is no question that he viewed that as a direct insult to his idea that we are going to go back into the consumer electronics business,” says Peddle. “He put all of us on a short leash that day.” The counter-proposal landed many of Commodore’s top executives in Tramiel’s cross hairs. “He was unhappy with Chris [Fish]; that’s why he screwed around with Chris,” says Peddle. “He was unhappy with Dick Sanford; that’s why he ultimately forced Sanford out. … He was unhappy with all of them, because he perceived that we didn’t see his vision.” When Peddle had returned from Apple, he had repaired his relationship with Tramiel by successfully releasing printers, disk drives, and new revisions to the PET line. Unfortunately, it did not last. “The personal relationship was very good and then it got bad again, which happened to almost everybody with Jack,” says Peddle. According to Peddle, they were more than happy to carry out Tramiel’s plan. “The problem was, we did see his vision. We knew what he needed to have done, and we could see how he could get between here and there. But we had a different vision. We had a vision of what ultimately became the PC business and he didn’t share that vision.” The discussions eventually deteriorated into small groups of executives talking together. Tramiel allowed the discussions to continue for a while, then stood up and banged his palm on the table. With all eyes on him, he announced, “The Japanese are coming, so we will become the Japanese.” The statement had everyone pondering the meaning of his words. Tramiel identified the Japanese as his main competition early on, just like other companies such as Apple. In the early eighties, Japan was firmly on the minds of many North Americans. Since the late sixties, Japan had taken over American industries one at a time: small electronics, household appliances, televisions,

radios, stereos, and automobiles. Now they had their sights set on the personal computer market. Tramiel knew their primary advantage was their ability to deliver low-cost products. “When the general managers gave him a huge pushback on the second day of that meeting, his gut level response was, ‘Hey, guys, we can’t wait for the Japanese to undercut us,’” says Tomczyk. To become Japanese, Tramiel wanted to undercut their prices. “I think he originally set some price goals that left them aghast,” recalls Robert Russell. Tramiel was a master tactician in the world of business, and he took the battle to his enemy. “Jack wanted to fight the Japanese in their home market with a lowcost machine, rather than the Japanese coming over with a low-cost machine to fight us in the US,” says Russell. Tramiel had seen calculator prices drop dramatically and almost put him out of business. Now he wanted to be the one driving prices down. Although Commodore had previously introduced new products to the North American market, he decided to release the low-cost computer to Japan first. North America would follow. “It wasn’t exclusive to Japan, but Jack wanted to scare the Japanese right on their own home ground and sell it there,” says Russell. Tramiel discussed competition with Japan in a March 1986 interview with German magazine Data World. “The Japanese can only be successful in the computer industry if there are no longer people like me. … The Japanese think on a long-term basis and need a three-year plan. They are not innovative and can only be successful if innovative people disappear from the industry.”

Tony Tokai (right) meeting with a young Kay Nishi.

Ironically, Tramiel was more like the Japanese than he was to his contemporary North Americans. The Japanese lived and breathed the philosophy of Sun-Tsu’s Art of War. To Tramiel, business was war. His statement meant Commodore would not wait around for the Japanese to take over yet another industry pioneered in the western world. One of Tramiel’s most famous quotes is, “The minute you’re through changing, you’re through.” Now he was going to change Commodore. Appropriately, the 1979 hit song, Turning Japanese by the Vapors summed up his new philosophy. Peddle could not help but feel resentment as he watched Tramiel chip away at his lead engineer role. He knew the fundamentals of computers and wanted nothing more than to design the next great machine. Now he was supposed to sit back quietly while Tramiel decided Commodore’s next generation of computers. [1] On May 31, 2002, Iwata ascended to President and CEO of Nintendo when Hiroshi Yamauchi, the company’s president since 1949, retired. He went on to launch the massively successful Nintendo DS and Wii products. [2] From his 2005 keynote address at Game Developers Conference.

CHAPTER 16

Small Computers 1980

In the first half of 1980, Commodore engineers participated in at least five major computer shows. Although the shows cost money for Commodore to attend, Jack Tramiel ed participation. His engineers were often engaged in tedious, meticulous work, cooped up in chaotic labs with poor ventilation and acrid smells. The reward for many engineers was to be able to show their creations off to others. Tramiel motivated his engineers by giving them deadlines to display their prototypes at shows, spurring them to work faster. At the spring Hanover Fair in , Commodore had a large and diverse assortment of finished products for display. Many of the products from the computer systems division made it to the show due to Chuck Peddle’s decision to isolate his engineers at the Moorpark lab. On display were the CBM 4032 and 8032 computers, both containing 32 kilobytes of memory. The business community was glad to see a true 80-column PET finally. The Phoenix team had delivered the 8050 dual disk drive, which doubled the storage capacity (double density) from the 2040 to 340 KB per diskette. Commodore announced a suggested price of $1,695.00 for the unit. Peddle also displayed two new printers, the 4021 and 4022, which used Japanese components. One of the most interesting products was the introduction of the 8010 modem. “The modem stuff was a direct result of the work that was going on at Moorpark,” says Peddle. The 8010 was an acoustic coupling modem, which meant the was required to manually dial a number to the connecting computer and then place the

receiver in a specially designed coupling device. The modem retailed for $395.00 and was capable of delivering the groundbreaking speed of 300 baud or approximately one line of text per second on a 40-column screen. The impressive yield of new products validated Peddle’s creation of the Moorpark research lab. However, the product that Tramiel was most eager to display was the Japanese ColorPET, which he hoped the attendees would embrace as a true competitor to the Apple II. Yash Terakura personally delivered the prototype. “I took that one to Hannover Messe,” he recalls. “At that time I met Michael Tomczyk.”

Prominent Commodore advertising outside the Hanover Fair.

Tramiel’s personal assistant was rising within the Commodore ranks, and becoming one of the inner family. Months later, he would visit Japan to help launch Commodore’s next computer. “I got to know Yash and his wife very well and he was one of my favorite people in all of Commodore,” says Tomczyk. Aside from Terakura’s prototype was the rushed color PET from the west coast engineers. “We went to the show and showed it,” says Charles Winterble. “I doing a lot of scrambling around trying to get RGB monitors and stuff like that from .” Tramiel closely watched the attendees to gauge their reactions to both computers. He had his fingers firmly on the pulse of the marketplace, and if the market did not react, it would not be worth taking the prototypes to production. Although Tramiel was pleased with his engineers for delivering their prototypes so quickly, it was clear the attendees were unenthusiastic about the computers. Already other systems like the Atari 800 and TI-99/4 boasted superior graphics, and it was clear the graphics of the color PET prototypes did not match up. After the show, Tramiel gave up plans for a ColorPET product. * * * Bill Seiler worked extra-long hours to get his TOI prototype ready for a demonstration at the National Computer Conference. “If you don’t put your foot in the sand, the engineers will never get it done,” he says. “If there is a show coming, we’ve got to get it working.” Tramiel spent very little to send his engineers to computer shows in California. His employees either took the Amtrak train or flew Air California, a regional airline, for under $20 per trip. On May 19, 1980, the engineers travelled to Anaheim, California for the four-day NCC. “The Moorpark development team was there: me, Bill, Glen, and Scott,” says John Feagans. The four day conference, held at the Disneyland Hotel convention center, was a playground for computer aficionados. Attendees took in dozens of presentations

on an incredible number of cutting-edge computer topics. The engineers shared rooms at the Disneyland Hotel. During evenings, they spent time in the Disneyland theme park. Bill Seiler was a curious man by nature and often became so engrossed in technology he forgot his surroundings. One evening after the conference activities, a laser beam captured Seiler’s attention. “Bill was playing with a helium balloon in the lobby of the Disneyland Hotel, running it through a laser beam,” says Feagans. While other attendees took in the rides at Disneyland, Seiler studied the blue laser beam attached to the Disneyland Hotel. Lasers were all the rage in the early eighties and Seiler was fascinated. He purchased a silver Mylar helium balloon with a long string and floated the balloon in front of the laser, studying the way the balloon scattered the light. Hours later, as his fellow employees returned from the park, he was still engrossed by the laser. Unlike CES, NCC had lots of computer savvy people in attendance. It was the perfect venue to demonstrate prototypes and new products. Apple attended the conference, and announced the Apple III—the computer Chuck Peddle had previously contributed to. The engineers brought along their latest Moorpark creations, which they displayed to the public at the Commodore booth. “We showed the 2030 single disk drive, which later became the disk for the VIC-20,” says Feagans. “We also showed a printer I had worked on.” However, the TOI prototype was not working well enough for public display. “We built some and it sounds like we may have gotten that one off to some show,” says Seiler. “But there were still problems with the display and it wasn’t working quite right.” Instead, the engineers gave back-room demonstrations of the computer in their hotel room. “We demonstrated Bill’s TOI computer, which was in the suite,” says Feagans. The engineers hoped to elicit and suggestions from attendees. The display flickered and produced random dots across the screen that would gradually deteriorate, until the display crashed the computer. “I actually have schematics for the TOI prototype and it was pretty messy,” says Bob Yannes. “I Chuck’s group was having trouble with memory reliability as the

system heated up. It was a real kludge.” The demonstrations were tightly controlled, and the engineers used an aerosol can of cold spray to keep the components cool. Seiler would attempt to remedy the problem within a month, when Tramiel expected the engineers to display the Apple killer at the June CES show in Chicago. It seemed unlikely he would be able to solve the problems by then. The Moorpark engineers were in trouble. * * * Commodore’s next computer was started by 22 year old Bob Yannes. A product of the seventies, he looked more like a teenager, resembling a young Luke Skywalker with glasses. “Bob was kind of a thin, young guy with blonde sandy hair,” recalls Charles Winterble. “[He was] just the nicest guy in the world.” Yannes had a fondness for the 6560 VIC chip, but so far he had been unable to find anyone to develop a product for it. In May 1980, he saw his opportunity to produce a VIC based computer when Sinclair released the ZX80. “It was in the spring of 1980 that the Sinclair ZX80 came out, and all the great press about it was really frustrating to me,” recalls Yannes. Most Commodore engineers were familiar with the British developer. “I met [Clive] Sinclair at some show in America,” says Bill Seiler. “I talked to him for quite a while. He was a pretty nice guy.” The ZX80 used a low-cost Zilog Z80 microprocessor. “He was there with his little Z80 based thing,” says Seiler. “It had a membrane keyboard on it and it was driving a video monitor. I think it had BASIC. I was impressed by how he was doing the video with the main processor, the Z80. It was really cheap.” Yannes, on the other hand, was not impressed by the machine. “I thought the Sinclair ZX80 was just terrible. [The video display] was black-and-white. Every time you typed on the keyboard, the display went blank because they were doing the display update in software.” Yannes felt he could offer more value for a lower price. “Here’s this crummy little Sinclair computer out there for 250 bucks, and we can do a real cool one for 200 bucks,” he explains. “To hear all this good press about all these people being excited about it, I just said, ‘This is crazy. We have this little video chip here that can make a great little color graphics computer for less than the

Sinclair. We can even put a real keyboard on it and make a nice product instead of that horrid thing.’” Yannes began forming a picture of a Commodore computer using the VIC chip. “I had sort of come to the conclusion that no one in the outside world seems to be interested in this chip for whatever reasons. Fine, why don’t we do something with it? It would have color graphics and the screen wouldn’t go blank when you typed on the keyboard. You could play games and things.” “Bob said he would put this prototype together that might show more capabilities,” recalls Charpentier. “Using the knowledge he gained doing his senior project, he was able to put his prototype together.” Yannes was aware of the one weakness in the VIC chip. “All we needed was to get over this mentality that a 22-column display was not good enough,” he says. “That’s when I made the first prototype of what would ultimately be the VIC20.” The VIC chip also received the new color generator from the failed 6564 chip. “Meanwhile, since we had already finished the work on the new video section, we dropped that back into the 22-column 6560 VIC, still hoping to generate some outside sales interest in the chip,” says Yannes. Yannes also wanted to improve the sound capabilities of the chip, but he was rebuffed. “The tone generators that were in the original VIC-I chip when I first started working there were not really very good,” he explains. “I never understood why we didn’t, as the new color generator required an all-level mask change, so there wouldn’t have been any additional cost and no more risk than adding the new color generator.” Bill Seiler, who previously created sound effects for Allied Leisure, was impressed with Yannes’ work. “Yannes was working on incredible sound things because he was really into audio stuff,” he says. Yannes’ project came together in his home. “There’s a story going around about how I built the VIC-20 prototype in my garage,” he says. “It wasn’t my garage, it was my bedroom. I didn’t have a garage.” Yannes assembled a crude prototype and did not even bother to permanently solder the connections. “I wire wrapped the thing,” he says. “I just took an old

memory board from a KIM-1 and put a perf board on the side of it, then hooked up a processor and the VIC chip. Then I built a little case out of some sheet plastic that I had lying around, and I put the old PET calculator-style keyboard on it.” After three days of work, Yannes had a computer that connected directly to his television set. “I called it the MicroPET,” he recalls. “I said it was a picocomputer. If the PET was a microcomputer, then this was a picocomputer.” Yannes showed his MicroPET to Al Charpentier and Charles Winterble. “He’s tinkering around, and he comes in one day, and Al grabs me and says, ‘You need to look at this thing.’” recalls an amazed Winterble. “What Bob had done was jury-rig a character memory ROM and put some interface chips into it, so the VIC could put characters on the screen. He demonstrated it and we said, ‘Holy mackerel.’” Through Jerone Guinn, Winterble knew that Tramiel was looking for a low cost computer that precisely matched what Yannes had put together. It was exactly what they needed at exactly the right time. Later, Yannes realized his managers saw his computer as a way to bring systems design to the east coast. “I guess when Al and Charlie saw it, they thought, ‘Hmm, this is interesting.’ I was pretty naïve about politics and stuff. I was just doing my job. What we decided to do was build up a much nicer prototype and when Jack Tramiel was going to be at MOS Technology for the general managers meeting, it would be presented to him.” Winterble decided to produce a more refined case. “We went to a local shop, and we had a little case molded,” he recalls. “It wasn’t even injection-molded, it was a vacuum-formed case—the kind of thing where you take a sheet of plastic and it’s vacuum formed into a shape. Then we put metal on the bottom of it and we put a keyboard in there.” Winterble and Charpentier showed it to their boss, Jerone Guinn. Winterble explains, “Then the issue became, ‘Well, what have you got here?’ The idea was you put Microsoft BASIC on here and we’ve got ourselves a little BASIC computer.” The MOS Technology team decided they would not tell the competing Moorpark engineers about the MicroPET. “They were being secretive working on

it before the meeting so Chuck’s team wouldn’t catch on and deliver,” explains Yannes. Without experienced programmers like Robert Russell or John Feagans, there was no way Yannes could modify the BASIC ROM to work with the VIC prototype in time for the show. As a result, the prototype lacked an operating system or BASIC. “It was an extremely minimal prototype,” recalls Leonard Tramiel. “It did its job to demonstrate what the machine could do.” Without an operating system, Yannes could not load programs into memory from disk or tape. Instead, he created a demonstration program and put it in a ROM chip. “I wrote this little software routine that would tell the features of the machine and scroll these screens up,” he recalls. The VIC chip’s scrolling effects were impressive. Messages popped onto the screen, then slid off the top as background music played. They could even slide down, left or right. Yannes considered the MicroPET a concept machine rather than a production machine. “It was just for show,” he says. “It was never a functional computer. There was no functional operating system or anything.” In May 1980, two weeks before CES, Tramiel and Peddle stopped by West Chester for a general managers meeting. “On his way back from New York, he goes to see those guys, and they show him this architecture which they had created on their own,” recalls Peddle. “Charlie [Winterble] was the one who engineered this presentation to Jack,” says Yannes. “Jack didn’t know it was coming, and none of the people at the meeting knew it was coming except for Charlie and Al.” Peddle believes their manager, Jerone Guinn, saw an opportunity to take over systems design. “Yannes didn’t understand that his boss [Jerone Guinn] was off pumping Jack up that he could do all that work on the east coast,” he says. Going into the meeting, Tramiel was already unhappy with Peddle and looking for an alternative. “I don’t know the politics involved, but I guess Jack had been trying for some time to design an inexpensive home computer, and they kept turning out things that were too complicated and too expensive,” says Yannes. Tramiel took his seat at the meeting, along with Peddle. After some discussion, Charpentier and Winterble revealed the MicroPET. The demo showed off all the

capabilities of the VIC chip. “It ended up looking very impressive, because you didn’t normally see that, even on high-end monitors,” says Yannes. “I kept the whole demonstration in black-and-white, and then the very last thing said, ‘Oh and by the way, it has color,’ and then it went into this color kaleidoscope program. It was a pretty dramatic demonstration.” Although Yannes only intended to demonstrate the potential of the VIC chip, Tramiel loved the entire machine. “I was not at the meeting, but apparently it blew Jack away,” recalls Yannes. “He wanted to do something with this right away. When he saw this, he said, ‘We want this. We want to do this.’” According to Charpentier, the conversation with Tramiel went something like this:

Charpentier: “Here’s what the chip can do. What do you think?” Tramiel: “All right, ship it.” Charpentier: “You mean the chip?” Tramiel: “No, no, no. I mean that!” (pointing to prototype) Charpentier: “This?” Tramiel: “That.” Charpentier: “This?” Tramiel: “That.” Charpentier: “Really?” Tramiel: “That.”

Despite the limited ability of the prototype, Tramiel was impressed. “The Sinclair didn’t do much either,” says Peddle. “That was one of the reasons why it didn’t bother him as much when those guys showed him what they showed

him.”

The MicroPET running a color kaleidoscope program.

Tramiel wanted his low-cost color computer released by the end of the year—an ambitious goal considering the time it takes to design a computer and set up production lines. In order to expedite the design phase, Tramiel planned to show the prototype at the June CES in Chicago, a mere two weeks away. June was typically when finished products were shown and orders placed by distributors. After the meeting, the managers called Yannes in to commend him. “They called me into the meeting,” he recalls. The rookie engineer felt intimidated by Tramiel. “He was a formidable figure. He just looked like someone you didn’t want to mess with.” Tramiel offered to give the nervous engineer a PET computer. “I already had a PET though,” he recalls. “I was so naïve when he asked me if I had a PET computer. I said, ‘Yes, I do.’ Later on Charlie told me, ‘He was going to give you one so you should have said no.’”[1] Yannes feared the west coast engineers might be upset with him personally. “I wouldn’t be surprised if Peddle’s group thought we had sabotaged them by giving them a bad 40-column chip [the 6564], then blind-siding them with the VIC-20 prototype. I had no agenda other than to beat the pants off the ZX80 and get the 6560 chip used in something! Charlie and Al may have seen this as a political opportunity, but I was fairly naïve at the time and was just trying to help the company I worked for.” According to Peddle, afterwards Tramiel said, “I wish my team in Moorpark had done this.” Peddle watched helplessly as engineering shifted away from his group. “All of a sudden, they are the new corporate heroes, and they are going to build this thing that ultimately became the VIC-20,” says Peddle. Jerone Guinn, the Texan who previously took over the MOS Technology division, was now attempting to take over the computer systems division. “Jack is approached by [Guinn] about how his team could develop this stuff and he said, ‘We don’t need Moorpark anymore,’ because his team could do all these

things.” After the meeting, Peddle and Tramiel quarreled. “Jack and I had a fight,” recalls Peddle. “Jack was going to show it at the Consumer Electronics Show, and I said, ‘Look, I don’t want us to look stupid because we built a pretty good reputation. I don’t want us to get screwed up by showing something that doesn’t work.’” Peddle argued that the machine needed more features, such as BASIC. “There was a lot of competition back and forth then between us and MOS to win this project,” says Seiler. “Chuck was really lobbying for the fact that it had to have BASIC.” Tramiel replied, “You just have sour grapes because you guys didn’t do it.” Dejected, Peddle departed Pennsylvania back to California on the PET Jet. Tramiel had made an impressively quick decision. Prior to the morning meeting, he had no knowledge of the prototype. “Right after the original prototype was shown to Jack at the managers meeting, he just basically said, ‘Do it,’” recalls Yannes. “I have a feeling he probably came down hard on the California guys, knowing Jack.” After the general managers meeting, MOS Technology had to prepare more prototypes. “When Jack said, ‘Okay, we want to show these at CES in June of 1980,’ we quickly built a few more prototypes to bring to the show,” says Yannes. “Literally we had two weeks to do it.” Before Yannes could begin work on the prototypes, he received an animated phone call from Peddle. “As I recall, Chuck Peddle called from the flight back to California,” says Yannes. “This was back before they had cell phones in airplanes. He wanted to call in with a problem he found in the design with it, which I kind of chuckled at, because this was just a prototype that I threw together. It never was intended to be a production design. He was trying to shoot it down as quickly as he could.” Peddle views the call differently. “I was trying to get him to tell his boss that what he did wasn’t a production thing. It really wasn’t a complete unit.” * * *

After Peddle landed, he broke the news to his Moorpark engineers. “The west coast guys got a little bent out of shape,” recalls Yannes. “It definitely created some issues between the west coast group and MOS Technology.” Bill Seiler felt no ill will towards the MOS Technology group, but he felt some rivalry. “There was a lot of competition between the California group and MOS Technology,” he says. Peddle was not willing to accept defeat. “Chuck shoved something together when he realized the guys at MOS were doing an alternative,” says Spencer. However, Tramiel had asked Peddle to display an Apple killer at the June CES, and that would be impossible without a working 6564 chip. Instead, he resolved to bring a better VIC-based prototype to the show which showcased his vision of a low-cost computer. “Soon after we started, we took the VIC color chip and started playing with it,” recalls Peddle. Peddle asked Bill Seiler if they could design a computer, including BASIC, in two weeks. “That was the way we did things there,” says Seiler. “The milestones were driven by the show.” Seiler told Peddle, “Yeah, we can do this low-cost computer out of the VIC chip.” He was unconcerned about the 22-character limitation of the VIC chip. “I don’t think the Sinclair machine had very many characters on the screen either,” he says. Due to the secrecy maintained by MOS Technology right up until the unveiling, the engineers had little time. Peddle, in turn, would keep their prototype a secret from the rest of Commodore until CES. Even in the late seventies and early eighties, Seiler stood out in a crowd because of his massive, unkempt beard and long, blond ponytail that stretched almost down to his waist. “He had the body of a gymnast—relatively small, like a gymnast would be,” says Peddle. “He always wore cutoffs and sandals.” Yash Terakura, who also wore his hair long, got along well with the engineer when visiting the US from Japan. “Bill Seiler is like me. He’s very independent and he’s different,” says Terakura. “Bill and I were good friends.” While not at Commodore, Seiler enjoyed the unique California culture. “He

swims, surfs, body surfs, wind surfs, and runs,” says Peddle. “I probably got him into running more than anybody.” Seiler sold his surfboard shortly after moving to California. “Surfing was too hard,” he says. “I had to paddle around. I always thought it was too much work so I took up wind surfing.” After becoming proficient at windsurfing, Seiler soon discovered the burgeoning sport of kite boarding. “It’s a little faster to learn than windsurfing, but it’s a lot more dangerous,” he says. “It’s a real trip but it’s pretty violently crazy. The kite is incredibly powerful. It pulls you out of the water.” Unfortunately the sport ended his ocean activities. “I blew my knee tendons out doing that one day, so I gave that up,” he says. “That was the last water sport I did.” He was also a competent programmer and wrote a software package called BASIC-Aid for the PET, which added some useful commands to the existing Commodore BASIC. He also wrote Extramon, a machine language monitor program used by developers to create high-level machine language programs for the PET. Seiler was Moorpark’s top engineer, but it would be impossible to design a new computer in two weeks. Instead, he used a standard PET motherboard as the basis for his computer. “I kludging a VIC-20 chip onto a half of a Commodore PET board,” he says. “I sawed one in half. The top half of the PET board was the video circuit. The bottom half was the processor and ROMs and memory.” After carefully removing the video section of the board, he replaced it with a VIC chip. The result was a messy but functional tangle of multicolored wires. “It was effectively a PET under the table running the screen,” says Peddle. “It was a VIC chip and it was color, but to make it work he just basically hacked together a regular PET.” The engineers knew the prototype was not in a final form. “We’d make some kind of product that would work, but it wasn’t necessarily producible, if you will.” The original PET ROMs only worked with the CRTC chip, not the VIC chip. To make his color display work, the ROM code required changes. Seiler turned to his friend and co-worker, John Feagans.

Outputting color using the PET character set required the engineers to come up with a new concept of meta-characters. “We came up with that crude way of doing color,” he says. “We just invented some color characters you could just print too. So if you wanted a blue horizontal line, you put the blue character and then start printing your graphics characters. If you wanted to suddenly be red, you print the red character or the green character. It was pretty basic, because people could put them in print streams.” The meta-characters resembled inversed PETSCII characters, except when contained within a print statement, they changed the color of the text that followed the meta-character. While Feagans modified the ROM code, Seiler turned his attention to the outward appearance of the computer. He needed a case for the motherboard. As it turned out, a ghost from Commodore’s past would provide the solution. “I digging a calculator out of the dumpster,” says Seiler. “It was a red and white calculator. It had a printer on it, so it was kind of big.” It required all of Seiler’s ingenuity to make the partial PET motherboard fit inside the gutted calculator case. “I had to mechanically get the prototype all jammed in the box,” he recalls. “I folding the boards over and sticking cardboard between them to get them inside this little calculator case. We jammed it in there. It wasn’t real pretty but it wasn’t that ugly either.” He carefully enlarged the keypad opening to accommodate one of the early rubber PET keyboards. The power requirements of the prototype were small, so Seiler used an AC adapter to power his computer. “We used a wall wart [informal term for AC adapter] to power it,” he says. Seiler patched up a hole in the plastic case using a Commodore bumper sticker. “There was a hole in it from the printer part of it.” Seiler calls his prototype a “G-job” because it was the type of fun project engineers created in their garages around Silicon Valley, usually on their own time. “It was like a garage job,” says Seiler. “Chuck mentioned G-jobs are just for fun. It’s exciting when you are doing the designs and brainstorming and coming up with new things.” The engineers wanted to load PET software on the G-job, which contained a cassette port. “We had a library of PET games we could adapt,” says Seiler. “By that time we had the secondary cassette.”

Unfortunately, the PET ROM code that controlled the cassette unit did not work with the G-job. “We were using the VIC clocking system, which was a different frequency from the PET system,” explains Seiler. “The cassette system was really dependent on the clocks.” There was no option other than to work around the clock. “We worked all night to get it done,” recalls Peddle. “We had two days to do it in.” It was a difficult problem, because the original code written by Peddle had no comments. “He had to get his arms around the cassette code,” says Seiler. “I working with him for a couple of hard days and nights. We hacked the cassette routines to work.” On the day before CES, the engineers began converting PET software for the Gjob. Although the PET software was written in the same BASIC language, the engineers needed to convert the games to display properly on the smaller 22column screen. They also inserted the color meta-characters into the print statements to colorize their games. “We modified a bunch of the old Commodore PET games to be color,” says Seiler. “Immediately we had to adapt blackjack because that was one of Chuck’s favorite idle-time programs. He liked blackjack and played it all the time on there. So we had to adapt that and make it color.” Leonard Tramiel was off for the summer break and helped the engineers. “Leonard did a lot of that. He liked doing that stuff,” says Seiler. “He was writing up new games and playing around with that. Anything else that was sitting around we could adapt pretty quickly.” Peddle now had a prototype to display at the CES show. There was now a horse race within Commodore to see which prototype would dominate CES. The PET group at Moorpark had skilled programmers and could incorporate an operating system, but they had little time. The east coast group was ahead, but all they had to show was a colorful demonstration program. The attendees at CES would ultimately decide. [1] Jack later showed his gratitude to Yannes. “We gave Yannes, from Tramiel, a $5,000 check a couple of months later, which was a lot of money for a kid,” recalls Winterble. “What’s really cool is that he used a good deal of it to buy his mother a new kitchen.”

CHAPTER 17

Powerless Struggle 1980

On Sunday, June 15 1980, the day before CES opened its doors, Commodore employees descended on Chicago. Among them was Leonard Tramiel, working the summer for Chuck Peddle on the west coast. Everyone stayed at the elegant Conrad Hilton in downtown Chicago. Bob Yannes attended the show, a rare treat for the rookie engineer. It was up to him to watch over the MicroPET. “The reason I went there was just in case it broke. I was the only one who knew how to fix it.” CES was an expensive show. It cost between $100,000 to set up a smaller booth in June and up to $1,000,000 for larger two-story booths used in January. These booths were so large they even housed offices on the second floor. For this show, Jack Tramiel went low-end, purchasing a single-story booth located at the rear of the convention hall. Every year CES introduced the latest in consumer electronics. This year Mattel introduced the Intellivision, which received intense interest. The big item at the show, however, was still the Atari 2600 VCS. Atari displayed their systems and games in a huge two-story booth located in prime territory on the show floor. Radio Shack was not present, since they sold their products through their own chain of retail stores and thus did not take orders from other distributors. Apple no longer attended, since they relied on specialized computer dealers to sell the Apple II, rather than retail stores. COMDEX was their preferred show. Texas Instruments set up a booth displaying their TI-99/4. The system was

plagued by delays and production problems. A year and a half after TI announced the computer, the fledgling system had fewer than 30 programs available, all made by Texas Instruments. This time, Tramiel held the high ground over his old enemy. In 1980, Commodore’s booth looked almost insignificant compared to others on the floor. The consumer electronics division of Commodore was really the main reason to attend CES. The booth prominently displayed calculators, digital watches, and their newest product, a digital thermostat. Ken Hollandsworth and a small group of salespeople were responsible for this division. The Commodore booth housed product displays and an information desk with brochures and price lists. There were a pair of small meeting rooms, each with a desk and a pair of chairs, used for g new s to distribute Commodore products. On average, Commodore signed a dozen new s each hour. Chris Fish, an inner family member and former VP of the computer systems division, was about to regain Tramiel’s favor. He had been on his punishment tour since late 1979, g up new distributors for Commodore. “Jack sat down with him at CES and basically told him that his purgatory was over,” says Peddle. “[He said] he was going to give Chris European management. That was his comeback into the fold.” Peddle brought some of the cutting-edge technology he and his engineers developed at Moorpark. “One of things we decided to do was a backroom suite of Commodore capabilities,” recalls Peddle. “We did a show-version of the big disk drive, we had some speech going, and we had the CCD [digital camera].” Peddle used the closed-room demonstrations to gauge interest in digital photographs and even primitive video communications. “What we did were these nice little hokey demos, making them look better than they were; knowing that we had to have some more speed in the modems.” The June CES show was a curious venue to display the prototypes. Normally, manufacturers displayed production models and took mass orders from retailers. However, Tramiel wanted to gauge how the market reacted to his low-cost game computer. Commodore did not have a reputation for fun and games in 1980— they were CBM, the business machine company.

The MOS Technology team had tried hard to undercut Moorpark with their secrecy, but Moorpark succeeded in developing a prototype in time for the show. “We went out to the show and the guys from the west coast were there too: Chuck Peddle, Bill Seiler, and John Feagans. They had come up with their own thing using a VIC chip,” recalls Yannes. “I didn’t even know it was going to be there. I’m not sure that Al and Charlie knew either. I think they got that built literally hours before they left for the show.” As the two engineering groups met, they eyed each other’s prototypes discerningly. “The competition got more heated up when MOS got involved at the show,” says Seiler. “We had it in a little back demo room.” Yannes was impressed with the Moorpark prototype. “Theirs was fully functional! Actually, it was very close to what the VIC-20 finally was because it was based on the PET architecture,” he remarks. “The one that the west coast group had done actually had BASIC running on it.” Seiler had his first look at the east coast prototype. “We walked up to it and played with it,” he recalls. “It was basically the same thing, but it was actually nicer looking than mine. It didn’t have any programming language. As far as Chuck was concerned, it wasn’t capable of doing much. You had to plug game cartridges in. It was more like a game [console] than a computer. Chuck’s fundamental premise was that it had to have BASIC so the customer could program it, because that’s what people wanted computers for back then.” The keyboard and case were the primary selling points of the east coast prototype. “The Yannes prototype that he took to CES only played the suite of demo apps that had been written for the VIC chip four years earlier,” says Feagans. “He added the original PET keyboard. The memory space was only 4K because it was originally envisioned as a chip for videogame consoles.” Although the engineers had worked together in the past, there was now a tense atmosphere at the show. “There was definitely animosity there,” says Seiler. “At least we had adapted the PET BASIC to our little chunky calculator prototype and come up with a clever way to do color.” Feagans believes Tramiel preferred the east coast prototype because of the superior presentation of the case and keyboard. “The sizzle sold Jack on the concept,” he says.

Of the two, Peddle obviously favored his team’s prototype. “Yannes had one machine and Seiler had another machine,” he recalls. “We just showed something that was more computer-like than what Yannes was showing. We were trying to make something potentially real.” Although Yannes was impressed with the rival computer, the presentation was less refined. “It didn’t look like the VIC-20 at all,” he says. “I think it had the old PET calculator keyboard. They didn’t have time to get a case together for it, so they used this old Commodore desktop calculator case, and they shoved it in that. I it had a Commodore bumper sticker stuck over the front of it to hide the fact that it was a calculator.” The modified calculator case was tiny, at about five by nine inches and three inches deep. “We had these two machines there, one of which looked pretty good and didn’t do anything and another one which was actually functional but looked pretty horrible,” recalls Yannes. “The other one was clearly built out of parts scavenged from a scrap box somewhere. No one would believe for a second that that was what the final product was going to look like.” Commodore made no formal announcement of the MicroPET, though they printed up a stack of leaflets with a product description. “It wasn’t shown out on the floor,” recalls Yannes. “It was in a little room in one of the booths. I think the only people that were getting to see it were people that were invited. They may have showed it to the press, too.” A large meeting room with tinted glass walls housed the MicroPET prototypes. “I think we had half a dozen there,” recalls Charles Winterble. “They break and they are sabotaged. You’ve got to watch who is touching them.” As the show began, Peddle was still upset with Tramiel and Jerone Guinn. “Only because they were there with something that didn’t work,” says Peddle. “We kept telling everybody, but Guinn was so fucking stupid he was telling Jack it was something else. I had a right to be upset.” Peddle was correct about the incompleteness of the east coast computer. It lacked an interface for a cassette unit, disk drive, or other peripherals. With no operating system, there was no way to type commands. It was not a real computer. For the first time ever, Yannes met Chuck Peddle. “I talked to Chuck Peddle once at that CES show, and he was cordial to me,” recalls Yannes. “I was

worried that he was going to be upset, and for all I know, he was upset with everyone else, but I think he just said, ‘Well, this guy is just an engineer. He’s not actively trying to undermine my empire.’ He was very nice to me.” Outside, scores of people trafficked by, stopping to press their noses up to the smoky glass. What they saw was a four-foot high counter with a row of small computers, each with a small glowing color television set. Yannes stood behind his machines like a proud father. Al Charpentier gave a quick rundown of the VIC features and hoped-for features of the machine and asked people to excuse the faults, promising that the real version would be much better. Tramiel was jubilant that he now had a physical embodiment of the computer he desperately wanted. He stood off to the side, carefully noting the reaction of those who saw the computer and occasionally asking how many they might be interested in ordering. As word spread, curiosity in the MicroPET began to mount. Soon, people began pleading to go inside, and some even got a little testy when the marketing people turned them down. “Everybody seemed to be pretty excited,” recalls Yannes. “The public started clamoring for something like this,” says Seiler. “It looked like a good competitor against the Apple II because it was built inside the keyboard but it was a lot smaller. And the price point was being pushed down. It was a $199 computer.” During lunch, several Commodore employees gathered in the cafeteria. Tramiel broached the subject of giving the machine a better name than MicroPET. In keeping with friendly three letter names, Tramiel suggested calling it the VIC, since the video chip was the main feature of the new computer. VIC would stand for Video Interface Computer. Japanese companies often used CES as a chance to examine the inner workings of new products months or even years before the actual product release. While the meeting room was unattended, a pair of Japanese engineers slipped in to get a closer look at the prototypes. One had a small notepad ready to jot down specifications, and the other held a small screwdriver set. “That actually did happen,” recalls Charles Winterble. “They came in and started taking it apart.” Before they could begin their study, someone noticed movement inside the office

and chased away their rivals before they could open the lid. * * * In all the excitement over the MicroPET, few noticed the absence of the TOI— except for Jack Tramiel. “We missed the show because there was nothing from MOS Technology that we could use,” says Peddle. “He would have expected something that was further along. What he wanted us to have at the show was a product that he was going to take to market.” Peddle acknowledges that some blame for the delay was due to the requirements outlined by the Moorpark engineers. “Their argument might be that the architecture they were working on at that time was wrong and they had to change it. That’s probably true too. But it was them driving the architecture, and we were just writing the specs for that product.” Peddle tried his best to win back the VIC computer with the G-job, but Tramiel viewed it a different way. He came to the show expecting to see the Apple killer he had asked for. “Jack went through and reviewed the things at the show,” recalls Peddle. “It was the first time he had a chance to see all of it himself. He went around and talked to everybody.” The reaction was not what Peddle hoped. “He was basically pissed at me because we weren’t moving fast enough on the Apple killer. Jack turned [Seiler’s G-job] down in an angry way. He said, ‘This isn’t what I told you to do. You’re failing.’ He’s right, we were.” Tramiel believed Peddle was trying to placate him with the G-job. According to Peddle, he said, “Listen, what you’re trying to do with this VIC computer is to cover for the fact that you haven’t done what I told you to do, which is create a product that we can put out this year and wipe out Apple.” “He thought we were screwing off, because we didn’t have it done,” recalls Peddle, who felt he had done everything he could have. “The bottom line is our definition of the product involved us having a chip. If we wanted to be cost competitive, we needed a chip rather than trying to do it in standard stuff. So we built our future around a chip which was delayed a few times. The chip didn’t happen on schedule.” Tramiel refused to acknowledge Peddle’s plea that it would take time to develop

the new technology. “Jack didn’t like excuses,” says Peddle. “He had very little patience with us.” Rather than blame MOS Technology for the delays, he blamed Peddle. “Jack chooses to get angry at whoever he wants to get angry,” says Peddle. “There are a lot of messengers with spears in their bellies. * * * On the final day of the show, Peddle noticed Tramiel courting a new engineer. “Jack was always looking for a hero to replace me,” he says. “It always pissed us off when he did it.” The attempts to replace Peddle seemed like ingratitude on Tramiel’s part, considering the success Peddle brought to Commodore. Peddle believes Tramiel’s apparent enmity towards him stemmed from insecurity. “Jack really did not want to be dependent on anybody,” he says. “The more successful I was with helping him, the more that feeling of dependency increased rather than decreased.” After the meeting in London, Tramiel wanted an engineer who would listen to him. “Jack convinced this guy by the name of Tom Hong to leave Apple and come to work for Commodore,” explains Peddle. “Hong had a really good stock position at Apple, and he walked away from it, because Jack had promised him some serious stock options at Commodore in return for coming over. He was hired to replace me. Tom was given the assignment of getting the VIC-20 ready for production.” After the show, Peddle allowed the Moorpark G-job to fade away. “I told Seiler, ‘Okay, we had a chance to show it, Jack said no. Let’s go ahead and focus on getting the TOI done.’” Their pre-CES marathon engineering was for nothing. It was demoralizing. * * * After CES, Tramiel gave Tom Hong and the engineers at 3330 Scott Boulevard six weeks to create a production model of the VIC computer. After that, he expected an assembly line to start producing them in mass quantities. “After CES, there was a big rush to get it into production,” says Seiler. “Suddenly it became a great big project.”

With the previous PET computers, Peddle and his team handed off a complete design to the production group. There was one problem: the VIC prototype was not a finished design and the production group did not do hardware design. Peddle believes Jerone Guinn led Tramiel to believe the MicroPET was ready for production. “Guinn shows this hacked-up thing that Yannes does and tells Jack it’s ready to go. It wasn’t ready to go. Yannes should have itted that he didn’t know how to make it go. The only one who knew how to make it go was Seiler.” Even worse, Tom Hong wanted to start with Bob Yannes’ prototype as the basis of the production model, rather than Seiler’s more advanced G-job. “The original one from the west coast ended up in my hands. It was a chip demonstrator board,” Robert Russell says derisively. Russell, a rookie engineer, had never designed a computer in his life. He was primarily a software developer. “It kind of was dumped on me,” he recalls. “They were kind of like, ‘Okay, Bob, take this hardware and put the software on it.’ It was like, ‘Argh! You’ve got to be kidding me.’ It had none of the peripheral ports figured out and not enough ROM to sneeze in.” Russell felt abandoned on the new project. He could not expect any help from the east coast engineers because, as far as Jerone Guinn was concerned, his engineers had delivered a complete design and their job was done. Now that they had proven themselves, it was time for them to move onto other projects. “The VIC-20 wouldn’t have happened without the guys on the east coast pushing for the thing and championing it, but they didn’t give us anything we could build a computer out of,” says Russell. Nor could he expect any help from the forsaken Moorpark engineers, a mere 10 minutes away. As far as they were concerned, Tramiel said the design was finished and their orders were to work on the Apple killer. “They didn’t want to have anything to do with the VIC project,” says Russell. “They were busy on the TOI.” Faced with a difficult situation and an inexperienced new manager, Russell went ahead anyway, intending to do the best job he could. “I kind of took it as my project to make sure it got done,” he says. “I was a big staunch er of the VIC.” Tramiel and the other managers were calling the new computer the VIC ever

since CES, but Russell was unenthusiastic about the name, which sounded like a plumber. He made a play on the word VIC, calling their computer by the codename Vixen. “That was the codename really early on,” he says. Despite having Tramiel’s full , Russell had problems finding proper development equipment to program the Vixen. “Moorpark on the west coast was kind of where the top notch, original engineering team went to,” explains Russell. “So they ended up with most of the equipment. I actually had to go over there and debug the VIC-20 stuff.” Regrettably, the Moorpark equipment was in use during the day. “I would work in the office over on Scott Boulevard during most of the day, and then I would drive over there and work most of the night.”

Robert Russell deliberating on the Vixen.

The Vixen would be a physical departure from the familiar design of the PET. The new design was a computer in a keyboard, relying on s to provide a television. It would need a video connector, a cartridge slot for games and other programs, and ports for joysticks, cassette recorders, disk drives and printers. With so many parts of the Vixen left to design, it would be impossible for the programmer to complete the whole system in a month. Russell would need to recruit experienced engineers from within Commodore to complete the redesign. Back at Moorpark, a funny thing happened. Prior to CES, Peddle and Seiler had attempted to win back the VIC project. Now the project came back to them when Russell attempted to enlist help from Commodore’s systems engineer, Bill Seiler. Although Russell worked out of Commodore headquarters at 3330 Scott Boulevard, he often paid visits to Moorpark. “I’d go over there and hang with those guys in the conference room when they would talk about the project, because they wanted me to do stuff on their [TOI] system, too,” he says. Unfortunately, he soon found out that Peddle did not want anything to do with the VIC computer. The project was not under Peddle’s authority, and helping the project only helped his rival, Tom Hong. “Chuck Peddle was not a VIC-20 er at all,” says Russell. “He wanted the TOI, and that’s what all those meetings were about.” If anything, Peddle wanted to see the VIC project fail so Commodore would refocus on his business computers once again. In response, the frustrated young engineer made fun of the TOI during meetings. “There was the acronym first, and then there were people who added meaning to it. I was throwing Tool of Idiots at them all the time and that wasn’t going over real well,” he laughs. “Of course, they called the VIC something totally degrading too.” The marketing savvy of the engineers was not nearly as accurate as Tramiel’s, and they had little confidence in the VIC computer. “Everybody pooh-poohed it

because there was only a 22-character screen,” recalls Peddle. “A bunch of people argued that nobody would buy it.” Peddle himself opposed the VIC computer. “Chuck was an unwilling participant when the VIC-20 started rolling,” says Feagans. Russell was primarily a software engineer, leaving him to beg and cajole Seiler into furthering the hardware development of Vixen. “Bill Seiler was really a top notch hardware guy who really made all the original hardware hang together,” recalls Russell. “I needed him to help me keep that alive, because I pretty much got abandoned on the VIC.” Unfortunately, Seiler could spare no time for the Vixen. Russell suspected the engineers were still stinging from Tramiel’s rejection of the G-job. “They obviously wanted to kill the VIC project. I kind of had to choose sides,” he says. As the lone west coast VIC er (aside from Tramiel himself), Russell began to receive favoritism from Tramiel. “I was working on the VIC-20, standing in the middle of the engineering cube area,” he recalls. “I had cubes with the VIC stuff out in front, and [Jack] came into there and said, ‘I need someone to carry my bag to the airport to go to Vegas.’ And he pointed at me! I carried this little duffel bag that weighed maybe three pounds or something.” Although Russell had planned a day of engineering, he soon found himself aboard the PET Jet bound for the entertainment capital of the world. Jim Finke, a potential Commodore recruit, was aboard. “We flew out with the guy who later became the Commodore president, and [Jack] was doing the interview and carrying on with that guy in the PET Jet to Vegas,” recalls Russell. Making deals seemed to be in Tramiel’s blood, even when he spoke to his engineers. During the flight, Tramiel offered Russell the same deal he had offered Peddle years earlier for the PET. “He offered me a buck on each VIC-20 if I didn’t take salary,” he recalls. The deal had the potential to make Russell a multimillionaire. “He was trying to see what he could get,” says Russell. “He was always doing that. Serious or not serious, it was one of those things that you never called Jack on. He would shake your hand, and it would be a done deal.” Russell was intrigued by the offer, but without a salary it would be impossible to

make car and house payments. “At the time, I was like, ‘How can I live?’ So I turned it down because I didn’t think it was going to pay off.” Russell’s refusal demonstrated the low expectations for the sales performance of the computer. Although the engineer was one of the main ers of the VIC computer, even he did not think sales would exceed 50,000 in a year. Upon arrival in Las Vegas, Tramiel gave Russell permission to tour for as long as he wished. “He gave me a handful of casino chips from the Sands Casino. He told me, ‘Tell the pilot when you want to go back, and have a good time.’” It was a surreal experience for the young engineer from Iowa. To this day, Russell is unsure why Tramiel wanted him along for the trip. “Who knows? I think Jack just wanted to be nice to me.” After gambling his chips away, Russell rode a cab back to the airport and flew back to San Jose alone on the PET Jet. * * * After the flight to Las Vegas, it was clear Russell held favor with Tramiel. “People with access to Jack could do quite a lot,” explains engineer Andy Finkel. “We had an old saying at Commodore at the time that various people had this card they kept in their wallet, called a Jack card. Bob had one of those.” In most card games, the Jack is never the top card. At Commodore, the Jack dominated all the other “suits.” With no one in the company prepared to dispute Tramiel, the figurative Jack card was a powerful weapon. “You could be talking to somebody and having an argument or discussion, and they’d pull out their Jack card and throw down on the table and obviously they’d win the argument immediately.” Bob Russell discovered how to use his Jack card to make things happen. “Once people knew that you were a Jack favorite, you could do pretty much anything you wanted,” he explains. Russell used his Jack card to give the Vixen project the momentum it needed. “That was when I rode Jack’s dissention,” he says. “I conned Seiler into helping me rebuild stuff and design things like the memory add-in board, so you could actually have enough memory to do something on it.” When Russell attended his next Moorpark meeting, he used his Jack card to gain more resources for the project. “I had meetings, and I said, ‘Well, Jack told me to

do it.’ I played the Jack card all the time.” “The US engineers were pulled kicking and screaming into the VIC,” recalls Tomczyk. “They kind of came grudgingly along.” Russell later discovered that the Jack card gave him the power to make nearmanagerial decisions. “I could bullshit my way through anything, because people were scared to go up against Jack. They were like, ‘Yes, sir! Open the doors.’ Jack would back you up, even if he hadn’t told you to do it.” Seiler and Feagans helped incorporate several features of the rejected G-job into the Vixen. “All us engineers were driven into the corner at Moorpark and told we have to finish this thing in a month and a half and get it into production because Jack is screaming that we have to build a bunch of these,” recalls Seiler. Unfortunately, TOI development began to suffer. “The TOI was sort of pushed aside.” It frustrated Peddle to see his engineers diverted to the VIC project because his future at Commodore depended on delivering the TOI to Tramiel. Seiler created a memory map, which allowed programmers to access the various functions of the VIC chip. Feagans adapted the BASIC programming language to the Vixen. Seiler also instructed Russell on how to implement his color metacharacters to produce colored text. “In addition to making BASIC and OS run, the text editor needed to be extended to accept color codes,” says Feagans. “Bob Russell did this part.” Peddle could not hide his disappointment as the VIC computer began to adapt many of the features from the PET. “He sure was annoyed when Bill mapped the VIC into the memory map and I had BASIC and the OS running,” says Feagans. The VIC-20 was a game computer, but it only contained a single joystick port. One joystick was limiting, compared to the Atari 400 computer, which had four joystick ports. However, the oversight was understandable since most current games like Asteroids, Sub-Hunt, Night Driver, and Pong used paddles. Pac-Man, which Bally-Midway had not yet released in North America, popularized the joystick. The single port ed two analog paddles. “The paddles were on the VIC chip,” says Yannes. “It could either an X-Y analog joystick or two conventional rotary knobs.” Curiously, Commodore never released an analog joystick for any of their computers.

Russell copied the same connector used on the Atari 2600: a standard D-shell connector with nine pins. The connector was not proprietary to Atari, which meant Commodore could copy it without fear of a lawsuit. Other computer makers, such as Sinclair and a line of Japanese computers, also used the D-shell connector, establishing a standard for joysticks. Since the PET did not have a cartridge port, Seiler had to design one from scratch. Vixen cartridges were easy to distinguish from Atari 2600 cartridges because of their size. Since the Vixen would use the slot for memory expansion as well as games, the slot required more address lines. This resulted in massive game cartridges compared to other systems. The VIC 6560 chip was capable of high-resolution graphics up to 192 by 200 pixels. This required a lot of memory, so high-resolution was only available if the purchased a memory expansion unit. Without it, there was not enough memory to contemplate a bit-mapped screen. Since Commodore was not supplying a monitor with the Vixen, the engineers had to be wary of the different types of televisions that existed. To accommodate the different border sizes around various television sets, the engineers implemented a thick border, leaving a usable display of 176 x 184 pixels. As a bonus, s could change the color of the border to one of eight different colors. Televisions standards were different around the globe. North America and Japan used the NTSC standard, which the VIC 6560 chip used. European television sets used the PAL standard, so MOS Technology created an alternate VIC chip called the VIC 6561. Tramiel loved the different television standards because North American Vixen computers would not work on European television sets. This meant he could charge different prices for computers and software in different regions with no risk of importers taking advantage of the price difference, as they had with the PET. Russell had almost complete freedom working on the Vixen. “Most of the time there was no real head of engineering. There were just engineers who had the responsibility to get things done. That was how it worked with the VIC-20.” After a month, a production computer emerged. “We had really tight time schedules and we pretty much did that to our time schedule,” says Russell. The result was closer to Seiler’s prototype than the east coast prototype. “There

should be no question in anyone’s mind that the primary architect behind the VIC-20 was Bill Seiler,” says Peddle. “We didn’t have to go back and invent all kinds of new things except for the I/O stuff. We were just basically putting it all together to make it work.” Robert Russell agrees. “Seiler had most of the hand in what was really the design,” says Russell. “He was the guy who made sure there was production hardware.” Bob Yannes affirms that the final design of the VIC-20 was nothing like his prototype. “If you look at the VIC-20, it is a PET, except instead of the 6845 video controller, it has the VIC chip in it,” he says. “It has the same I/O structure and the same processor, and the kernel code that was running in there was the PET operating system kernel.” Tramiel had hired Tom Hong from Apple to make the VIC computer happen, but Chuck Peddle believes Hong had very little to do with the project. “He was hired as a stud from Apple who was going to come in and make the VIC-20 happen,” says Peddle. In reality, Peddle’s own engineers made the VIC-20 happen. “The reality was that Hong was overrated. Jack got sold on a bill of goods, which he often did.” With the marathon engineering to complete the Vixen, Seiler earned his yearly pilgrimage to Siggraph. “That was my favorite thing,” he says. “Siggraph was the IEEE national computer graphics show. I used to get the company to send me to Siggraph every year.” The engineer travelled to Seattle, Washington for the week of July 14 to attend the show. “Siggraph was always fun because there were two or three nights that they had these videos of all the supercomputer graphics that were generated that year and they had winners and everything,” he says. “I learned a lot about what was going on with ray tracing and neat things were starting to show up that are now buried in almost all the graphics software and hardware.” * * * Throughout the redesign of the Vixen, Tramiel sensed his Moorpark engineers were not fully ing the computer. “Chuck Peddle was resisting this a lot and he was in charge of engineering,” says Tomczyk. “When the people in charge of engineering are resisting something, you give it to somebody who

wants to do it.” Tramiel invited Tony Tokai and Yash Terakura to the west coast for a discussion. “Jack called me so Tony Tokai and I came over to the United States,” says Terakura. “We had discussions with Jack and Chuck and everybody.” The focus of the discussion was how best to complete the VIC computer and take it to market. “Ideas just kind of came up while we were there talking,” says Terakura. “Jack wanted to make a computer for Japan first and make sure it was a viable, feasible product in Japan. He believed that if it sold in Japan, it could sell anywhere.” It was a bold strategy to bring the fight to the Japanese rather than waiting for their computers to invade America. “He wanted to prove that a US company could survive in Japan,” explains Terakura. “We were making calculators in those days and not even getting 10% of the market in Japan. So Jack always wanted to be number one. He said, ‘If we can be number one in Japan, we can be number one in the world.’ That was his philosophy.” Peddle wanted his engineers refocused on the TOI computer, instead of the constant distraction of the VIC machine eating away at their time. Michael Tomczyk recalls the meeting. “They really didn’t want to do this computer. In fact, Japan said, ‘If you don’t do it we’ll do it.’” The next words spoken by Tramiel were precisely what Peddle was hoping to hear. According to Terakura, Tramiel asked, “Why don’t you take this one back and finish it up in Japan?” With the VIC computer relocated to Japan, his engineers could get back to working full time on the TOI. “I said, ‘Okay, fine.’ Then I talked to Tony and he said, ‘Can you do it?’ And I said, ‘Yeah, I think so.’” Bob Yannes re the day Commodore US found out what Tramiel’s famous statement really meant. “People were saying the Japanese are going to take over this market just like they’ve taken over every consumer electronics market,” he recalls. “Jack’s quote was, ‘We will become the Japanese.’ No one quite really understood what that meant, but it sounded good, and you really didn’t question Jack anyhow. But I guess what he really meant was that we were going to use the Japanese to produce the VIC-20.”

The move to Japan surprised Bill Seiler. In the past, almost every aspect of the PET was designed in North America. “Once we did the prototype, somehow it got shipped to Commodore Japan,” he says. In July 1980, Terakura and Tokai loaded a prototype machine onto the PET Jet and returned home. “I took that one back to Japan and we designed the thing,” says Terakura. The downtown Tokyo offices at 7-3-38 Akasaka were modest compared to the US Commodore headquarters. “In Tokyo, we were on the fifth floor of the Yamakatsu power company,” says Terakura. “We were right next to the Canadian Embassy and there were some Canadian Embassy offices in the same building too.” At the front of the offices was a 400 square foot space for displaying Commodore machines. The small offices housed ten staff total, including sales, marketing, istration and engineering. “A lot of people worked 24/7,” says Terakura. “They worked at home or stayed at the office on weekends.” “It was just a small office. It wasn’t much,” says Terakura. “I think the biggest part was my engineering section, because we had so much junk so we had to have our own section. We had an open area we were using. The production was done in another part of Japan.” Engineering was the focus of Commodore Japan. “It’s a little different from a US environment,” says Terakura. “It’s just one large area and you have a bunch of desks and tables, pretty much like an open lab. We didn’t have any CAD systems in those days, so we had to have drafting tables. And we had two mechanical engineers working on the deg.” Like many of Commodore’s international offices, the premises were unimpressive. “No Commodore facilities looked extravagant or elegant,” says Tomczyk. “They were all kind of bare bones offices. You would be surprised where some of these wonderful inventions were developed. It was like wherever you went in Commodore there were sloppy benches with wires and circuit boards and monitors strewn all over the place.”

An unidentified Commodore engineer working at a drafting table.

Once back in Japan, Terakura opened the machine and pored over the circuit board, documenting the layout of circuits. The production design was literally a redesign of the prototype, which they could then produce using assembly line equipment. “They took the PC board from the west coast and redesigned it for their insertion equipment,” says Yannes. With the motherboard already completed by Seiler, the Japanese merely redesigned the layout of the circuit board. “The basic schematics and idea were done by Bill Seiler already,” says Terakura. “So all we had to do was just tweak it and put it together in the housings and all the other stuff—the power supplies and everything else.” With the computer-in-a-keyboard concept, there was no room for an internal power supply. This meant engineers had to design an external power supply. The power brick made the Vixen look less refined, which is why marketers rarely showed the computer with the cables and power supply in the same photograph. Heat was a major concern in the Vixen design. As Russell recalls, “The original VIC-20 prototype would get so hot that the heat sink would rot off the motherboard, so we put in a huge, over-designed regulator in that.” For the next month, the Japanese designers pieced together a marketable product. “The actual hardware deg and all the PCB design, and housing design, and all the other stuff was done in Japan,” says Terakura. The VIC-1001, as the Japanese named it, had a unique design. “All the plastic design, all the case work, and all the molds that had to be cut were being done out of Japan,” says Russell. “Whether your product came out or not was dependent on Tokai.” Unlike the metal PET case, the Japanese created the case using injection-molded plastic. Commodore Japan had a small drafting group for their designs. “They had a big design bureau inside where they had the drafting people,” says Russell.

A designer named Nishimura was responsible for the case designs of calculators and now computers. “Nishimura was a mechanical designer,” recalls Terakura. “He was the one who designed the housings and stuff and did all the toolings for the plastic housing.” Nishimura, who was located in California at the time, communicated with Terakura to figure out where the various ports would be located around the computer. “We did discuss a lot about where we put the plugs and whatnot,” recalls Terakura. The Japanese designer produced a memorable case. Other computers, with their right angles and sharp corners, seemed not quite as friendly. The VIC-1001 case, with smooth rounded edges, looked like a cozy pillow with a microprocessor inside. Perhaps the most consequential change from the original prototype was the use of a standard keyboard, rather than the rubberized chiclet keyboard. Surprisingly, the decision for a more expensive component came directly from the top. “Jack Tramiel insisted on using a regular keyboard instead of membrane,” says Terakura. “People liked the regular keyboard. It gave the VIC an expensive look.” The keyboard made the VIC computer stand apart from the competition. “The smartest thing—and I think this was Jack’s idea—was putting a full-sized keyboard on it,” says Yannes. “What you had on the market in the price range at the time were things like the TI-99/4, which had a calculator keyboard, and the Atari 400, which had a membrane keyboard. I don’t think those companies realized that people didn’t take those products seriously, particularly because of that keyboard.” Chuck Peddle believes the keyboard decision was motivated by the earlier blunder with the PET 2001. “That’s why the VIC-20 is built around a regular keyboard,” he says. “We knew that you needed a regular keyboard because we got punished by the market until we put one into our product.” Although the VIC computer was not very powerful, the keyboard made it look legitimate. “Its graphics weren’t as good, it didn’t have as many characters per line, it had much less memory than either of those machines, but the real keyboard on it just made it look like a real computer whereas the others looked

like toys,” says Yannes. The keyboard came directly from the business PET computer. “The business PET had a full-sized keyboard,” says Seiler. “They adapted that to the VIC-20. The Japanese guys put all that together.” Terakura also broke with Japanese tradition by including a sophisticated manual for the computer. “It was the first time in Japan anybody made a very intensive manual for computers,” he says. “People were surprised to see the thick manual with complete schematics. Until then, manuals were written by either marketing or sales people. But I wrote the manual from the perspective of an engineer. Until then, schematics were considered proprietary information and trade secrets. I released the schematics because all the main chips used were from Commodore MOS. The VIC chip was not in the market.” It is hard to underestimate Terakura’s contributions to the final success of the computer. “Bill made the basic concept of VIC and early proto, then I finished it up and made the marketable machine which was VIC-1001 in Japan,” he says. * * * Jack Tramiel hoped Chuck Peddle and the rest of Commodore would the VIC-20 when it became clear he wanted to conquer the consumer market. Unfortunately, most of his executives and engineers did not. The disagreement was a festering wound and continued to worsen. A power struggle was taking place. While Yash Terakura developed the VIC-1001 in Japan, Tramiel and Peddle’s relationship began to deteriorate rapidly. Tramiel’s new saying was, “We will make computers for the masses, not the classes.” However, the PET computer had evolved into high-cost CBM business machines. “In Jack’s mind, and it’s true, we drifted away from the original entry level concept,” says Peddle. “The business computer stuff was never his thing.” Tramiel’s son Leonard saw both sides of the conflict. He was friends with Peddle, who was something of a mentor to him, but he also clearly understood the agony Tramiel endured from the rebellion. “I’m not sure that [Peddle] liked taking instructions very well. He had this idea of where things should go and how they should be done,” recalls Leonard. “Chuck was a pretty self centered

person—not in negative way, but he just had a strong image of himself.” Leonard understood why Peddle was pushing for a more advanced business computer. “Chuck Peddle was not a go-with-the-flow kind of guy, but you can’t really develop anything new if you are a go with the flow kind of guy,” he explains. Perhaps the biggest obstacle was the language barrier between engineers and management. Although Peddle was part marketing person and part engineer, he seemed to have an inability to explain his ideas to Tramiel. “Other people didn’t agree, and he was not very good at convincing them,” recalls Leonard. Peddle its he did not try hard enough to convince Tramiel. “We had a pretty good team that could see ahead architecturally and we didn’t really explain this very much at Commodore,” he says. Instead, Peddle focused his conversations on executives who already agreed with his business concept. “We basically just went back to the management, and management at that time was Dick Sanford, Chris [Fish] and these other guys more than it was Jack.” Leaving Tramiel out of the discussions was a mistake. Peddle’s position as lead engineer was beginning to erode. “He was trying to push Chuck out of engineering or something,” says Seiler. “We had gotten the VIC off the ground and he sort of squeezed Chuck out.” According to Peddle, Tramiel came to him to discuss the low-cost computer. “You came to me to make a cheap machine for the masses,” said Tramiel. “What you’ve done is started doing a machine for the classes.” Peddle replied, “No, that’s not really true. What we’re doing is we’re letting the architecture take us where it should take us.” Unfortunately, a technology barrier prevented Tramiel from agreeing with Peddle. “That wouldn’t have been something he would have even understood,” comments Peddle. “He was a smart guy, but computer architecture would have been beyond him.” Tramiel was convinced his low-cost computer was the future of Commodore. “He had fallen in love with the Sinclair stuff and he just wasn’t listening,” laments Peddle.

Peddle continued pushing for two separate computer divisions. “It was obvious that the company needed to go in two totally independent directions,” says Peddle. “Commodore could have absolutely been a dominant world player in the high-end market (which ultimately came to be the PC market) and the game oriented fun computer market.” Bill Seiler was glad to work out of Moorpark, away from Commodore headquarters where Tramiel and Peddle had their discussions. “There was too much ranting and raving when Chuck went into Jack’s office,” he recalls. “I usually just kept my head low and tried to stay out of the politics.” Dick Sanford, who considered himself a friend of Peddle, was often present during meetings. “I do recall that every time Chuck didn’t get his way he’d get upset,” says Sanford. “He didn’t have free reign. He wanted to do research all day long. So it’s just a difference of opinion between what Jack was trying to accomplish and what Chuck was trying to accomplish. I suspect that Jack had private conversations with him. Obviously, I don’t know what was said in those conversations.”

Demonstrating Moorpark’s TOI prototype.

Peddle had caused a rebellion, turning Tramiel’s executives against him. He was threatening his position at Commodore, and Tramiel knew he had to do something. It was not difficult for him to identify the managers he would have to deal with. “Sanford is a problem. Fish is a problem. I’m the biggest problem,” says Peddle. Tramiel told Peddle to stop work on the business machine. The engineer contemptuously ignored his employer’s instructions. “I didn’t stop the work,” he reveals. “It was an R & D group.” With Commodore’s appetite for engineers increasing, Peddle developed a second source of engineers by creating a small research lab at the California Polytechnic State University in San Luis Obispo. “We set that up because we were looking to get people out of the colleges there like we had at Iowa State,” says Peddle. He was continuing his plan of creating a company culture similar to HewlettPackard. Unfortunately, Tramiel did not share his vision.

CHAPTER 18

Big in Japan 1980

When Tony Tokai returned from Japan with the prototype, the Moorpark engineers were astounded by the sleek design. “I going into Jack’s office on Scott Boulevard and seeing the first one,” marvels Seiler. “The prototype came back and we all looked at it and said, ‘Wow! It looks like a real computer. That looks really cool.’ It was like a fat keyboard that had the whole computer in it. Everyone liked it because it had a full-sized keyboard.” The new design prompted Seiler to take a harder look at this earlier design. “We suddenly said, ‘Holy shit, we’ve got to fix this,’” he recalls. “We really threw out the whole thing that was in the prototype because it wasn’t cheap enough.” The engineers were on a tight timeline. “We had two months before they wanted to start shipping these things,” he says. Memory was the most important consideration in order to meet Tramiel’s mandate of a low-cost computer. Al Charpentier designed the VIC chip to use relatively expensive SRAM. “We had to solve the problem of using static memory,” explains Seiler. “It was fairly expensive. So we came up with some weird scheme where there was just enough static memory in it to come up and run a little program.” To keep costs down, the engineers settled on only five kilobytes of total memory. “I think you only had for the 2K of usable space,” says Seiler. “There was a real kludgy way I had to work the memory out so that the cartridge worked. The memory space was not as nice as it should have been.”

MOS Technology had a surplus of one-kilobyte RAM chips, which engineers no longer used in PET computers. Tramiel made sure his designers used these onekilobyte RAM chips to clear out inventory. Robert Russell remained an active engineer on the project, developing the ROM code. “Software was done in the US side,” says Yash Terakura. Part of Russell’s devotion to the project was because of the new experience of working with a foreign culture. “I was enjoying working with the Japanese people,” he says. The engineers carried over many of the goals from the TOI project into the redesign of the VIC computer, among them the need for a disk drive. “We knew we needed a floppy because Apple already had a floppy drive,” says Seiler. In order to connect a disk drive to the computer, the engineers designed a serial port, which also allowed printers and other peripherals. The concept was similar to the IEEE-488 connector of the PET but less expensive. The serial port originated with the TOI project. “It pretty much had the PET expansion bus on the back but we knew we couldn’t put the IEEE-488 on it,” says Seiler. Chuck Peddle hired a former Xerox PARC engineer named Robert Metcalfe to help his engineers figure out the port. “He organized the meetings at Moorpark with Metcalfe because we were trying to design an expansion bus,” says Seiler. “We knew we couldn’t put the expensive HP IEEE bus in it so we tried to come up with something cheaper.” “That summer in San Jose, Robert Metcalfe was a consultant who worked for Commodore before he founded 3Com,” says Feagans. “He was hired as a consultant to look at networking.” Metcalfe pitched an idea he was currently working on. “Metcalfe came over and explained the whole Ethernet thing,” says Seiler, referring to the technology for local area networks. “I him drawing it all over the blackboard. We all sat there and said, ‘Man, this is a super-great idea!’ And then he started explaining how expensive it was to do it with the hardware that was available. This wasn’t going to fit it in this cheap computer.” Although the engineers ended up rejecting Metcalfe’s method of networking computing devices together, their expansion port design was influenced by him.

“He convinced us of the idea of serial,” says Seiler. “So we kind of made a serial version of the IEEE-488 for the VIC-20. The disk guys ran off and started working on a simple single floppy board. And it was serial. It was quite a bit slower than the IEEE because it sent all the bits and it was all done pretty much in software again.” Metcalfe also explained the groundbreaking projects he had witnessed at Xerox. “He told of all the marvelous things going on at Xerox PARC,” says Feagans. “He influenced me to the importance of networking for personal computers.” Feagans was also influenced by Metcalfe’s description of graphical interfaces, in which s control the computer by clicking visual icons, rather than typing commands. The actual designer of the TOI serial interface was Moorpark engineer Glen Stark. Peddle recalls, “He was the guy that defined the interface that was patented for Commodore that allowed it to fake the IEEE interface over a single wire cable.” “The serial bus was meant to be much lower cost,” explains Russell. “On the IEEE, we always needed a second processor to run the interface on the peripherals. On the serial bus, the whole idea was to do it with one processor that could run the interface and the drive.” As a result, Commodore could manufacture cheaper disk drives and printers. Although the serial interface cost less, it was slower. “Obviously IEEE was a heck of a lot faster because after all, it has this big parallel stream,” says Russell. The new interface was also smaller, reducing the size of the computer. The IEEE-488 connector was 5.5 centimeters in width, versus a 1.5-centimeter diameter for the round serial connector. “The [IEEE-488] cables alone cost a fortune, so that was a big reason we didn’t want to do that anymore,” says Russell.[1] Seiler also integrated the proprietary port from the PET computer. This expansion port allowed companies and hardware hackers to create additional peripherals for the VIC computer. Although Glen Stark came up with the serial interface concept and design, Russell implemented the code. “He was the guy that helped Glen and got things done,” says Peddle.

According to Russell, he was given the objectives of the serial interface and told to make it happen. “I was left as the implementer,” says Russell. “I had to figure out how to make the whole serial port system work.” Once he finished coding the serial interface for the TOI, Russell decided to include it in the VIC computer. “That’s something I did on my own. I wanted to have serial communications, so I wrote a software serial interface for the VIC. If I hadn’t have done it, nobody would have cared, and it wouldn’t have been in the computer.” Russell also had to modularize the cassette code written by Peddle years earlier. “About eight people went back and tried to document it,” says Peddle. Bill Seiler had previously examined the code and given up. “Chuck never put any comments in his assembly code,” says Seiler. “Assembly code without comments is scary. Chuck said he liked to look back over it without the comments and refigure it out. I think he was just lazy and didn’t want to type in comments.” All had given up on the task until Russell put his mind to it. This earned him the respect of his fellow engineers by taking on difficult projects with no promise of glory at the end. “Russell was a great guy,” says Seiler. “The code was really convoluted—a lot of interrupts going on—and it was hard to figure out what was going on.” Russell went in blind since the code was undocumented. “This cassette code had this bizarre frequency domain nonsense without comments,” he recalls.

Robert Russell (left) and Bill Seiler working in the Moorpark labs.

Eventually, Russell had to convince Peddle to help him with the code. “Chuck’s cassette code from the PET was so convoluted only he could make it work,” says Feagans. “Chuck Peddle had to give me a lot of his time to help document the cassette code because nobody had a clue how that worked,” says Russell. “I sitting up in the corporate side in an office with Chuck Peddle, who was over at Moorpark by that time, going over the code with him and him trying to what the heck he had done.” Russell liked working with Peddle, despite his ambivalence towards the Vixen. “He impressed me, because he had personally done things like the cassette code on the PET even when he was leading the project. I really ired his work ethic.” In the end, Russell was able to reduce the code size and make even more room for other functions. “He’s the guy who figured out the cassette code finally,” says Seiler. “He found out about a third of it wasn’t even doing anything and he threw it away. It made more room in the ROMs.” Video was the most important aspect of the VIC code. “I had one guy working with me in the [Scott Boulevard] building working on the display code,” says Russell. “Some of the toughest stuff that had to be cleaned up was the display code. That was kind of a modular piece, because we had already split it off to do an 80-column version [for the PET].” The VIC character set was stored in a four-kilobyte ROM chip. Each character was eight pixels wide and eight pixels high. Perhaps the biggest improvement of the Vixen over the PET was the robustness of the character graphics. On the PET, programmers were limited to building their graphics from the PETSCII character set. With the Vixen, programmers could now modify and redefine each character. For example, the letter A could be redefined as a stick man. The Vixen would also benefit from advancements to the TOI operating system, namely the kernel concept originated months earlier. “John Feagans was the

architect of what became the kernel concept, which is the separation of the actual operating kernel versus the BASIC interpreter,” explains Russell. Although Feagans originated the kernel design, Russell ended up programming the kernel code for the TOI and Vixen. “He wanted to move on to other things, so I picked it up,” says Russell. One of the least friendly aspects of the VIC-20 was the operating system. The prototype sent to Japan used BASIC 2.0, with decidedly unfriendly commands. In order to list the contents of a disk, s had to type LOAD“$”, 8 and then type LIST. On other operating systems, such as /M, s merely typed DIR to see a listing. However, Commodore engineers predicted most s would use cartridges and cassette drives, so there was no pressing need for commands to copy, delete, or list files. Although the keyboard design came from Japan, the North American engineers developed the character set. “The original keyboard layouts were done in Japanese,” says Russell. To design the character set, Commodore Japan sent a worker to North America to help Robert Russell. “I worked with a Japanese girl to do the keyboard layouts because I don’t know anything about creating Japanese symbols,” says Russell. Russell produced a special Japanese ROM chip capable of displaying Japanese Kanji characters. It could also display regular upper case alphanumeric characters, making it one of only two Japanese computers capable of displaying both Kanji and the western alphabet. Eventually, Tramiel sensed the Vixen was good enough for the marketplace and he put a stop to development so it could enter production. “We can always do one more thing,” says Seiler. “Engineers will keep engineering. I think Chuck called it creeping elegance. A lot of it doesn’t ever pan out to be worth the money or effort we put in.” Even though the design was more or less complete, the project would take additional time. “We had to do all the testing and everything,” says Terakura. Chuck Peddle attributes much of the success of the project to Terakura. “Yashi Terakura deserves enormous amounts of credit for the work he did,” he says. “Yash was the one who made the keyboard stuff work on time. Success has

many fathers, and in this case, it did have many fathers.” * * * Robert Russell enjoyed favoritism from Tramiel because of his for the Vixen, but he was not immune from his boss’s wrath. With the Vixen prototype done, Tramiel wanted to show the computer to his investors. “I was supposed to bring prototypes out for the meeting where the TOI and the VIC were shown to the analysts in New York City at the Four Seasons [Hotel],” says Russell. A poor presentation could cause investors to lose confidence in Commodore. “This was a very important thing,” says Russell, who intended to ride with the prototype on the commercial airliner. Unfortunately, one of his managers gave him different instructions. “My managers on the west coast wouldn’t let me hand-carry the stuff on. It got put in luggage, and of course the box got lost. It didn’t show up.” Tramiel gave an embarrassing presentation without the prototypes. “I went with them to the introduction to the analysts,” says Russell. Afterwards, things seemed normal. “I got a beautiful dinner at the Tavern on the Green in New York with the engineers and Jack.” Russell noted the tight bond between Tramiel and his wife, Helen. “She was a very motherly type of person,” he says. According to Russell, Tramiel would playfully say, ‘Mama, cut my food for me,’ and Helen would slice his food and feed him. “It was kind of embarrassing as a twenty-one year old but still kind of funny.” After the meeting, everyone traveled from New York back to Pennsylvania and MOS Technology. Russell was about to receive his first Jack Attack. “It actually happened in the president of MOS [Technology’s] office [Jerone Guinn],” says Russell. “Jack called me into the office and chewed me out to within an inch of my life.” Although Russell was just following orders, he felt the full impact of Tramiel’s displeasure. “He was just very much angry, very much frustrated, and wanted to know why I had screwed up.” Decades later, Russell still seems incredulous about the event. “It was just because I followed my boss’s instructions. It wasn’t under my control.”

Despite explaining that his manager would not allow him to carry on the prototypes, Tramiel felt Russell had erred in his judgment. He told him, “Why did I put you on a jet to fly you out here? I didn’t want to see you, I wanted the product here, and I wanted it to be working!” Jack Attacks were intensely personal, and every word came straight from Tramiel’s heart. “You’re just kind of like, wow, sensory overload,” says Russell. “The guy is really calling you an idiot and who knows what else. He could definitely make you feel like you were a pretty small pimple on his butt.” To invoke a Jack Attack meant an employee had truly caused the company damage, and Tramiel took that personally. The 22-year-old Russell feared the worst. “I thought I was getting along and going places in this company, and here it looks like my career is over,” he recalls. Russell listened to Tramiel’s words and began to make sense of it. “To Jack, it was like, ‘You don’t follow the stupid thing your boss is telling you to do. You do what is right.’ Use your own common sense is what he was really saying.” After the verbal thrashing ended, Tramiel asked, “Do you have anything to say?” Russell replied meekly, “Yes, sir. I understand completely. I will never ever let you down like that again.” The effect of a Jack Attack was electric. “If you understood the concept you could walk out his office alive, which I did,” says Russell. “I’ve suffered a Jack Attack or two, but I always said I understood and never made the same mistake twice. He had no patience for fools.” “[Jerone Guinn’s] secretary was outside and couldn’t believe I walked out of there alive,” recalls Russell. “She said, ‘I was so sure you were fired.’” The unfortunate accident had one good effect for the engineers. Afterwards, Tramiel allowed them to use the PET Jet to transport hardware. “It was convenient for us to go to shows because we could load it up with all our equipment and go to or Switzerland with it,” says Charles Winterble. “This way everything was with us.” Russell looks back on his Jack Attack with humor. “It’s a badge of honor,

especially if you were one of the survivors.” * * * Commodore’s marketing department had little say in the development of the VIC-20. “In my opinion, they were better off for it,” says Leonard Tramiel. “I never understood why an advertising person should be involved in product development. They don’t know how products work, they don’t know what can be done and can’t be done, so I don’t know how they can design new products.” After Terakura departed with the prototype, Tramiel only allowed his personal assistant, Michael Tomczyk, to communicate with Japan because he felt too much discussion would slow down production. “Jack told me at beginning that all decisions having to do with this computer had to be run through me,” he recalls. “He refused to make me the product manager per se, but he said, ‘I’ve told everyone associated with this computer that all decisions relating to it have to be endorsed by you, so you’re going to have to do this by persuasion.’” This indirectly gave Tomczyk a degree of control over the project. Tomczyk had been hired months earlier, in April, and now he held an important position on one of Commodore’s newest products. “Mike decided that he wanted to be in charge of this new computer that Commodore was sort of wanting to put out but wasn’t sure if it really wanted to put out because it only had 22 columns instead of at least 40,” recalls Commodore engineer Andy Finkel. “Mike decided that the price point was more important and he could make the VIC-20 a success.” “I at one point Mike Tomczyk came in on the VIC-20 and said, ‘Okay, well I’m in charge of marketing this thing,’” recalls Yannes. “We had never heard of him before or seen him before, and we were like, ‘Okay.’” Yannes appreciated Tomczyk’s on the VIC project. “He was very enthusiastic.” Tomczyk communicated using a Telex machine, which was similar to a FAX machine except it handled only text. Tomczyk sent a flurry of costly telexes to Japan, which caught Tramiel’s attention. One day, as Tomczyk was standing in the hall, he was surprised to feel someone kick him from behind. When he turned around, it was Tramiel, who was giving a warning about the costs. Tomczyk wanted to control the marketing of the new computer. “Back at that

time, prior to 1984, marketing as a term or concept didn’t exist,” says Leonard Tramiel. “You had sales and advertising and product development, and they were done by separate groups.” Tramiel sent Tomczyk to Japan to watch the release of VIC-1001 in preparation for the US release. * * * Commodore Japan’s general manager was Taro Tokai, known as Tony by westerners. Tokai was one of the most important general managers because of his ability to negotiate with parts suppliers in Asia, and he was skilled in setting up and operating manufacturing plants. “He was also getting the stuff out of Korea, and he was overseeing Korea and Taiwan until they built up the Hong Kong operation,” says Russell. Tokai gave Commodore access to the same chips Japanese electronics companies relied on. “If you went over to Japan you would find they had ten thousand chips that weren’t available in the United States,” says Russell. “At that point in time, the chip technology was already shifting to Japan. The NECs of the world were catching up with what they were doing in the US, especially peripheral chips and integrated chips.” When the RAM chip shortage hit the world in the early eighties, Tokai continued supplying Commodore. “We were really scrounging to buy RAM and stuff like that,” says Russell. “He was a big supplier of parts for the computers.” Tokai spoke formal English, and just as important, blended into western business culture. Like the other general managers, Tokai was always present at meetings in London and trade shows such as CES. Physically, Tokai was startlingly thin. When he traveled to North America, his fellow managers were worried about his health and encouraged him to eat more. He had a long face with a pencil-thin moustache and spoke in a soft, deep voice. “He was an extremely brilliant world class business leader who understood the implications of the first home computers,” says Tomczyk. “He ed and encouraged his small team of engineers and software programmers to work on this computer and make it happen.”

Tokai was the first Japanese businessman Russell met. “Not knowing much about the Japanese, I was really impressed,” says Russell. “Then I found out he was the wild maverick of Japan. He had a huge reputation in Japan for being this bright, non-traditional Japanese manager.” Russell was in awe of Tokai’s ability to market American electronics in Japan. “He was a sharp guy,” says Russell. “He was marketing things like the VIC where he’s going up against Japanese products in their own market, and he was selling PETs. I mean, my God, how did he do it? All those other guys had specialized keyboards with all different types of shift keys to generate all the character sets.” Russell soon learned that Tokai was an exception compared to his Japanese contemporaries. “Most Japanese managers, and even people below them, were company men,” says Russell. “Tokai wasn’t like that at all—he was his own man. He was going to be successful on his own. Commodore was a great vehicle early on, but you could tell he was independent of that.” In many ways Tokai was the managerial equivalent of Yash Terakura. “He was pretty much like me,” says Terakura. “Very independent and he didn’t fit in with Japanese society either. He was pretty much an oddball.” Much of Tokai’s managerial style was absorbed from the United States, allowing him to combine the best aspects from Japan and America. “He was stationed in the US with the Ricoh company and I think he kind of picked up the lifestyle in the US,” says Terakura. “He liked it. It’s better than the Japanese lifestyle.” Tokai’s unusual style earned him praise in the Japanese press. “They ran articles on him in Japan that said this is the type of person the Japan manager needs to become,” says Russell. “Honestly, when I meet the Japanese managers that I’m impressed with, they are still all the Tony Tokai types. There are a lot of mildmannered ones that I’ve run across in my time in Japan, but the ones who blow you away are like Tony Tokai.” The Japanese media was intensely interested in the new microcomputers at the time, and Tokai ran one of the top companies. “A lot of the media interviewed him,” says Terakura. “At that time, Apple, Commodore, NEC, Panasonic and all those were big computer makers. He was one of the top five. He was just like Jack Tramiel in the US and he was a pretty prominent person in Japan.”

Tokai was one of the rare Commodore managers receptive to Tramiel’s low-cost VIC computer. He had seen the market share of the PET in Japan plummet from 50% to only 15% in the space of three years and was eagerly awaiting a new product to reverse the trend. His most ferocious competitor in Japan was NEC. In 1976, NEC released the TK-80, the first Japanese home computer. TK stood for training kit, and it was similar to the KIM-1. Next, NEC released the Compo BS/80 in 1978, which included a keyboard and television output. In 1979, NEC released the PC-8001, a more sophisticated entry with eight colors and high-resolution graphics of 160 x 100 pixels, plus upper and lower case text. By the early eighties, the PC-8001 owned 45% of the Japanese personal computer market. “We were launching it in Japan specifically because we couldn’t afford to make the huge volume we needed in the United States,” recalls Tomczyk. “We had to acquire capital and facilities to ramp up manufacturing. So it was better to take it in Japan, test the waters, and do a launch there.” As general manager for the Japanese region, Tokai could market Commodore products as he saw fit. He decided to call Vixen the VIC-1001 to retain continuity with the PET 2001. “The reason it was called the VIC was the chip itself was called the VIC chip,” says Terakura. “The reason it was 1001 was because the PET was 2001, and it was smaller successor so it was 1001. Simple as that.” With the internal electronics reorganized for manufacturing, the final design was ready for production. In just over three months, the Japanese had produced a clean, small, easy to manufacture, reliable machine. By September, Tony Tokai had a minimal production process ready, and VIC-1001 computers began rolling off the assembly line. * * * To properly launch the VIC-1001, it was important to have software that could demonstrate the capabilities of the new computer. Terakura himself had some software skills. “When I was in Japan I developed some games—very simple ones,” he recalls. “In Japan there were a few games but it wasn’t many as I recall.” Tokai did not have to look far to find skilled programmers eager to develop

quality software for the computer. He was familiar with the “groupies” who came to Commodore Japan after their classes ended to help out the engineers. “That was my hangout, and I was not alone,” says Satoru Iwata. “There were others there who also looked at those early computers, and thought the same thing I did: how could we play games on them?”[2] The programmers initially developed games for the PET computer, using the rubber-keyboard PET-2001 models they all owned. “We became friends, formed a club, and soon rented an apartment in the Akihabara district of Tokyo, where we began deg our own games,” recalls Iwata. Their initial games for the PET had been very disappointing compared to the games they played in the arcades. In early 1980, Iwata was finishing up his two year program at the Tokyo Institute of Technology. “I don’t really how, but I managed to keep up with my class work and graduated from the institute,” he recalls. On February 21, 1980, five of the groupies formed their own company. “That group of friends is what became the company known as HAL,” says Iwata. Terakura re the period. “Iwata, who is now the head honcho of Nintendo, was working for HAL.” Although Iwata was a young man, several of the programmers were still in high school. “In of the programmers, a number of them were just really young,” says Commodore software developer Andy Finkel. “There were a couple of senior guys so they weren’t entirely kids.” The groupies were fixated with the PET 2001 computers, and the obvious similarities of the monitor housing and keyboard from the film 2001: A Space Odyssey. In the film, a super intelligent computer named the HAL 9000 controls a complex space station. The Japanese game programmers decided to name their company HAL Laboratory. “We thought that name was very cool,” says Iwata. “The name came from the computer in the movie 2001: A Space Odyssey.” According to Iwata, the entrepreneurial instincts of the young programmers were not appreciated by his parents. “I left to become only the fifth full-time employee of HAL and when I told my father this, you can imagine, it was not the happiest moment in the history of my family,” he recalls.

When the young programmers saw that Commodore’s Yash Terakura was working on a small computer meant for video games, they knew where they wanted to spend their development efforts. The programmers were particularly taken by a game called Pakkuman, first released in Japan on May 22, 1980. The Namco game featured a small yellow character that gobbled up dots and ate blue ghost monsters. “What happened was a group of Japanese programmers got really frustrated that they couldn’t have Pac-Man on the computers,” recalls Michael Tomczyk. Each programmer focused on developing an arcade version of a game for the VIC-1001, with much attention spent on Pac-Man. “We worked until midnight or later every night,” says Iwata. Although the VIC-1001 was not a good development system, the programmers could use more powerful PET computers to code and compile, then run the software on the VIC-1001. “They had a lot of things first which were really convenient for the average computer enthusiast to be able to develop their own software outside of Commodore,” recalls Finkel. “They got the hard drives first, which are kind of nice for development. They had the memory expansions first.” Tony Tokai made sure the HAL programmers had early versions of the VIC1001 so they could bring them back to their rented apartment in the Akihabara section of Tokyo and program games through all hours of the night. The developers used an interface to connect the VIC-20 cartridge slot to the PET computer so that the cartridge could load right from the expanded PET memory. When done coding the game, they burned the code to a ROM chip and placed it into a cartridge. “They turned Pac-Man into a VIC-20 game on a cartridge,” says Tomczyk. The young HAL Laboratory programmers also made conversions of the most popular arcade games of the day by Atari, Namco, and Taito—Space Invaders, Avalanche, Galaxian, Night Driver, Rally-X, and Lunar Lander. HAL Laboratory pushed the graphics and sound of the VIC-20 to the limit, producing accurate reproductions of the original games. “The Japanese folks were doing basically perfect clones of arcade games,” marvels Commodore software developer Neil Harris. A programmer named Hitoshi Suzuki developed a clone of Atari’s 1979 game

Lunar Lander. Incredibly, he created software sprites that were as smooth as those done in hardware. After completing this game, Suzuki began development of another 1979 Atari game, Asteroids. Suzuki’s brother ported a version of Atari’s 1976 arcade game, Night Driver. Another programmer created a clone of the 1978 Atari arcade game, Avalanche, [3] which shared similarities with Breakout. The games were of such high quality they could only come from arcade players obsessed with recreating the arcade hits in their own homes so they could play them at will. “In those days, to see that stuff on that little VIC, and the fact that they were actually playable, was pretty much blowing me away,” says Russell. As one of the oldest HAL employees, Iwata kept order among the programmers. “People sometimes ask me what I did when I was hired at HAL,” he explains. “The answer is that I was a programmer. And an engineer. And a designer. And I marketed our games. I also ordered food. And I helped clean up. And, it was all great fun.” When the groupies had several games in beta version, Iwata approached Tony Tokai of Commodore. According to Finkel, “The kids in Japan wrote some games and went to Commodore and said, ‘Hey we’ve got these really nice games. Want to sell them?’” Surprisingly, Tokai said yes, despite the dubious legality of the games. “Japan’s copyright laws were different then,” says Andy Finkel. “It kept evolving as the legal community started noticing computer software.” “The world was a different place then,” explains Neil Harris. “Intellectual property rights were not as well developed in this field at that timeframe. That was the situation in Japan then and those laws and precedents were put in place later.” With a small but impressive library of games ready to demonstrate the computer, Tokai was ready to launch the computer in Japan. * * * Earlier in the year, Tokai had developed a distribution channel for the low-cost retail computer. “It was sort of different, so he developed some different markets

because we were getting into the department stores rather than just normal electronic outfits,” says Terakura. “That’s something he started and he developed a new type of market for the computer sales.” Prior to launch, Michael Tomczyk tried convincing Tokai to call the computer the Commodore Spirit, but found out it was not appropriate for the Japanese culture. “I wanted to call it the Commodore Spirit, but at the last minute the Japanese said, ‘Spirit doesn’t mean Casper the Friendly Ghost. Spirit means ghoulish, fiendish, horrible zombies from hell that will come and eat your soul. So we can’t use the word spirit, it’s bad.’ So we went back to VIC-1001.” Although the VIC-1001 was not yet in production, Tokai was able to demonstrate a prototype computer with the same look and all the functionality of the final product. In late September 1980, the VIC-1001 debuted in the Seibu Department Store in downtown Tokyo. Seibu was the Japanese equivalent of Bloomingdales and sold high-end products on several large floors. At the time, the store hosted a computer show exhibiting products from approximately one hundred companies. Tony Tokai and Michael Tomczyk manned the small Commodore booth in the store. According to Terakura, the visiting American was there to learn. “Tomczyk did marketing in the US and had nothing to do with Japan. He visited Japan but never was involved with the VIC-1001.” The competition must have felt threatened by the low price of the VIC-1001. “[Jack] named a price of $300 and they were like, ‘Oh, you’re nuts’, but he scared them,” says Russell. The rival Japanese companies were stunned by the new offering. “It was a surprise to them because the computers with that low price did not come with all the amenities,” explains Russell. “It was sort of a different concept.” Curiosity overcame the rival Japanese engineers. “We had every engineer in Japan trying to sneak into our booth at that department store and unscrew the case of that first prototype computer,” says Tomczyk. “Every time we turned our back there would be somebody trying to unscrew the case, looking at the circuit layout inside that computer to see how we did it.” Tomczyk was not alarmed by their actions—in fact, the opposite. “It was

extremely flattering and empowering because it told us that we really had something there. The engineers were almost rabid to get inside and take a look at what we had. It was very cool.” Tokai and Tomczyk kept a screwdriver close at hand. “We had to keep literally physically shooing them away and put the lid back on with our screwdriver,” says Tomczyk. Tomczyk was new to Japanese culture and wondered if this was considered normal behavior in the electronics industry, and he asked Tokai about it. “Tony said he had never seen anything like that before in his life,” he recalls. “We were totally shocked and amazed and we high fived each other. We said, ‘Now we really know we’ve got something here.’” That evening, after the close of the show, Tokai called Tramiel long-distance to report the results of the debut. Through the static filled phone line, Tramiel could barely hear Tokai’s soft voice, but he managed to hear orders of over 1,000 computers on the first day. This was a success, considering there were probably only a few hundred thousand personal computers worldwide. Tramiel slept well, knowing his low-cost computer might drive back the Japanese competition. [1] The IEEE-488 cables were available from only one source, a company called Belden. According to Jim Butterfield, “In the year preceding the VIC-20, these cables became unavailable for some reason. Commodore found itself in the position of building disk drives and printers for their computers, but being unable to hook them up.” [2] From his 2005 keynote address at Game Developers Conference. [3] The coin-operated Avalanche arcade games used the MOS 6502 chip.

CHAPTER 19

Mass Destruction 1980

There have been infamous accidents involving corporate jets—infamous because usually only famous people own jets, so accidents tend to receive a lot of publicity. Commodore was about to the list. “There was a disaster with the [PET Jet] that was very dramatic,” recalls Charles Winterble. Just prior to the flight, the PET Jet had taken some engineers to Europe. “Scott Patterson and I had gone to Basel, Switzerland to present at a Commodore developers conference,” says John Feagans. On September 2, 1980, Jack Tramiel and his wife Helen boarded the jet. They had been visiting the east coast and were now returning home, along with financial controller Dick Sanford, VP of sales Dick Powers, and consumer division manager Ken Hollandsworth. The group made a short flight to Chicago, where they picked up two Commodore customers. At every airport they stopped, a Fixed Base Operator refueled the jet, loaded luggage and hot meals, and drove the engers onto the tarmac. “The car pulls up to the airplane with your luggage and that is put onboard, and then you board the airplane,” says Sanford. Tramiel hoped the flight to California would impress the customers, a hefty father and son team who owned a PET distributor. “There were two customers that were out of Chicago,” says Sanford. “They were just big men. They were purchasing product from Commodore and then they were reselling it.” The flight ended up terrifying them.

Because the father and son had never flown on the aircraft before, the pilot stood at the front and gave a brief safety demonstration. “The appropriate safety features are described by the pilots to those who are flying on the aircraft who have never flown before,” recalls Sanford. As usual, no one expected to use the emergency procedures. The PET Jet had a permanent crew of a pilot and copilot. “It was the company owned plane, so it was our own pilot,” says Sanford. “The pilot was a younger fellow, about 30. The copilot was in his 50s or so. The fellow who was older had been a pilot in Vietnam.” Takeoff and landing in smaller jets was always the most nervous part for engers and crew, since small jets were more susceptible to runway vibrations and wind. As they bumped and rattled down the runway, the two distributors gripped the armrests a little tighter. Soon, they were on their way home. The flight was normal to begin with. “We had taken off out of Chicago on our way to San Jose with these customers,” recalls Sanford. “It was beginning to get dark.” Whoever sat closest to the on-board kitchen was responsible for serving food and beverages aboard each flight. Today, Ken Hollandsworth happened to be closest to the kitchen so he began brewing a fresh pot of coffee for everyone. “There was the service section of the airplane with the coffee urn and all that business,” says Sanford. “Dinner had been packed on board, so once the plane leveled out, dinner was ed around. There was no service and no stewardesses.”

Jack Tramiel and Jim Finke dining in the PET Jet.

Unknown to Hollandsworth, by starting the coffee pot he set in motion a deadly chain of events. “There was a design flaw in the aircraft,” says Sanford. “The design flaw was that the coffee pot, sitting in its receptacle, was just a little bit out of kilter. As the plane moved, it would rock back and forth in its setting, and it wore away the insulation around the power bus.” Tramiel was sitting with the customers discussing a deal when they sensed a change in the cabin. “We were starting to eat and the lights flickered,” says Sanford. “I looked up and you could see a fire. The main wiring of the aircraft that went from forward to aft was burning.” It was a dangerous electrical fire. Due to the flickering lights, the pilots were immediately aware something was wrong. “The captain came back quickly and grabbed the fire extinguisher,” says Sanford. “While the fire extinguisher worked, it did not put the electrical fire out. Of course, the smoke in the plane is now beginning to build. He came back to me and asked me to get the fire extinguisher that was behind my seat, which I got out for him. He went and fired it again and it didn’t put the fire out.” It was inconceivable that the fire extinguisher failed to affect the electrical fire, perhaps because it was difficult to smother the flames on the ceiling. “Well, you’ve got a serious emergency at this point,” says Sanford. The pilot and engers stared hopelessly as the flames danced around on the ceiling. “The fire slowly burned this wire from forward to aft and it was melting the metal around it,” recalls Sanford. “The metal was just dropping these molten drops down. I would suspect if one of those things hit you, it would have gone through your arm.” To make matters worse, the jet soon suffered a complete power outage. “All electronics ceased at that point,” says Sanford. “It’s pretty dark out. The only light we had at this point, other than the flashlights that the captain and the copilot had, was from the moon.” The pilot returned to the cockpit and began discussing options with his copilot. “You could see the pilot and copilot going through their flight books looking for

whatever they were looking for,” says Sanford. The entire dashboard in front of the pilots was dead. “Of course, we didn’t know that at this point.” In the enclosed fuselage, the air became toxic. Not only did the synthetic materials give off horrible fumes, but the flames quickly ate up what little oxygen remained in the air. For the seven engers and two crew, it was now a desperate fight for breathable air. “The smoke was really building up,” says Sanford. “It was the first time I’d been in a fire, and it was like a knife at your breastplate cutting right in.” The heat was oven-like in the small cabin, and everyone’s faces were dripping with sweat. Combustibles burned violently, causing the plastic to sizzle and bubble. Suddenly, Tramiel found himself laboring for air, as though he was suffocating. “I looked over and could see Jack, and his eyes were starting to roll,” says Sanford. An amateur pilot himself, Sanford felt frustrated at his inability to help with the situation and was glad to be able to take some action to help Tramiel. “We picked up all the dinners that had been on the trays in front of us and threw them in the head,” he says. “I got Jack on the deck in the middle of the airplane, went to the head and got a cold compress and put the compress on him.” The urgent need to get everyone out of the poisonous environment suddenly hit Sanford. “Obviously everybody is a bit concerned at this point,” he says. “So I say to the pilot, ‘Get this goddamned airplane down!’” Sanford did not yet realize the problems the pilots were dealing with in the cockpit. “Unbeknownst to us of course, we had no electronics at all,” he says. “We had three radios, two of which apparently ceased with the electronics going out. They had one mayday channel left.” So far, the pilots were unable to raise anyone on the channel, so they maintained their altitude until they had a new flight plan. “I had flown in the past,” says Sanford. “Obviously, the biggest thing is you don’t want to give up airspace because that’s your cushion. If an aircraft is in trouble, you don’t want to go down because you may not be able to get back up. So these guys keep flying.” The young pilot was instrument rated, and had over 2700 hours of flight experience. “He was a young man, 30 years old, and he was a computer pilot,” says Sanford. “And when the computers went down, he froze.”

Sanford sensed dissention from the former military pilot. “You could see arguing going on in the cockpit. There’s no question it was arguing,” he says. “This guy had been in Vietnam. The scuttlebutt was, and nobody ever proved this or got the real facts on it, but he had been a CIA pilot or something.” The military pilot had been in difficult situations before, and felt he had the experience to save them from disaster. “The copilot in the right seat took over the airplane,” says Sanford. “He basically told the young man to sit the hell down and let me take this thing in.” Sanford gives credit to the pilots for not venting the smoke, which seemed like the obvious thing to do. “The problem with continuing to fly in that condition is the smoke continues to come,” he explains. “But you can’t eject the smoke. These guys were smart enough not to do that.” At high altitude, as soon as they vented the air, the oxygen masks would have dropped automatically. “What would have happened is it would have blown the airplane up,” says Sanford. “Oxygen and fire don’t mix. It would have blown it to smithereens out of the sky. They had enough sense about them and training to not eject the smoke.” It was an excruciating predicament for the pilots. Sanford and the engers, who were unaware of the full situation, began to feel panic and yelled at the pilots to do something. “In the meantime, the smoke is almost overcoming to you,” he recalls. “They aren’t dropping the plane down. We don’t know what’s going on up forward other than get the goddamned airplane down!” Eventually the pilots succeeded in ing a nearby aircraft with the mayday channel. “The mayday was picked up by a Frontier Airlines plane that was somewhere near,” recalls Sanford. The Frontier Airlines pilot relayed the distress call to the nearby Iowa City airport. “They notified them that we were in trouble and they vectored us into [Iowa City], Iowa.”[1]

The pilot (left) and copilot of the PET Jet.

The pilots began their decent into the Iowa City Municipal Airport. “We’ve given up airspace but they understood at this point they could fly this thing in,” says Sanford. It was now a waiting game for the engers as the flames continued to spread down the length of the aircraft. “This burning thing had ed me and I’m avoiding getting hit with this molten stuff coming down,” recalls Sanford. He leaned to one side as hot metal rained down, each drop melting the synthetic carpet and releasing horrid fumes. Sanford believed they were all dead now. “We thought this plane is going down,” he says. “We’re on fire, and we’re going to lose our lives from smoke inhalation. What do you think about when you are about to lose your life? We had plenty of time to think about that.” For Sanford, the worst part was not being able to do anything. “What really ticked me off is, here I am flying around up there on this missile—that’s really what an airplane is—and I had no control over what was going to happen. It was going down and I was dying tonight. And I can’t do anything. I’m stuck here.” The engers were severely weakened and delirious from lack of oxygen. Several blood vessels in Tramiel’s bulging left eye had burst, giving him a grotesque appearance. The lack of oxygen was a blessing in disguise, however, because without oxygen, the progress of the flames slowed down dramatically. “There was certainly not a lot of conversation going around,” says Sanford. “Everyone was in their own thoughts.” Eventually the pilot brought the aircraft down low enough and the pressure outside of the PET Jet almost equaled the pressure inside the cabin. It was now safe for the pilot to vent the air without triggering the oxygen masks. “As we come down towards the airport, now they can eject the smoke because we are low enough. The smoke is ejecting and the cabin is getting a little bit clearer,” says Sanford. “We don’t have oxygen masks on, but the smoke is now thinner, so it’s a little easier to breathe. Your chest is still pretty uncomfortable at that point, however.”

The lack of electrical systems forced the pilots to control the PET Jet manually. “They also lost the ability to automatically drop the landing gear,” says Sanford. “They had to manually put it down. So the pilot comes back and hands me a flashlight and says, ‘Dick, look out the starboard window and see if my gear is down.’ I stepped over Tramiel and confirmed that it was.” Decades later, Sanford laughs about the perilous situation. “I handed back the flashlight and said, ‘What the hell about the nose gear?’ He just obviously kept going. Nobody could confirm it was down.” The pilot had managed to get them over Iowa City, and they were approaching the airport. To the engers, it looked like there was snow on the runway at first glance. “They dropped down low enough, and now we can see out the windows,” recalls Sanford. “So [Iowa City] goes ahead and they foamed the runway, as you’ve seen in aircraft movies. We can see all these lights going around and around—the fire engines, all that business. I thought, ‘Oh my God, this is like Airport 1975.’” The engers believed their ordeal was nearly over. “So they turn and line up for the runway that was foamed,” says Sanford. Normally aircraft slow down their speed prior to landing using air brakes, which are flaps that reduce speed without affecting lift. “At this point, they realize with all the electronics and everything else that went out, their flaps didn’t work,” recalls Sanford. “That means they can’t slow the aircraft up as much as they wanted to upon landing.” The copilot made a last second decision. “They wanted to go to another runway, so they byed the foam runway,” says Sanford. “They start to land on another runway. The reason they did that was because either it was the longer runway or they had better wind in their face.” On the ground, the fire department crew watched as the PET Jet overflew the runway they had covered with fire-retardant foam. “The airplane hit the ground at 180 knots, and we should have come in at 120 knots,” says Sanford. Without electrical control, it was impossible to reverse the engines to slow down the PET Jet. “Now we didn’t have the ability to reverse thrust,” says Sanford. “That’s how you stop these damn things.”

Instead, the pilots would need to rely on the weak manual brakes. “We were coming in at that kind of speed, how do you stop? There’s nothing else to do so they shut the fuel down,” says Sanford. “They slam on the brakes, which doesn’t do much other than burn the brakes up immediately. We saw sparks coming on the whole side of the aircraft as it’s burning up the brakes. Now you can imagine as you are flying down this runway and it seems like both sides are on fire.” It was soon apparent the runway was not nearly long enough. “They go down the whole runway, and at the end of the runways they have these strobe light towers,” explains Sanford. “As the pilots got closer to the end of runway, they were able to steer the plane by stepping on one of the brakes. That turned out to be the right brake. That caused the plane to drive off the runway into a field.” While all this was going on, the flames rapidly closed in on the aft section of the jet where the engines and fuel resided. The fresh air had reinvigorated the fire. Sanford feared it would hit a fuel line, instantly turning the PET Jet into a ball of fire. “You’re thinking, oh my gosh, you’ve got two minutes before this hits the fuel tanks and blows it up,” says Sanford. The vibrations increased as they hit the field. “Now you’re not on a smooth runway and this thing goes bumping along,” says Sanford. “That ultimately stops the airplane.” Eventually, they came to a stop in the middle of an open field. They had been in the air for 40 minutes. Everyone unbuckled as quickly as his or her shaking hands could move. “The fire is still going,” recalls Sanford. “At the same time, the pilots came back and opened up the doors.” Jack Tramiel still lay in the aisle where the smoke was thinnest. “Immediately the father and son, who were very big guys, they literally knocked Mrs. Tramiel on her rear-end and dove out of both sides of the airplane,” says Sanford. “We all helped Mrs. Tramiel, who is a little lady, out the door. And then we got Jack up.” The occupants lunged out onto the grassy field, tasting clean air once again. As soon as their feet hit the ground, they started running away from the flaming jet. “We just ran as fast as we could,” says Sanford. “You’re just running on adrenaline, faster and faster, as far as you can, waiting for this thing to explode. And finally you stop and look back.” The group stood in the cold field, watching the flashing emergency lights. “We

were making sure that everybody is okay. Nobody was hurt. By now, the fire engines are catching up, and they have those big lantern lights,” says Sanford. “The guys in the asbestos outfits and the big foam guns came flying out and they just enveloped the airplane with foam.” One paramedic attempted to fix an oxygen mask over Tramiel’s mouth, but Tramiel waved him away. The fresh air felt too good in his lungs. After a few minutes, he was able to catch his breath. “And then the adrenaline starts to run down, and now you’re standing out there in the freezing field,” says Sanford. “They trudge us back into a hangar. I guess they took statements and that was it.” After the harrowing ordeal, all seven engers retired to the airport lounge for drinks. “We all inhaled alcohol real fast because we were just really high from what went on,” says Sanford. Travelers who saw the motley group of sootcovered, disheveled, red-eyed engers must have thought they were street people who somehow made it into the airport. Tramiel did not care. It was the second time in his life he had narrowly avoided death. “We flew out the next morning on United Airlines.” News of the incident shocked the Commodore engineers who had flown in the PET Jet many times. “We were in the plane a couple of weeks before then, and all of a sudden this thing nearly goes down,” says Winterble. Surprisingly, Commodore was publicly quiet about the drama. Tramiel and Gould both owned many Commodore shares, and they feared the accident might alarm investors and hurt the share price. Winterble believes the copilot’s quick thinking and refusal to land on the foamed runway saved the lives of his engers. “The copilot was a hero,” says Winterble. “If the copilot hadn’t taken control of the situation and gone beyond the call of duty, they would never have made it.” * * * In the aftermath of the accident, investigators searched for the cause of the fire. As an electrical engineer, Bill Seiler took an interest in the investigation. “The main power bus had been shorted out by the coffee maker,” he explains. It was the frayed electrical cords around the coffee station.

There were approximately 250 Westwind jets in service, with the potential to cause a similar disaster. “The FAA got this and sent this emergency message around the world to anybody who had these Westwinds,” says Sanford. “They came up with half a dozen that were not in trouble yet, but the fraying had begun. This incident hopefully saved some other lives.” The incident served as an example of what to do in a fire emergency while in flight. “That whole story is taught in flight schools about the wisdom of these guys,” says Sanford. Although the copilot continued his career,[2] the young instrument rated pilot retired after the emergency landing. “My understanding is that the pilot never flew again,” says Sanford. “Obviously, in anything, the captain is always in command, but the copilot is the guy who saved our lives.” Engineers determined the damage to the PET Jet was not extensive and the insurance company sent it to the Midwest where technicians would rebuild it. “It went to Kansas City where they fix these things,” says Sanford. People interpret the events in their lives in many ways. According to Michael Tomczyk, Tramiel told him, “It was a message. God was telling me, ‘Don’t fly so high.’”[3] * * * Chuck Peddle’s recent unhappiness with Tramiel began when he hired Tom Hong. He believes Tramiel gave Hong more credit for the VIC-1001 project than he deserved. “Jack had taken off for some reason during that period,” explains Peddle. “He gave this guy Hong the assignment of coming up with the design and getting it ready for production. When he came back, it was done, and I think he really felt this guy had done it.” At the time, Peddle believed the VIC-1001 would be a failure. “We had suggested to Hong that he take credit for it because we didn’t want to get into an argument,” says Peddle. “We frankly felt we were secure enough that we didn’t need any credit.” When the VIC-1001 turned into a success in Japan, Tramiel felt he had capable systems and production engineers. According to Bob Yannes, “Once he saw that we could deliver and given that Chuck Peddle and his team were getting

rebellious, Jack gave us control over system engineering.” Tramiel was slowly taking away the computer systems division from the man who started it at Commodore. In September 1980, Peddle confronted Tramiel by surprising him after his recent flight into California. “I picked him up from the airport one day because I wanted to talk to him about the design center,” says Peddle. “He wasn’t happy because he told somebody else to pick him up. So he felt I had him trapped, which I did.” With a captive audience, Peddle laid out his feelings about Commodore’s engineering. “Jack and I had this conversation about the things we needed to do,” he recalls. “I told him some stuff I wanted to get done. I almost told him I had finished the VIC-20 for him.” Peddle also held a long-standing grievance that he was no longer running the computer systems division, which he was promised when he first came to California. Peddle told him, “Jack you’re not taking care of me. You’re not working with me the way we agreed on.” As Peddle drove on, their discussion turned into an argument. Peddle felt emboldened, and said, “Jack, every time you haven’t listened to me something has gone wrong. Instead of arguing with me, you should be treating me better.” The conversation did not help Peddle’s cause. “I could tell he was pissed when I dropped him off,” says Peddle. Tramiel was now certain his engineer had defied his orders to stop work on the business computer. He felt his engineer was out of control. It is easy to see why Tramiel was unhappy with Peddle. Without Tramiel, Peddle would not hold an influential position in the computer industry deg future products and running several R&D labs. Peddle simply wanted too much freedom. “You had Chuck almost running a separate computer division within Commodore,” recalls Kit Spencer. It was a disdainful attitude towards the man who gave Peddle so many opportunities. Tramiel questioned the success of Moorpark. “It wasn’t as effective as Chuck’s original PET development or 6500 microprocessor development with a small team,” says Spencer.

What came next horrified Peddle. “The next day he announced he was breaking up Moorpark,” says Peddle. The memo told his Moorpark engineers to report to Tom Hong at the Santa Clara offices. “When they disbanded Moorpark, all those guys went to work for Hong.” “I was reassigned to Pennsylvania,” he says. “If you want to talk about Jack Attacks, he came in the morning and gave orders to [lawyer] Richard [Blumenthal] and they disbanded Moorpark in an hour. There wasn’t even a chance to wrap it up and document it or anything. It was a very abrupt decision.” Tramiel also closed the other research centers in Phoenix and San Luis Obispo. The latter was not a big loss, according to Peddle. “I never really got it going,” he says. “Commodore trashed it so soon.” The Phoenix team, which specialized in data storage hardware, had some impressive products in the works, including a hard drive. As with the Seagate floppy drive mechanisms, Commodore used an outside supplier for the hard drive mechanism. “We chose Memorex or one of those guys and I don’t think they ever finished the drive we selected,” says Peddle. The Phoenix team was also on the verge of releasing several different models of eight-inch floppy drives with different storage capacities. The first drive, the 8060, was a single drive that could store 750 kilobytes. The 8061 dual-drive had 1.6-megabyte storage capacity and the 8062 dual-drive had 3.2-megabyte capacity. Peddle believes Tramiel closed the centers in anger. “I don’t think that happened because of Jerone Guinn. I don’t think it happened because of Tom Hong. I think it happened because I pissed Jack off that night,” he says, referring to the drive from the airport. “Was he already unhappy? Yes. Was he upset with me because I was pushing him into a separate business? Yes. Were those all triggers to what happened? Yes. He decided to show me, ‘It’s my company and I’m going to make it what I want it to be.’” The closure of the R&D centers hurt Peddle. “I felt really bad about Moorpark,” he says. “It was something that we did very well.” Around the same time, Commodore began a migration to the east coast. Tramiel was tired of losing employees to other companies in Silicon Valley. He was also looking for ways to reduce his taxes, likely from his observations of Irving

Gould. “I think he wanted to get out from under California taxes,” explains Peddle. “He wants to spend more of his time in New York. He was also thinking about some way to get dual citizenship as a Canadian.” As a Canadian citizen, Tramiel would be exempt from taxation when selling Commodore shares. Later that week, Tramiel asked Peddle to begin preparations to move back east, presumably for a punishment tour. “He wasn’t trying to fire me,” says Peddle. “He was just sending me back to work for [Dick] Sanford in Valley Forge. He fully expected me to take my punishment and hang around for a little bit until I got back into favor.” However, Peddle did not intend to leave the west coast. He had grown accustomed to the warm climate of California and more importantly, he wanted to stay in a strong technology environment. Peddle talked with Tramiel and Sanford about his impending move. “The only option for me was I had to move to Pennsylvania and go into purgatory like those other guys,” he recalls. “I basically told them, ‘Hey, that’s not our deal. I don’t want to go.’ Sanford basically said, ‘Well, then I’ve got to terminate you.’ I said, ‘Okay, terminate me.’”[4] * * * When Tramiel closed down the Moorpark labs, it left the TOI in a lurch. Many of the key engineers working on the TOI were now gone and there were no executives pushing for the computer. With so much emphasis on the VIC-20, the TOI project became less important to Commodore. “Once [the VIC-20] was shown to Jack prior to the June 1980 CES, I think it was all over for the TOI,” says Bob Yannes. “It was having too many problems, it was too expensive, and it wasn’t really what Jack wanted.” Although Charpentier struggled with the 6564 chip through most of 1980, the problems remained. “It was absolutely technical issues,” says Robert Russell. Bill Seiler had previously attempted to overcome problems with the chip through hardware hacks, but nothing worked. “If Seiler could have solved it with hardware fixes, I’m sure he would have but they couldn’t really make it work because of the core chip,” says Russell. Seiler assigns equal blame for the failure of the chip between MOS Technology

and himself. “It was probably 50-50,” he says. “We had just started from a wrong beginning somewhere by trying to kludge the VIC-20 chip to make it do something. It was probably a bad approach instead of starting all over again and thinking the whole problem through.” Eventually, MOS Technology gave up on the 6564 chip. “There was really little we could do to help Peddle’s group,” says Yannes. “We had already told them the proper solution was to design a new chip, but there simply wasn’t time or budget to do that.” The TOI is a story of failure, where the engineers pushed too close to the limits of RAM speed. “They were up against the edge of technology,” says Russell. However, Al Charpentier was not defeated. He would soon begin a redesign of his graphics chip, and it would be even more ambitious than the 6564. * * * Bud Fry had put together the project to develop a scientific calculator for Commodore back in early 1979. Unfortunately, by autumn of 1980 it was clear the project was going nowhere. Neither of the acquisitions produced revenues for Tramiel. “He bought this LCD company and it’s not working out. He bought this CMOS company and it’s not working out. So he doesn’t have a calculator business anymore,” says Peddle. Bud Fry’s due diligence had not revealed that Frontier was incapable of manufacturing the CMOS chipset. “We went to the manufacturing place that was supposed to be able to build the CMOS, and they didn’t stand a chance of building it,” says Peddle. “They had been overrated when they came into the company.” Commodore was unable to nurture their investment. “To get CMOS built took a significant investment,” explains Peddle. “At that time, the Japanese told everybody what the rules were. If you didn’t go out and invest $100 million in good process equipment, you couldn’t play.” Irving Gould was indirectly responsible for the CMOS failure. “Every time he would try to go out and raise money to make that kind of investment, Irving would tell him no,” says Peddle. “Jack tried to back door it by buying this company.”

The failure of Frontier meant Commodore made little use of Micro Display Systems’ LCD manufacturing capability. Worst of all, since the time Tramiel had purchased MDS from Seiko for 100,000 Commodore shares, Commodore stock had skyrocketed. The deal became worse and worse as Commodore stock increased in value. Most businessmen would write it off as not doing their proper due diligence, but Tramiel did not let it go. Instead, he launched a lawsuit against Seiko, claiming they misrepresented MDS, in the hopes of regaining some of the shares. In autumn of 1980, Bill Mensch completed the designs on the 4500, the 4501, and the 4503 scientific calculator chips, but once again, he would collect no royalties from Commodore. “All the things we got Bill started on died, not because of anything he did,” says Peddle. Tramiel’s decision to abandon the 4500 chips was due to technical limitations of the chips. “The 10 microamp standby refresh power doomed the design to failure in the marketplace when compared to the existing 1 microamp designs for scientific calculators,” says Mensch. “Commodore lacked the market understanding to guide the decision making and furthermore to commission a static RAM redesign.” Just prior to completing the 4500 chips, Mensch began hearing rumors that Tramiel believed Commodore owned WDC, even though that was clearly not the deal they made in Las Vegas. “He told everybody that he knew that he owned WDC,” recalls Mensch. Tramiel might have been testing Mensch to determine how tightly he would hold onto the WDC in case Tramiel decided to sell it to recoup some of his losses. Since Mensch was largely dependent on Commodore for expenses, he chose not to contradict or confront Tramiel. Although Mensch had designed the chips in good faith, Tramiel put him on the firing line, along with the rest of the calculator program. “When that fell apart, Jack basically decided he didn’t need a design center,” says Peddle. “He threw Bill to the wolves.” It started when Mensch purchased an office for the Western Design Center at 2166 East Brown Road in Mesa, Arizona. The action seemed to provoke Tramiel. “Because I bought my office, Jack no longer felt he controlled me,” says Mensch. “The relationship ended abruptly after I had purchased the office.”

Another possibility is that Tramiel wondered how Mensch raised enough money to buy the building when all he was paying Mensch was office expenses. According to Spencer, “It was important to have that trust with Jack and personally he honored several handshake deals with me although he could be ruthless to people in business—especially if he felt they didn’t trust him.” Perhaps Tramiel no longer felt Mensch trusted him. Tramiel decided to salvage anything of value from his WDC deal, so he instructed a Frontier employee named “Mac” to scavenge the WDC offices. “He sent Mac over and he had a U-Haul van and he picked up anything he thought was Commodore’s, put it in a truck and left. And that was it, that was the end of it,” recalls Mensch.

The Western Design Center in Phoenix, Arizona.

Mensch was among the best silicon designers in the industry and could have helped Commodore’s microcomputer business down the line, yet Tramiel severed the relationship coldly. For his part, Mensch claims he honored their handshake agreement. “I had an agreement with Jack that I would work exclusively for Commodore. I was loyal to that to the letter,” he says. “I had no other client because when I said I would exclusively work for him, I translated that to I won’t even set up, or discuss the possibility of doing work outside Commodore.” With his sole customer no longer paying the office bills, WDC ran into financial problems. “At the end when he pulled up stake, I probably had a couple of thousand dollars in my bank ,” he says. With no money for payroll, Mensch’s Western Design Center looked like it might become a casualty. “I had to lay off all of my employees, so that’s what I did,” he says. The sudden abandonment surprised Mensch, but he swore he would persevere. Mensch reflects on Tramiel, “He put more companies out of business than probably anyone in the electronics industry, with great pride. He put me into business. I figured if I could survive with Jack Tramiel for two years, I could survive anywhere.” [1] Dick Sanford, John Feagans, Bill Seiler, and Michael Tomczyk all state the airport was at or near Des Moines, but the National Transportation Safety Board report of the incident states it was Iowa City. Furthermore, Iowa City is closer to Chicago along the flight path. [2] Sanford was sad to hear of the copilot’s fate. “He’s the guy who saved our lives. He ed away several years later with cancer.” [3] Michael Tomczyk, The Home Computer Wars (Compute! Publications, 1984), p. 186. [4] In court, Sanford later denied having this conversation.

CHAPTER 20

Return of the Curse 1980

Chuck Peddle believes his conflict with Jack Tramiel stemmed from the failure of Tramiel’s calculator business and the rise of Peddle’s computer division. “We’ve got a pretty good business going and Jack discovers that he’s out of the calculator business,” he says. Tramiel’s habit of using unproven managers in positions of power frustrated Peddle. “Jack always had the hero of the day,” he says. “He would bring people in who would tell him that he was the greatest guy in computers and that he could make them more money. Jack would say okay, and he would hire him. Then, instead of me being in charge of them, I’d be working for them for a week. That wasn’t my deal.” Commodore soon gained a reputation for the constant turnover of management. “They came and went pretty fast,” says Peddle. “Sometimes I would hire someone and all of a sudden Jack would see what they were doing, fall in love with them, promote them, and give me another assignment. It was like he couldn’t leave it alone.” Tramiel constantly stirred the pot before Peddle’s projects could congeal. “Most of the time I was doing the right things for him,” says Peddle. “I messed up with the floppy, but the rest of the time I was doing the right things. So it had me in an unhappy frame of mind.” Peddle disagreed with Tramiel’s attempt to enter MOS Technology into the memory market. During this time, Peddle believed it was at the expense of his computer systems program. “My view was that he had not given me the

resources because he had all these other guys fooling around with the memory business.” Unfortunately, MOS Technology was unable to keep up with the Japanese. “The Japanese changed the way memory worked and got more density over the Americans,” says Peddle. “It was just obvious that they were going to kick our ass for a while.” By the time MOS Technology began developing SRAM chips, DRAM began to dominate the market. “We started Commodore with static-RAM memory and discovered it didn’t work,” explains Peddle. “We had to go out of the memory business because we fucked it up big time.” In his race to pull out of the calculator market, Tramiel continued to break agreements. “Jack didn’t keep his word whenever he felt it was in his interest not to keep his word,” says Peddle. “I saw him do it to about five people.” Peddle was most hurt when he saw his own friends at the receiving end of Tramiel’s tactics. “He took my buddy who helped me do the original PET [Petr Sehnel], then gave him a promotion and then broke him out of the company. He took my buddy [Larry Hittle] in Phoenix who helped us with the KIM-1 and effectively broke him after cutting a deal. Some people would say he broke the deal with Bill [Mensch].” Tramiel also broke his June 1980 promise to Chris Fish to allow him to run the European operations, possibly because Fish ed the business computer. “Jack hired this president, Jim Finke, as European President,” says Peddle. The broken promise demoralized Fish. “Effectively, Chris took the attitude, ‘Screw this, Jack. Just as you are promising me that I’ll be head of Europe, all of a sudden you bring this guy in.’” Fish now held enough Commodore stock that he could easily resign. “Chris was just capable of cashing out his shares, which at that time were worth about three to four million dollars,” says Peddle. A week after Tramiel closed down Moorpark, Fish decided he had taken enough and left Commodore. “I think Jack thought he would go away and come back after a while.” * * * When the conversation with Dick Sanford to terminate Peddle took place, Peddle

was already planning to leave Commodore. “I had already talked to Chris [Fish] and [John] Paivinen about starting a new company,” he recalls. “The two of them approached me at the same time and said, ‘Why not go do this yourself?’” Dick Sanford does not feel Tramiel pushed Peddle out of the company. “That might be Chuck’s perception but I don’t believe Jack wanted him out of engineering at all,” says Sanford. “Chuck saw, as all entrepreneurs do, an opportunity to take his skills and technology to build his own company. And he and another individual in the company put a partnership together.” “Chris and I actually started the idea. We decided to do it over drinks at the Buena Vista after running a marathon,” says Peddle. “Chris had been part of this discussion on spinning out the company. John [Paivinen] came up with the idea that the world was ready for a new computer.” The trio planned to build a personal computer using some of the work developed at Moorpark. “I was working on all these pieces, most of which either indirectly or directly got incorporated into our PC machine,” says Peddle. Most of the engineers at Peddle’s former research labs were disappointed with how Tramiel had so casually discarded their hard work. Peddle approached Bill Seiler, Glen Stark, Scott Patterson, John Feagans, and Robert Russell to his new company. “Jack closed Moorpark down and we came back to Scott Boulevard,” recalls Seiler. “I was sitting there and I Chuck said he was going off to do a new company and he couldn’t tell us anything about it. Chuck was always very careful because of the lawsuits at MOS Technology. He got sued by Motorola. After that, Chuck was very careful about how he extricated himself from a company.” In a move that was sure to antagonize Tramiel, Peddle began recruiting Commodore employees for his new company. First up was Tom Mitchell, a Commodore production executive. “We were very careful how we recruited people out of Commodore,” says Peddle. “We expected to get sued so we developed this paper that people had to sign.” The engineers were determined to prevent another lawsuit. According to Seiler, “He said, ‘Don’t have anything but what is in your head when you leave Commodore. Don’t bring any paper. Don’t bring any prototypes. Don’t bring anything.’”

Contrary to Peddle’s assertion, Robert Russell re the Commodore engineers getting together for discussions. “I went out and did a whole bunch of offsite super-dooper secret meetings,” he recalls. “We were at a Japanese place eating shabu-shabu that night. The whole bunch was there.” Both Seiler and Stark accepted almost immediately, but Feagans and Russell had reservations. “Feagans was approached,” says Peddle. “John turned us down.” Russell felt there were some bad feelings between Feagans and his defectors. “They acted as if they didn’t want John and they kind of thought he wasn’t on their side,” says Russell. At first, Russell felt it would be best to leave Commodore because the west coast was losing its importance. “It was pretty clear that it might have been a good idea to leave and go up with those guys,” he says. Russell was working on an advanced BASIC for Commodore at the time. “[It] was designed to do high accurate math for business stuff, and it had all the Microsoft bugs expunged from it,” he recalls. “They said, ‘Can you bring that with you? Nobody’s going to notice.’ It just struck me wrong. As much as I liked Chuck and his team, I really didn’t want to go.” Seiler, who held a deep interest in computer graphics, had previously begun to figure out some of the problems with creating high-resolution graphics, which had stymied the team earlier with the TOI. “I figured out a way to put the [character set] in the DRAM so that the character font was programmable,” says Seiler. “You could change the character font.” This also made it easy for Seiler to create a bitmapped screen. “The way I did graphics is, I put different characters on the screen in each location and just played around in the font memory, and you could draw bitmap graphics.” Peddle had explicitly forbade his engineers from bringing any hardware with them to the new company. “I was working on a graphic PET and I had it working at home,” says Seiler. “He made me throw it out or give it away.” Seiler ended up handing the prototype to Russell. Peddle and his defectors formed Sirius Systems Technology in Scotts Valley, California. The man who saved Commodore by creating their computer business was gone. “I’m sure it was a total surprise to [Jack] that I wandered out the

door,” says Peddle. It took Peddle many years to figure out why Tramiel mistreated him at Commodore. “Why, after I’ve done all these things for him, why does he force me out?” asks Peddle. “It’s because I effectively stole his company without ever intending to. … The guy whose name is on the door has a real hard time letting go. Jack felt his name was on the door at Commodore.” Tramiel soon realized the man who replaced Peddle was inadequate for the task. “He thought Hong had done the VIC computer design so he thought he had me replaced,” says Peddle. “Jack turns around and looks at this guy, and gives him some other assignment, which he screws up.” Hong found himself on the receiving end of a Jack Attack. During the meeting, Tramiel determined if Hong could learn from his mistakes. “A lot of people didn’t get the message,” says Russell. “They thought he was attacking them. If you got confrontational, I’m sure it could end up with you getting fired. I saw a lot of people leave that way.” Hong did not get the message and did a poor job of representing himself in his meeting with Tramiel. “He wasn’t as capable as Jack thought he was, so Jack fired him,” says Peddle.[1] Hong was one of many managers Russell saw who did not survive their first Jack Attack. “I used to see my managers come out of there, and you were like, ‘Okay, is he going to come back to make the turn and come into the office? Nope! He’s going out the door. Okay, I guess he’s no longer my boss.’” As managers routinely fell to Jack Attacks, it became difficult to find anyone willing to work for Commodore. “They just didn’t have a director of engineering, because we couldn’t hire them after a couple came and went,” says Russell. “It was so bad that we ended up putting our high-speed printers in the head engineering manager’s office because it was usually empty.” * * * From his experience with the PET, Tramiel knew he could sell several peripherals for every VIC-20 computer he shipped. Financially, the peripherals were as important as the computers.

Commodore anticipated that the most important VIC-20 peripheral would be the cassette unit. Since the VIC-20 used the same cassette interface and code as the PET, Commodore continued selling the same cassette unit for a cost of $75. Robert Russell designed an unusual expansion device to allow PET peripherals to connect to the VIC-20. “I did IEEE cards for the VIC-20 and the IEEE kernel because we were trying to use the drives,” he says. Commodore did not release the product. “I don’t think any of those things made it as products for Commodore.” Russell also helped develop VIC-20 memory expansion units. “I was working on VIC peripherals under the Commodore brand, like the memory board,” says Russell. By inserting more memory, s could load larger programs into memory. The VIC-20 was theoretically capable of up to 32 KB of total memory. At the VIC-20 launch, three memory cartridges were available: three kilobytes for $39.95, eight kilobytes for $59.95 and a special three-kilobyte cartridge for $69.95, which allowed high-resolution graphics and additional BASIC commands. Memory expansion became an integral part of owning a VIC-20 and Commodore sold almost as many expansion cartridges as they did computers. Eventually, Commodore sold a 16-kilobyte expansion cartridge for $109.95. * * * The few Commodore employees who ed the VIC project gained power within the company beyond their official title. On the west coast, rookie engineer Robert Russell gained influence by pushing the Vixen development forward. Michael Tomczyk also ed the computer, and soon took charge of the unwanted project. In a short period, he suddenly eclipsed senior marketing executives due to Tramiel’s unequivocal backing. “He went to Jack and convinced him to let him do something new for Commodore, which was a launch team dedicated to the VIC-20,” says Andy Finkel. “Since it was pretty much a new title at Commodore, Mike got to define a lot of his own role.” Tomczyk grew up in small-town Northern US, near Lake Michigan. Since his youth, he had yearned for something greater. “When I was a young kid growing up in Oshkosh, Wisconsin, I used to pray, ‘Please allow me to do something

significant with my life because I’m growing up in Oshkosh, Wisconsin.’ I wanted to do something that helped the world in some way, but I was in fucking Oshkosh.” After a tour of duty in Vietnam and a business degree, Tomczyk travelled to Silicon Valley in 1979. “I used to hang out at Apple almost daily. I would hang out with Steve Wozniak and Andy Hertzfeld and some of the people developing the Apple computers because I was planning to go into the industry and I wanted to learn.” Although he was not an Apple employee, he walked the halls freely. “I was kind of like a groupie at Apple. They let me come and go as I wanted. I could walk around even without an ID tag, which was forbidden at Apple. People used to ask me, ‘Where’s your ID tag? Go get your visitor badge!’ And I used to just wave them off.” After six months as Tramiel’s personal assistant, Tomczyk felt ready to make his contribution with the VIC computer launch. “Jack told me, ‘Just make it happen.’” Although working closely with Tramiel had advantages, it was also potentially dangerous. “Michael was very bright but he also had an ego problem at times,” says Spencer. “He’s like most people. We’ve all got pluses and minuses. Working with Mike, we were able to use many of his plus points. The skill is using the plusses and minimizing the minuses.” Tomczyk was inexperienced and it was likely he would make mistakes, and Tramiel did not tolerate mistakes lightly. Tramiel and Tomczyk, both army veterans, had good chemistry. “Michael Tomczyk and Jack had this great relationship,” says Russell. “Jack kind of wanted to rattle the whole marketing organization, because he wasn’t really happy with marketing.” Tramiel used Tomczyk as a tool against his own marketing executives, pitting one against the other. “Tomczyk was kind of this rogue marketer,” says Russell. “Jack would let him put whatever he wanted to on his business card.” Commodore was a loosely structured organization, so Tomczyk used a printing service to create ever-improving business cards. “He kind of let Tomczyk choose whatever title Tomczyk wanted, just to threaten the other marketing people. So some guy would say, ‘I’m VP of marketing,’ and Tomczyk would have a

business card printed that said he was Executive VP of marketing,” laughs Russell. “[Tomczyk] had a lot of ego and Jack used that ego. Jack kind of thought it was funny.”

Michael Tomczyk working for Commodore.

When injured executives complained to Tramiel, he refused to intervene. “Jack said, ‘You’ve got to fight it out for yourselves,’” says Russell. “I was always ending up on the other side of Tomczyk in these fights. The other marketing guy would corner me [and say], ‘Oh, we’ve got to watch out for this Tomczyk character! He’s a big threat to the company.’ I didn’t give a shit because I was an engineer. Because I went to the shows they knew I was in good with Jack. They would always try to get me involved in their schemes.” Tomczyk stood out from other managers in the company. “He had a marketing background which was really rare in Commodore at the time,” says Neil Harris. “Mike had a huge amount of energy and a very strong vision of what needed to be done in order to make the computer a success.” Although Tomczyk held influence, he really wanted the product manager title for the VIC computer. The product manager is responsible for tying together marketing, advertising, engineering, product , production, packaging, and distribution. Unfortunately, product manager was the one title Tramiel would not allow Tomczyk to print on a business card because he did not believe in product managers. When Tomczyk became the central communications node for JapaneseAmerican discussions on the VIC project, he took the lead. “I interacted a heck of a lot with Tomczyk,” says Russell. “He was kind of always there when the VIC was real. When it went to the west coast, Tomczyk showed up soon afterwards.” One of the first things Tomczyk did for the North American release was name the computer the VIC-20. Since then, there has been speculation on why the number 20 is used. Some thought it was 2001 with the 01 dropped. Others thought it might be the sum of the ROM memory: 8 kilobytes of BASIC 2.0 plus 8 kilobytes kernel plus 4 kilobytes character set ROM. However, Tomczyk maintains the number has no meaning and he chose it because it sounded friendly. In his new role, Tomczyk attempted to steer the marketing of the VIC-20, not the

design. “I don’t think he really had anything to do with the design of the VIC20,” says Bob Yannes. “He had some overly optimistic ideas of what the VIC-20 could do. It was a very primitive computer.” Tramiel wanted to manufacture large quantities of VIC computers in the United States and in Europe. However, Irving Gould would not raise money using Commodore stock. As a result, Tramiel was forced to raise development money by selling off Commodore assets and pulling out of the calculator industry altogether. “All the calculator people [in MOS Technology] went away,” says Peddle. “I think the only people left were the game guys.” Tramiel wondered if he would be able to redirect his retail calculator marketing people into computers, or if it would be better to start over. Tramiel consulted with Tomczyk, who told him outright he thought the marketing department was weak. A thirty-megaton Jack Attack was imminent. When Tomczyk came through marketing the next day, he saw only empty desks. The whole department was vacant. “I my shock,” he recalls. “I came back one day and I saw that Jack had fired the whole marketing department. I couldn’t believe it.” Later Tomczyk ran into Tramiel and asked, “What did you do?” Tramiel answered, “Well you said you weren’t happy with them so I fired them.” In truth, Tramiel no longer needed them. Tomczyk felt alarmed that he might have been the cause, and he asked, “So now what?” It was time to start building a marketing department for low-cost retail computers. “You go hire the marketing people you think we need and I’ll hire the marketing vice president to run the group and we’ll divide up the tasks that way. And we’ll get the new marketing group put in place.” Tomczyk was now a trusted inner family member, and Tramiel gave him important responsibilities within Commodore. “I went off in a daze and started networking around to see who I could find who knew anything about computing.” Tramiel tapped an executive named Dick Powers to run the new computer-marketing department.

Tramiel was now fully committed to retooling Commodore into a retail computer company. His inspiration came directly from Atari, who had been able to achieve mass distribution of their low-cost computers in retail stores like Sears. Given the success of the Vic-1001 in Japan, Tramiel felt Commodore now had a chance to capture a share of the North American market. Success would depend on how widely Commodore could distribute their computer and how well they could . Dick Sanford was primarily responsible for launching the VIC-20 in the United States. “Dick Sanford was a key guy in Commodore,” says Spencer. “He had been financial controller close to Jack and he was general manager in the USA when VIC was launched.” Although the VIC-20 was going to sell in retail stores eventually, at first it would sell through the regular PET computer dealers for a price, until quantities were available. In October 1980, Sanford renamed and reorganized the existing PET dealers into the Commodore Dealer Network in anticipation of the VIC-20. The same month, Sanford organized a meeting of the regional sales managers in Valley Forge, Pennsylvania. After Sanford addressed the group, Tramiel announced Tomczyk as the new director of marketing for the United States. Tramiel had fulfilled nearly all of Tomczyk’s ambitions. As director of marketing, he could make a difference. He had risen very high very fast. While Sanford handled the distribution issues of the VIC-20, Tomczyk would handle product marketing materials and packaging. “Michael was very involved with the VIC in the States in the early days and Dick was involved as the general manager on the project,” says Spencer. The two were almost polar opposites in personality. “Dick and Michael were very different characters,” says Spencer. “Dick was very capable and his background was financial, not marketing. He’s definitely a much better ant than I am and maybe as general manager of a big company too. But he didn’t have a marketing background.” On the other hand, Tomczyk had a marketing education but very little practical experience. “Michael and he had more difficulty working together as Michael

didn’t have experience or quite the same personality,” says Spencer. In November 1980, while Tramiel was busy trying to get manufacturing plants set up in and United States, Sanford sent one of his assistants to help Tomczyk develop a business plan for the VIC-20. Tramiel had no official schedule for the release of the VIC-20 in North America, so Tomczyk set an introduction date for the second quarter of 1981. Tomczyk needed some business cards printed for the CES, so he casually asked Sanford if he could use the title of VIC-20 product manager. Sanford did not object, so Tomczyk printed them up. Tomczyk’s plan was to allow the idea of him as product manager to sink into everyone’s consciousness until no one could deny it. He hoped Tramiel would either not notice or not care enough to revoke the title. * * * When Commodore released the PET in 1977, the company exerted almost no effort to encourage software development. Only Chuck Peddle courted software companies, such as Personal Software. Despite this, a software industry spontaneously developed around the PET. Apple continued gaining s due to VisiCalc, but in 1980, Personal Software finally released a version of VisiCalc for the PET. The company was concerned with piracy, so they included a unique copy protection in the form of a special dongle that plugged into a spare socket on the PET motherboard. According to Peddle, the PET version of VisiCalc was superior to the original Apple version. “We had an 80 column screen by that time and Apple was still limited to 40 characters,” remarks Peddle. “You could do those tables in a heartbeat.” Commodore UK recognized VisiCalc as an important application for the business market and later sold PET computers bundled with the spreadsheet. The availability of genuinely useful applications for the office allowed Commodore to dominate the European business market. If Personal Software had created VisiCalc for the PET first, Commodore would likely have dominated both the North American and European markets. “It would have put it everywhere,” says Peddle.

Though everyone knew VisiCalc was a winner, it took a while for VisiCalc to gain popularity. Software pioneer Chris Crawford summarizes the growth of VisiCalc and Apple. “It wasn’t until VisiCalc set-in in 1980 and 1981 that the Apple really got going.” VisiCalc was a clear turning point for the Apple II. Originally, Wozniak designed the computer as a game machine, but it was too overpriced to compete with the Atari 2600 VCS. VisiCalc transformed the Apple II into a business machine and Apple did almost nothing to make it happen. “It made Apple the definitive computer in the US market,” says Peddle. Even with the boost in sales provided by VisiCalc, Apple remained in third place. Tandy sold close to 300,000 TRS-80 Model I computers by the end of 1980, due to their massive distribution that now amounted to almost 8000 stores worldwide. This resulted in $150 Million in computer sales, compared to Apple with $117 million, an incredible increase from $47 million the previous year. Considering Radio Shack equipment cost even less at retail than Apple II computers, this puts Radio Shack total sales far higher than Apple. Commodore had a remarkable year with $125 million in revenues and profits of $16.2 million. However, it was clear Apple had faster growth in the previous year, a direct consequence of VisiCalc. In December 1980, Apple was set to take their company public. However, an earlier deal with former Commodore executive Andre Sousan came back to haunt them. “Apple is getting ready to go public,” recalls Peddle. “They took a look at the company and discovered that this one guy has all the rights to all the machines ever for the rest of his life in Europe. They basically said, ‘We don’t want that on our prospectus.’ So they went to Andre and they said, ‘What does it take to get you to go away?’” Sousan wanted one thing: Apple stock. “I think he wound up with as much stock as Jobs and Markkula,” recalls Peddle. “He had them by the shorties and he’d earned every penny of it. He absolutely put money into them when they needed money, he backed them, and he made Europe happen.” The Apple IPO made Sousan a wealthy man. “When Apple went public, they were the first, just incredibly overhyped initial public offering,” says Peddle. “They had just an incredible amount of money that they raked in as a result of

that IPO, which carried them through until they actually started making boatloads of money, after the Mac came out.” In the end, Sousan’s dismissal at the hands of Tramiel helped him enormously. “Jack screwed him and this guy just made fucking millions—hundreds of millions,” says Peddle. The dismissal also hurt Tramiel, who was never able to penetrate the French personal computer market with the P.E.T. computer.[2] The lack of proper funding by Gould had other consequences on Commodore, including the potential to fall behind the competition technologically. “We got out in front of Apple and companies like that but didn’t maintain it,” explains Robert Russell. “You say, well why didn’t you maintain it? Well, they had a lot more engineers than us and a lot more for product development at times when there were only ten or twelve engineers [at Commodore].” While Apple enjoyed proper funding for long-term success, Gould forced Tramiel into short-term tactics. With the successful IPO, it was inevitable Apple would soon begin to dominate the US marketplace. There was now little hope for the PET computer in North America. * * * After the success of the VIC-20, several Commodore employees ascended into Tramiel’s inner family. “It was approximately the top 20 trusted people in the company,” says Tomczyk. “It included everyone from people with no title who worked in engineering to people who had titles like vice president. It didn’t matter what your title was. It was the people that Jack really trusted and who had a record of achievement.” John Feagans had always been important to the success of the PET. After he refused to depart with Peddle, Tramiel reopened the Moorpark R&D center. “After the Peddle team left, I remained in the Moorpark R&D office,” he says. “I was soon ed by a new project which was a tape drive mass storage device based on VCR diagonal striping technology.” Commodore also continued developing a hard drive. Feagans’ loyalty to Tramiel was beyond question. He began to enjoy rewards reserved for inner family , including flying on the corporate jet in November 1980. “Jack flew me and my wife to Las Vegas for Comdex, and also our honeymoon, on the loaner PET Jet after the belly landing in Iowa,” says

Feagans. “That was my first meeting with Jim Finke.” At the newly reopened Moorpark offices, where the disk drive designers now resided, Feagans continued working on the concepts introduced to him by Robert Metcalfe, including networking and graphical interfaces. The others began work on a data storage device using VCR tape. Robert Russell, the young engineer hired in 1979, entered Tramiel’s inner family after his heroic efforts making the VIC-20 a reality. Russell worked out of an office at 3330 Scott Boulevard, in the same building as Tramiel. This meant Tramiel often dropped in on his engineers to monitor the progress of his projects. “When I got some things done, he liked me,” says Russell. “He kind of dragged me around. It was real exciting for a young guy out of college being dragged around by the president of a company to shows in Vegas and meeting people who were going to be presidents of the company and flying around in the jet.” Al Charpentier and Charles Winterble also entered the inner family after the VIC-20 success, due in no small part to the video chip engineered by Charpentier. The gesture meant a lot to Winterble. “He asked about you, about your goals, about your family, and what you wanted to do,” he says. “You had special privileges and he would care a little bit more about you.” Tramiel educated of his inner family in how he wanted them to do business. “When you worked with Jack, as I did in my job, you had to learn the business parts of it,” says Winterble. “In his words, he told me he wanted to teach me two things: he wanted to teach me to be Japanese and he wanted to teach me to be Jewish. That was his training program for me, and he did it! I learned about business and a lot of things in the consumer marketplace from him directly or indirectly. He sent me to Japan and had me working with a bunch of these people to learn. It was interesting.” Kit Spencer continued his record of success within Commodore, and remained a valued inner family member. “I was pretty close to Jack,” says Spencer. “I worked with Jack for well over a decade. This was well before Charlie ed Commodore and in those days, we were a pretty small company compared to what we became.

Tramiel made sure his employees knew that no matter what their position in the company, he rewarded them based on how much they contributed to Commodore’s success. “Too many people looked for positions to be appointed to and titles, but it wasn’t really that sort of company because it was changing too fast,” says Spencer. “There were jobs that needed doing and you got remunerated very much as a person rather than as a position.” Every year, Tramiel invited his inner family to a meeting in his office. “After I was there about a year, I was called into Jack’s office,” recalls Tomczyk. “I was standing in the hallway with a bunch of other people. They were all people like Kit Spencer, the general manager of England, and the manufacturing VP, and chip designers, and John Feagans.” Tomczyk wondered why everyone was waiting outside. “I showed up for the meeting with Jack, and everybody looked at me and said, ‘Congratulations!’ I said, ‘For what?’ They said, ‘You’ll see.’” The gesture was beyond anything Tomczyk had anticipated. “The fact that I was there meant I had been accepted into the family,” he says. “Everyone in the family got a 100% bonus and extra stock options, and that’s what Jack called the family.” * * * Peddle’s departure from Commodore marked a change in his personal life. His second wife, Shirley, had been involved with Commodore since he ed. Without Commodore, something was missing. “We had done all these wonderful things,” recalls Peddle. “She turned forty and decided that she wanted to try other things. I was starting a new company, so she went her way and I went mine.” When Tramiel found out Peddle was stealing Commodore employees, he became vengeful. “I take [almost] the whole team and go out and start another company,” says Peddle. “He was really pissed about that.” It was an open declaration of war. Since his return from Apple, Peddle had accumulated approximately three million dollars worth of Commodore stock, often through stock options that would mature in the future. The executives had a term for the arrangement, because it enticed them to remain with the company. “We used to call it the

golden handcuffs,” says Spencer. “Most of the key people who stayed at Commodore for a while were decently well rewarded. It wasn’t an overnight thing usually.” A month after Peddle left Commodore, Tramiel sued both Sirius Systems and Chuck Peddle personally. Forgiveness was not part of Tramiel’s belief system and it was no sin for him to pursue vengeful retribution on former employees through lawsuits. Tramiel claimed Peddle was in possession of stock certificates that he had not earned. “There were two suits,” explains Peddle. “One was to fuck me personally and the other was to just screw us up.” The lawsuits surprised Seiler, because they had carefully departed Commodore. “Even then he got sued by Jack,” he recalls. “There were like five pieces to it, like predatory hiring and stealing trade secrets. The most important one to Jack was to get back at Chuck and get his stock back.” Tramiel potentially had a case for stealing trade secrets. Peddle had certainly developed the idea of a new architecture in discussions with John Feagans and others while at Commodore. He even pitched the concept to Commodore executives in London in April 1980. However, Tramiel gave up his claim when he rejected Peddle’s proposal. The lawsuit dragged everyone into court. “I going to all these depositions and having to testify against Jack when we were at Sirius,” says Seiler. “They said we stole a bunch of technology. No! It was a completely different architecture. It was an 8-bit machine, this was a 16-bit machine. We redid the whole display. We did everything over.” Peddle maintained Sirius Systems had not infringed or used any intellectual property from Commodore. “There was no way we were possibly infringing on anything we did at Commodore,” he says. “What we were doing was totally different.” He decided to hit back at Commodore. “I countersued them for $10 million for spurious lawsuits. My $10 million countersuit had some merit because they were just persecuting me.” Even though the two sides attacked each other with lawsuits, Peddle and his partners maintained a friendly demeanor in the press, for a while. “We didn’t say anything bad about him,” says Peddle. “If you read the press, we said only complementary things.”

Unfortunately, one of the Sirius partners did something Peddle now regrets. “This is now getting real serious and we made a real bad screw-up—a seriously, seriously, dumb-assed, stupid screw-up. My partner [Chris Fish] got all mad because he sued us, and he said, ‘I’ve got information that can put Jack and Irving in jail and I’m going to put it on the line.’” Fish attempted to intimidate Tramiel and Gould out of the lawsuit. “In order to keep it from getting into the legal system, but to send a serious warning, we gave an interview to a guy who published effectively a bunch of stories about Jack in Barron’s,” recalls Peddle. The articles chronicled Tramiel’s past with the Atlantic Acceptance scandal and Gould’s dealings in the Bahamas. Bob Yannes recalls the effect the articles had on Tramiel. “There was a horrible write-up of him in Barron’s that just made him livid,” he says. “They were pretty scathing.” “So at this point we had taken a relatively bad situation and made it infinitely worse,” explains Peddle. “Now we had thrown down the gauntlet and said this is really personal between us and Tramiel.” Unfortunately, the allegations had no substance. “It turned out my buddy was just blowing fucking smoke,” says Peddle. “He didn’t have hard evidence. He had his speculation, which ultimately he recanted. So all it did was to really inflame these guys with no benefit.” The court case was to take place before a 12-member jury. “I was supposed to meet with Tramiel because the courts wanted us to settle the case two weeks before that time,” says Peddle. “But they scheduled it during the same time I was launching our new company in Vegas.” Tramiel and his lawyers showed up for the settlement hearing with no one to talk with. “I think Jack went to that settlement conference, having his lawyers telling him he was never going to win the case, and that we hadn’t stolen anything from him,” says Peddle. “I think he went to that settlement conference to make up.” Peddle its he could have attended the meeting. “The real truth probably is that I was so fucking confident I was going to win I was telling him to stuff it,” reveals Peddle. “I think not showing up for that—insulting Jack—was the worst thing we did.”

“I could have settled it probably by paying their legal expenses,” says Peddle. “It would have been like a quarter of a million dollars or half a million dollars. I had four million dollars worth of stock that day. That deal was on the table, and I never even went in and looked at it because I was so sure I was going to win.”[3] Peddle had previously given up his stock position at Commodore to move to Apple, and then given up his Apple stock options to return to Commodore. Kit Spencer recalls Peddle’s missteps with his stock position. “Chuck in the end didn’t make as much as he should have done. He did things badly by leaving and giving up things. Chuck had a habit of shooting himself in the foot, I’m afraid.” Peddle blames his mistakes on his righteous attitude at the time. “I was legally right and I was morally right. When you are that right you get your judgment fucked up,” he says. “Right doesn’t matter. I made terrible judgmental decisions about how I dealt with the law and I dealt with Jack because I was so convinced I was correct.” It would be up to a jury to decide if Peddle was entitled to his shares. The date was set for February 1981. [1] Hong later sued Commodore. “This guy wound up with zero stock in Commodore and zero stock in Apple,” explains Peddle. “He was pretty desperate. He actually sued Commodore, but I don’t think it worked out. Jack was pretty clever about how he did those things.” [2] The word pet means something else entirely in French. “Andre, who is a Frenchman, knew that pet in French is [fart],” explains Peddle. “So in , it was always P.E.T., with dots.” The association did not help Commodore’s marketing effort. [3] Peddle alternates between two, three, four, and in one interview even five million dollars, presumably because of the rising price of Commodore stock throughout the period.

CHAPTER 21

Failure to Launch 1981

In January 1981, Commodore was in full force at CES Las Vegas. Jack Tramiel’s recent acquisition, Micro Display Systems, had produced several different digital watches for the booth displays. Some had a black leather wrist-strap with a silver body and a black face with the Commodore logo. Others had adjustable metal wristbands and gold, oval shaped faces. Some models had built-in lights and automatic alarms, which were state of the art in the early eighties. Commodore pushed hard to sell the watches, and dealers had to purchase them in large quantities. The Commodore name was not one people associated with watches, however, and the products were not selling as well as Tramiel wished. Tony Tokai attended the show and brought along a few of the amazing HAL Laboratory games for the VIC-1001. The executives were curious how the VIC1001 fared in the past months. “Tony said it was wildly successful,” recalls Michael Tomczyk. According to Yash Terakura, “We know that because I was on the production side and we made quite a bit.” The North American computer industry fully expected Japanese companies to enter the personal computer market in 1981. However, as Commodore employees visited the NEC booth, there were no computers on display. Many believed the VIC-20 release in Japan a few months earlier caused NEC and other companies to withdraw their plans in the face of such cheap American computers. “The proof is that no significant Japanese home computer was introduced into the US market throughout those years,” explains Tomczyk.

“There’s no reason why Apple would keep NEC from coming into the market, since NEC and other computers like it presumably were in the same price range. It was purely a price-performance impact.” Tomczyk believes the VIC-1001 caused the Japanese to go into an 18-month product development cycle. “They just had to go back and rethink the whole thing. In those days, they were being very careful before they did anything, and we knew that.” Commodore boldly displayed the VIC-20 computers, which were in fact Japanese VIC-1001 computers with different labels. “We showed it to everybody,” says Charles Winterble. “We had crowds lined up. It was just phenomenal.” In the back room, Tramiel talked with potential customers with his cigar in hand, bellowing smoke like a steam-powered CEO. “Jack would frequently talk,” says Winterble. “There would be a crowd brought into the back room—investor types, the key retailers, the key sales guys, key customers.” Tramiel frequently repeated his mantra, “Computers for the masses, not the classes.” Tramiel relied on his east coast engineers to demonstrate the product. “Jack would do an introduction and then frequently me and Bob [Yannes] would put on a dog and pony show,” says Winterble. “I would give an overview of the technology, and if there was an engineer asking about bits and bytes it went to Bob.” Sam Tramiel was temporarily home from his punishment tour in Japan and he walked the floors of CES with his father discussing the competition. During one stroll, they ran into their supplier of BASIC. “Jack would often bump into Bill Gates and he would say, ‘You got that so cheaply,’” recalls Spencer. “Jack would say, ‘No we didn’t. If we hadn’t bought it you wouldn’t have been established as a company and as a standard.’ And both things are quite true. Essentially I think it was a great deal for both companies.” Commodore also displayed an offshoot of the CBM 8032 called the PET 3220. It was an electronic cash , something of a novelty in the early eighties. The cash was an 8032 with a smaller green-screen monitor, dot matrix printer, and cash keypad. “Basically it was a PET in a steel box on top of a cash drawer,” says John Feagans. “There was a receipt printer connected to it.”

The cash idea came from Chuck Peddle’s former adversary. “Jerome Guinn was the head of Commodore Cash Company. Prior to that, he was the GM in Santa Clara,” says Feagans. “The engineers I dealt with were Benny Pruden, Frank Hughes, and Jeff Bruett.” The team formed in August 1980 to create the product in Dallas. At the close of CES, Tramiel put Commodore into high gear to get the VIC-20 ready for North America. The task was immense. A factory would need to produce them in quantity. The computer needed a box design. Someone would have to write a proper manual. A library of launch titles would need to be prepared, with peripherals such as disk drives and printers. Commodore would have to fulfill thousands of dealer orders and someone would have to design an advertising campaign. It would be up to Tomczyk to develop advertising. Tramiel gave him two weeks to develop a presentation.

Frank Hughes (center) reclining with of his cash team in front of Commodore’s Dallas facility.

After CES, a few thousand VIC-20 computers began arriving from Japan in plain blue and white boxes. The Japanese had sold all the VIC-1001s they could manufacture for the Christmas season. It was time for Tomczyk to put his skills to work. * * * Tramiel allowed of his inner circle to formulate their own plans, but he regularly visited them to monitor their progress. “When Jack was in, the whole place was chaos,” says Winterble. “He was a terrific hands-on manager. Some people liked it and some people hated it.” Tramiel walked down the hall, popping into one office after another while leaving a trail of cigar ashes. He liked to stay in touch with every aspect of Commodore. Tomczyk recalls, “He used to make a motion by pushing his fingers together with his thumb and say, ‘You have to touch and feel and smell everything. You have to be in .’” Most executives enjoyed their interactions with Tramiel. “Jack has a way of making everybody feel extremely important,” says Winterble. He could also just as easily tear employees down in a Jack Attack. “If you didn’t deliver, it was Jack’s way of getting your attention,” says Winterble. “I doubt if there was ever a Jack Attack for somebody who delivered what they said they would deliver. You were allowed to screw up. If you screwed up stupidly or screwed up multiple times, that is what he reacted to. Just making an honest mistake would not get you a Jack Attack.” Everyone in Commodore knew all about Jack Attacks. “I have certainly seen— and more heard than seen—Jack have what became known as Jack Attacks,” recalls Neil Harris. “Really pounding the desk and yelling at people. Quite often, it was deserved. Personally my management style is a little different from that,

although I’m not worth what Jack became worth so maybe his way is more effective.” Michael Tomczyk had risen too high, too fast. As a result, he did not possess the proper experience to perform his job at the high level it required. Under pressure, Tomczyk managed to contract pneumonia but decided to work through the illness. Two weeks after CES, Tomczyk entered Tramiel’s office to present his marketing plan. Despite the sickness, he was upbeat and eager to receive Tramiel’s approval. His day was about to turn into a nightmare. Typically, a representative from an ad agency delivers part of a marketing presentation. The director of marketing also usually brings along some of his staff to suggest alternatives. Tomczyk’s inexperience showed. He went in alone, with no one to him or his ideas. To make things worse, his illness made it appear to Tramiel that he was indecisive. “You couldn’t bullshit him at all, and particularly in areas of marketing and sales, which he knew cold,” says Winterble. Tramiel smelled blood. “If you went to Jack and you were prepared for what you were going to do, you were fine,” says Harris, a one-time subordinate of Tomczyk. “If he asked you a question and you didn’t know the answer, and if you said, ‘Look, I don’t know the answer to that but I will get it for you within 24 hours,’ you were fine. If you went in and just started doing what people do, heaven help you.” Tomczyk started his presentation by showing a picture of a proposed box cover, with the theme “Lightning Strikes Twice” (referring to the previous PET success). The proposed box cover borrowed the famous image from Stanley Kubrick’s 2001: A Space Odyssey, showing the earth at the bottom of the frame with the VIC-20 floating in space and a bolt of lightning through the VIC-20. “He was already prejudiced when I had my Jack Attack,” says Tomczyk. “I believe somebody had gone behind the scenes and was jockeying to be in charge of my area, and had convinced him that I was not doing the job. I don’t think it was based on what I was presenting because he was almost totally ignoring my presentation. Somebody had gotten to him and Jack was someone who, once in a while, would let the politics get very weird.” Tramiel found the box derivative and amateurish, saying, “It’s not friendly! What’s friendly about lightning? People get struck by lightning.”[1] Tramiel

thought for a moment, and said, “I want to show people on the box, enjoying their computer. That, to me, is friendly.” Tramiel suggested the possibility of a spokesman for Commodore. “I haven’t given it much thought,” was Tomczyk’s reply. This seemed to set Tramiel off, who proceeded to criticize Tomczyk for failing to prepare. He then asked Tomczyk to continue with his presentation. Tomczyk showed the rest of his advertising images, which were all based around the lightning concept, which Tramiel had already dismissed. Tramiel was growing impatient. In Vietnam, Tomczyk had never encountered anything as ferocious as the Jack Attack he was about to receive. The door was open for everyone to hear. Charles Winterble often heard Jack Attacks echoing through the offices of Commodore. “A guy had to listen a lot,” says Winterble. “He didn’t get a lot of chances to talk. He would suffer the indignation of being told how poorly he was doing and he’d have to listen to it for a while.” Just down the hall, around the corner, the newly hired marketing people stopped their work to listen to the booming tirade. “I can’t explain exactly what it is like. I can’t even all the details of the Jack Attack,” says Tomczyk. “In my case, it was very irrational. He would give you a moniker, like you want to be an educator or you want to teach. You should be doing more. He would put a handle on you of some sort. But it wasn’t personal like a personality attack, he would characterize you in some way.” After what seemed like an eternity, the meeting ended. Though Tomczyk felt relieved to be out of the room, it was not over. This was just day one of his Jack Attack. As Tomczyk entered his office, several of his new employees asked if Tramiel would fire him. Tomczyk smiled reassuringly and told them, “No, I’m not going to be fired. It’s only a Jack Attack. It happens to the best of us.” He was correct on all counts. “If you made a mistake, you had to be open about it,” says Harris. “If there was something going wrong, you wanted to let somebody know so that it could get fixed in advance. Most people in political organizations don’t do that. You didn’t want to go to Jack and tell him a lie or make up some bullshit.”

Day two of the Jack Attack was even more depressing for Tomczyk. Tramiel threw him out as marketing director. “If you were in marketing at Commodore, you had a tough job,” says Winterble. “You had to deal directly with Tramiel, you took abuse, and you didn’t last very long.” Tramiel sent him on a mini-punishment tour by banishing him from the company headquarters. He took away his office and sent him to work in the crowded consumer products division. Compared to his old office, it was like working from a boiler room. Tramiel was now out to find someone else to launch the product. * * * Although Commodore was on the verge of success with a new product, the departure of key engineers the previous year had bruised the company. In early 1981, Commodore began to rebuild. They started with the PET Jet. “The plane was not severely damaged and was rebuilt,” says John Feagans. “I flew on the inaugural flight of the rebuilt PET Jet.” Surprisingly, the engers on the ill-fated flight in September were not reluctant to board her again. “I took the first flight on it when it was repaired— the second maiden flight,” laughs Dick Sanford. “It didn’t bother me. I look at it this way: pilots want to go home too. No pilot is going to go up there knowing he is taking a risk. If something is wrong with that airplane, he’s not going in it. That’s my attitude.” One question remained. Was Commodore still airworthy without Chuck Peddle? After Peddle left, the MOS Technology engineers began planning a new line of computers. “As far as Jack was concerned, the future of product development was going to come out of the east coast,” says Robert Russell. To motivate his employees, Tramiel created a bonus plan. “Commodore had put together a profit sharing plan for the employees where they had put a chunk of stock aside,” explains Russell. “What they would do is take the increase of the value of the stock and then divide it up among the employees proportional to their salaries.” Tramiel offered the plan only to his east coast engineers such as Bob Yannes, leaving Russell and the other west coast engineers out. Even though Commodore was in the midst of launching the VIC-20, Tramiel knew Commodore would need a new product after that to show at the January

1982 CES. “The VIC-20 had not achieved the kind of success that it eventually did,” says Al Charpentier. “It was just coming to market, so it was time to do another project.” No marketing input went into the planning of the new computer. “Usually we didn’t see the engineers while they were building something new,” says Neil Harris. “It was a very different process than I’ve seen in most companies—very engineering driven. Usually there is some input from marketing on what the market wants. I don’t think in that company there was any orientation towards that at all.” The engineers at MOS Technology knew they could not rely on the technologically inferior VIC-20 for long, so they began to discuss an improved 40-column color computer. “We said, ‘Okay, let’s talk about the VIC-40, quietly,’” says Winterble. Winterble and Charpentier discussed their plans away from Commodore. “Al and I were close friends and we would go to a little restaurant, Friday nights particularly, and drink too much wine,” recalls Winterble. “When we went there, we said, ‘What are we going to do? We’ve got this fantastic design area and we’re doing nothing with it.’” Winterble was trying to figure out a way to make MOS Technology profitable. “We couldn’t get MOS Technology sales to take off, so we needed some great stuff,” he explains. “Out of that came a bunch of chips.” The first chip was a revised 6502 processor. “We took the 6502 processor and made it into a core, and we could drop that into a whole bunch of different processors. Those were pretty much just cookie cutter designs,” says Winterble. The 6510 processor emerged from their talks. “The 6510 was just a 6502 with I/O on it.” They also proposed a more advanced chip to handle peripherals. “We did a 6526 chip, which was the CIA,” says Winterble. The CIA chip controlled peripherals, much like the PIA chip. “We called it Charlie’s Interface Adapter because I did the architecture on it,” says Charpentier. “It was later the Complex Interface Adapter. I started it, and Yannes finished it and did it right.”

The most exciting discussions were about sound and graphics. After two previous attempts, Charpentier and his team had failed to deliver a workable 40character color chip for the TOI. Now, in early 1981, Winterble was going to give Charpentier a chance to do it right. He told Charpentier, “The time has come for us to do a 40-character VIC-chip.” Rather than attempting to build from the VIC chip, Charpentier wanted to start from scratch. “Charpentier stepped back and said, ‘We’ve got to think this through from the beginning again,’” says Bill Seiler. “He came up with a different way to do it.” The engineers wanted to target the entertainment market. “It was to be a game chip,” says Winterble. “It would also be a computer, no problem, but games were hot back then. We said, ‘Let’s do the world’s greatest game chip.’” Winterble presented the project to Jack Tramiel. “I asked Jack if we could talk to him,” says Winterble. “I’m sitting in my office, which is on the second floor of MOS Technology.” Winterble heard Tramiel’s familiar voice in the hallway. “All of a sudden he’s in my office sitting down. He’s got this great big cigar and he’s flipping ashes on my floor. I always hated that damn cigar, but you didn’t complain.” Winterble took a breath and presented his case. He told Tramiel, “Look, we’ve got an idea. We want to do a 40-character game chip and it can also be a computer. We want to organize a project to do it.” “I didn’t even really finish talking and he said, ‘Do it,’” laughs Winterble. Winterble was still stinging from the failure of two previous graphics chips. This time, he wanted to make sure they planned the project. “I said, ‘Well wait a minute, Jack. We want to do a study on this thing. What I’d like to do is really dig into it, look at the competition, come up with an analysis, and do a businessplan. You know, do the whole thing the right way.” Tramiel realized his engineer was asking for time to plan the product. “He looks at me and says, ‘Okay, how much time do you need?’ We talked about it and he said, ‘I’ll give you ten days.’” Charpentier realized if he wanted to create an appealing chip for the marketplace, it would need to sur everything else available. To study the

competition, Charpentier and Bob Yannes played video games on the best game systems of the time. “We bought every computer we could get our hands on. At that time, it was game systems: Mattel Intellivision and all the Atari stuff,” says Winterble. Surprisingly, the Intellivision was their biggest influence. They also studied the Texas Instruments 99/4A. During this time, they compiled a list of all the features they wanted their chip to incorporate. “The entire resources of our little facility there at MOS [were occupied] for ten days,” recalls Winterble. “We took apart everything we could, analyzed the capabilities and tried to project the capabilities [into the future].” Charpentier was very interested in developing a graphical capability called sprites. Sprites are two-dimensional moving objects used in games. The process is fundamentally similar to cartoon animation, where artists create an object on cellophane and then overlay the object on top of a background. Programmers can create sprites without any specialized hardware, but the results are often slow and flickering. Charpentier wanted to create hardware-accelerated sprites. The engineers studied sprites in other systems. The Atari sprites, called “playermissile graphics,” failed to impress Yannes. “They are terrible,” he says. “They don’t have enough colors, they aren’t big enough, and they don’t do what you want.” In contrast, Yannes was impressed with the sprite capabilities of the Ti-99/4A with the Texas Instruments Extended BASIC cartridge. Their sprites could smoothly move to any point on the screen, and once set in motion, the sprites continued to move without any further program control. However, Yannes saw room for improvement. The Texas Instruments sprites were only one color and they were small. The MOS Technology engineers realized they had to improve on the sprites or else they would merely have the same technology as their competitors but two years behind. They would have to predict the next step in graphics technology. Al Charpentier originated most of the ideas and discussed them with Winterble and Yannes. “Al was really into video architecture,” says Winterble. “We had many a late night discussing the architecture of that VIC chip. We’d talk in the abstract about sprites; how many sprites, how the sprites would work, and how

the interrupts would work.” Charpentier was determined the chip would work with DRAM this time. “There’s only so fast you can go hitting the memory because you are sharing memory with the processor and other things that have to use it,” says Seiler. DRAM was much cheaper than static RAM because the circuitry took up much less space on a chip. “They were using dynamic RAM, which is radically cheaper than static RAM,” says Seiler. “The VIC-20 used static memory, and that was fairly expensive. A static memory cell uses five or six transistors, whereas a dynamic cell uses one for each memory bit. So you get a lot more memory in the same chip area on a dynamic RAM.” The engineers quickly decided their chip would multi-color sprites, which could move independently with minimal programmer intervention. They decided to better Atari and include eight sprites, compared to four. The engineers also devised some ingenious features to enhance sprite capability. Automatic collision detection would help programmers, since most games relied on collisions. They also wanted to make it possible for each sprite to double in size. Winterble compiled a massive report. “We put together great color photos and this thick report,” says Winterble. “We had gotten two hours sleep in ten days.” Their report was a proposal for a video game console. “The project that we presented to Jack for the VIC-II chip was actually to do a high-end video game system,” says Charpentier. Ten days after their meeting, Tramiel returned. “He sits down in exactly the same chair with another damn cigar, and I go through [the proposal],” explains Winterble. “It’s an inch thick and he’s paging through the pictures, looking at it, looking up at me, looking at it paging through it at about a page a minute. He reads about one sentence, looks up at me, and says, ‘Ok, damn it, now will you do it?’ He was just appeasing us.” Although Tramiel barely acknowledged the report, the research gave the engineers a clear goal for their next set of chips. Tramiel instructed Winterble to keep the project a secret within the company. “No one was to be told about it,” recalls Winterble. “In particular, marketing was not to be told about it. Jack’s statement to me was, ‘You’re marketing. It’s engineering driven and when it’s done we’ll tell marketing about it.’”

Tramiel allowed Winterble and his team complete freedom. “It was kind of funny, because I wanted to work with a budget,” says Winterble. “I had this business training and I wanted to use it for something.” When Winterble asked for a budget, Tramiel told him, “I’m not going to give you budgets because budgets are a license to steal. I’ll watch you, and if you spend too much money, I’ll stomp on you.” Winterble knew it was a once in a lifetime opportunity, free from marketing and management interference. “This guy turned over the company to us. It was amazing. He gave us the keys.” * * * While Charpentier worked on his video chip, Bob Yannes designed the architecture for a new sound chip. The 24-year-old Pennsylvania native had been dreaming of creating a dedicated sound chip since he ed MOS Technology years earlier. Yannes’ fellow engineers at MOS Technology liked and respected the quiet engineer. “Bob was a fair haired, bright, sharp young puppy,” says his best friend at the time, Robert Russell. “He did his work and wasn’t pushing to get his name on everything and take credit for things.” Throughout his early life, Yannes looked far younger than he actually was. “I think probably everyone just thought I was a little kid,” he says. The young engineer took an early interest in electronics, using what little money he had to purchase parts and kits from the local Radio Shack store. “Being a child of the sixties, I was into pretty much any technical stuff.” The young Yannes purchased a clock chip from Radio Shack. “One of the first electronic things I built was a digital clock,” he recalls. Using the instructions included with the clock chip, Yannes acquired the parts and wired them together on a perforated board. “I it had a really big LED display and it would overheat and then lock up. But for a first project, it worked at least.” He was also into photography and built his own darkroom. He merged his two interests by creating electronics to help with the photographic processing. “I built my own darkroom timer,” he says. Instead of an amateurish perforated

board, Yannes created his own etched circuit board for the project. However, the naïve hobbyist made a few mistakes. “One thing I didn’t realize was the etchant that they used, ferric chloride, reacts with aluminum very violently. I was using an aluminum foil pan to hold it, so suddenly it started frothing and foaming and smelling really bad. I learned my lesson with that and went to glass dishes for etching circuits.” Yannes was lucky to experience computing at an early age. “The first computer I ever played with was in junior high school, which was a time sharing system,” he recalls. In the early 1970s, while in high school, the new sounds of electronic music captured Yannes’ attention. The innovative movement was full of pioneers exploring the musical potential of new technologies. The sounds influenced the direction Yannes’ life would take. One song influenced Yannes more than any other. “Lucky Man from ELP (Emerson Lake and Palmer)—the synthesizer solo in that,” says Yannes. “When that came out, we all asked, ‘What made that sound?’ That drove a lot of people to find out what it was, and when they found out what it was they developed an interest in it.” As his interest in electronic music grew, Yannes began listening to electronic music pioneers from and Britain, such as Kraftwerk and Mike Oldfield. “I listened to all of them,” says Yannes. “I had their albums.” In 1974, Yannes discovered he could order analog music synthesizer kits. Due to his severely limited budget, Yannes settled on a company called PAiA Electronics in Oklahoma. “I had no money at all,” he recalls. “The PAiA kits were kind of crude but they were very inexpensive, which is why I could afford them. There were a few other companies making kits, but they were thousands and thousands of dollars.” The first PAiA synthesizer Yannes built was unimpressive. “They had their little Gnome synthesizer, which was forty bucks. It wasn’t really a music instrument but it made some interesting sounds,” he says. Crude as the kits were, Yannes was learning valuable lessons about sound. After high school, Yannes attended Villanova University. The same year, MITS released the low-cost Altair 8800 microcomputer. “When the first Altair 8800

computer came out back in ’75, a friend of mine got one and that really got me into microcomputers,” he says. Yannes began devouring computer magazines. “Byte was a great magazine,” he recalls. “In the early days, it was pure hackers and hobbyists. It wasn’t a mainstream magazine but it was published like one. It was big and thick on shiny stock but it was hardly polished. The early days were like the Wild West frontier.” In college, Yannes earned his tuition by assembling computers. “One of the things I would do to earn some money was to build computer kits for the science lab or the computer lab,” he recalls. “At a local Byte Shop, I built these systems and got them working. There were a lot of people coming in who wanted computers but they weren’t interested in building their own and particularly debugging it because a lot of them weren’t very reliable designs back then.” Using some of his earnings, Yannes purchased a more expensive synthesizer kit from PAiA. “I ended up getting one of their modular systems when I was in college,” he says. “That one was actually a useful musical instrument. I made some recordings with that.” Yannes was somewhat disappointed with the course material in university. “I was really interested in microprocessors and most schools are fairly behind the curve when it comes to technology,” he explains. “In one of our classes, we were studying the Intel 4004 processor. I was just so far ahead of that already, so I negotiated with my professors to do independent study courses and independent project courses.” Yannes favored the 6502 processor over the others. “It was one that I could actually afford, which was great. You could get an 8080 for $200 or you could get a 6502 for $25,” he recalls. “Back then, you tended to develop a particular fixation with a processor. You had your [Intel] 8080 people, you had your [MOS] 6502 people, and your [Motorola] 6800 people and they would get into violent arguments over which one was the better processor. Kind of like Ford and Chevy.” Music led Yannes to experiment more with computers. “I’m not a good musician at all, but I was very much interested in making synthesized music,” he says. “I hoped to have a computer with a multi-track sequencer that I could play the

information into and then edit it to correct my mistakes. I was also just interested in computer graphics. I thought it was really cool what you could do on a computer screen.” In 1978, his final year at Villanova, Al Charpentier visited Yannes on a recruiting drive. “He interviewed me on campus,” says Yannes. “I was aware of MOS Technology because I had been playing with the 6502 processor for a couple of years by then.” Yannes could not turn down the opportunity to work for the company that made his favorite microprocessor. “I basically said, ‘Okay, I want to go work for these guys.’ Al said one of the reasons he hired me is he figured someday my knowledge of music synthesis would be beneficial to the company. Al is a very smart man.” Yannes quickly learned some harsh realities about working in a competitive environment. “The first day I went to work at MOS Technology, I walk in and I have my own office and my own desk, and I went, ‘Wow!’ I opened up my desk drawer and in it was a letter laying off the person who previously had the office,” he recalls. “I really didn’t know what I was getting into, to be honest, but it’s what I wanted to do and I certainly have no regrets.” Yannes never thought of himself as a Commodore employee. “I always kind of viewed that I was working for MOS Technology, which was fine to work for,” he says. “Mostly, working for Commodore was fine. Probably they didn’t reward their employees as well as they should have.” For the next few years, Yannes’ mind soaked up knowledge of semiconductor design. Because of his youthful appearance, he was sometimes mistaken for a teenager who had wandered into the labs. “It certainly doesn’t help earn respect,” he says. According to Russell, the work at MOS Technology was not what Yannes wanted. “He was a frustrated musician,” he says. “He wanted to do sound stuff so bad, but he was basically working on the other stuff because he had to.” Yannes had an opportunity to use his knowledge when he improved the sound capabilities of the 6560 VIC chip. Then, in early 1981, Winterble finally proposed a sound chip for Commodore. Yannes felt like his whole life had been building up to this project.

[1] Michael Tomczyk, The Home Computer Wars (Compute! Publications, 1984), p. 195.

CHAPTER 22

The Friendly Computer 1981

With Michael Tomczyk no longer in charge of the VIC-20 product launch, Jack Tramiel temporarily assigned the task to his retail sales executive, Ken Hollandsworth. “The sales division was oriented around a typical consumer electronics sales model, where you would find retail stores and retail chains,” says Neil Harris. “And this was a product designed for the mass market.” For the time being, it would be up to Hollandsworth to oversee launch preparations. Commodore would attempt to sell the VIC-20 in the same markets they sold their watches and calculators, rather than only specialty computer stores. Tramiel attempted to court his Commodore UK marketing director to leave his post and come to the United States, but Spencer was uninterested. Spencer recalls, “I’d go for meetings there and he would often say, ‘Can you stay on for a couple of weeks and do a few things before you go back?’ and I said, ‘Yeah, but I’m going back, Jack.’” In early 1981, Tramiel gave his UK marketing director a larger role as head of European operations. Spencer relocated from the UK to Switzerland, with his wife and daughter slated to move once the school year ended. “I had actually already moved to Switzerland to start running the European marketing operations,” recalls Spencer. Until production increased to sufficient numbers, Commodore would sell the VIC-20 to North American computer dealers. “They started out selling them through computer stores like ComputerLand and places like that,” explains Bob

Yannes. A few days after his Jack Attack, Tomczyk moved into his new workspace at the cramped Consumer Products Division. There, he squeezed into a small desk near the sales order clerks. However, he was not off the VIC-20 project completely. Hollandsworth consulted with Kit Spencer, who had success launching the PET 2001 in the UK. Hollandsworth decided to follow Kit’s template for launching a computer, and he gave Tomczyk the task of creating a manual and launch software for the VIC-20. For the early orders to computer dealers, Commodore included the Japanese manual translated into English by John Feagans’ wife, but for retail stores the computer would require a far better manual. “We all agreed that one of the things that no computer had that was desperately needed was a very detailed manual,” says Tomczyk. The manual would be included with every VIC-20, and Tomczyk was determined to keep the material fun. The word “” was not in wide use at the time, so instead of calling it the manual, he named it, “Personal Computing on the VIC-20: a friendly computer guide.” The exiled Tomczyk had no employees at the time to work on the manual, so he hired outside consultants at first. “He wanted a friendly manual and Commodore at that time really didn’t have the kind of people that would do that kind of thing,” explains Andy Finkel. Instead, Tomczyk went to Bob Albrecht and George Firedrake, who had previously founded the People’s Computer Company and wrote their newsletter. “Some of the people who were ex-People’s Computer Company guys collaborated with Commodore on the original manual because we wanted something really friendly,” says Finkel. “That was an interesting experience because those guys go way back.” Tomczyk sent the rough translation of the Japanese manual to the writers in Menlo Park. They already owned several PET computers which helped then write the BASIC sections of their book, but Commodore also gave them an early VIC-20 computer. The PCC founders began the process of making computers easy for ordinary people. A month later, Tomczyk reviewed the early manuscript and was not happy with

the results. The manual was too friendly, to the point of uselessness. Tomczyk asked permission from Hollandsworth to hire programmers to refine the manual. “This was kind of a new thing for Commodore,” recalls Finkel. “Usually they had all the technical people in the engineering department, and marketing people in the marketing department. As the launch progressed, they decided they needed technical people in the marketing team if they were going to do a successful launch to consumers.” In early February 1981, Tomczyk hired Neil Harris, the Cornell University dropout who previously worked at a Mr. Calculator store in Philadelphia selling PET computers. Harris would become a key player in the push to get the VIC-20 into retail stores. Harris was a proficient BASIC programmer and often developed software on contract. Programmers tend to be introverts, but Harris was a rare extrovert. “I found that temperamentally I couldn’t really work as a programmer for more than six months doing only programming before I was ready to go stir crazy,” he recalls. “So I would do sales work which didn’t pay as well but at least I got out to talk to people more.” As a Mr. Calculator employee, he knew about Commodore’s VIC-20 release. “I had heard Commodore was launching an inexpensive color computer which spoke PET BASIC, which sounded like an amazing thing to me.” In early February, Harris received a tip about a new job opening at Commodore. “My mother called on the phone one day and said that she had heard on a talk show on the radio that this company called Commodore was having an open house for hiring.”

Neil Harris programming on a PET computer.

Harris was unsure if he was qualified to work at Commodore, but there was nothing to lose by applying. “I went out to the job fair at a hotel in King of Prussia, Pennsylvania,” he recalls. “I spoke to the human resources manager and gave her a quick description of my background. Her eyes got really big because I was already conversant in Commodore computers. She said, ‘I need to introduce you to Mike Tomczyk right now!’ She grabbed me by the arm and dragged me over to Mike, and he hired me pretty much on the spot.” Even in 1981, Commodore had a tough reputation. “When I ed Commodore after working for Mr. Calculator and being around the industry, the word on the street was that Commodore was a really hard place to work with a lot of turnover,” recalls Harris. “If you last a year at Commodore it is an amazing achievement. When I ed the company, that was my goal: ‘I’m going to last more than a year here.’” Harris worked beside the PET group, along with Tomczyk. He recalls, “When I ed the company, Commodore had a very small US operation that was focused on selling Commodore PET computers into computer stores for use in business.” Tomczyk was on a mission to improve marketing at Commodore. “He brought a marketing discipline, and particularly a product management discipline, that I had never been exposed to and was really alien to the company culture,” says Harris. Harris’ first task was to improve the friendly computer guide. “In my first week, Mike Tomczyk handed me the manuscript for the VIC-20 manual and said, ‘Look we hired an outside company to write the manual and they wrote this manual and it’s crap. It’s cute, but it just simply will not teach you how to actually work this machine. I need you to rewrite it.’” Harris replied, “Okay, where’s my computer?” Tomczyk told him, “We can’t get you a computer yet and you won’t have a computer for two weeks.” He gave Harris a set of style guidelines and set him to

work. It was an unpromising start for the new hire, but his knowledge of PET BASIC would allow him to make a good beginning before an actual VIC-20 arrived. “So I brought in my trusty Smith-Corona typewriter that I had used in college and I started banging out the manuscript,” he says. “I was trying to write something that would be friendly but would actually cover the ground you needed to get your feet on the ground and program.” Soon after, Tomczyk hired Andy Finkel, a close friend of Harris, who also attended Cornell University. “About a month into my tenure there, Mike asked me if I knew any programmers who could come in and help us,” recalls Harris. “And I thought of my friend Andy Finkel, who I had been friends with for many years going back to high school days.” Harris boldly told Tomczyk, “I’m going to bring this guy in tomorrow and you’re going to hire him right away. He’s fantastic and he’s just the guy we need.” Finkel had just finished his computer engineering degree at Cornell, and he was a Commodore PET owner and hacker. The two programmers were also diehard Star Trek fans, and worked at the first Star Trek convention held in Philadelphia. “I mostly ended up working security and Neil did PR and other things,” recalls Andy Finkel, who claims he did not wear a red security shirt or carry a phaser weapon. It was up to Finkel to keep hordes of overexcited fans away from the likes of William Shatner and Leonard Nimoy when they attended the convention. Without any office space, the programmers crammed into the service manager’s office. “Mike didn’t have a lot of resources or budget to work with and was working in a department that was not really appropriate for where you would have made a group like this report to,” recalls Harris. “So we basically had to beg, borrow or steal equipment to get anything done.” Soon, Harris and Finkel got their first look at the new VIC-20 computer. “They were always in my possession in the early days,” says Robert Russell. “When there got to be a few more we gave some to Tomczyk and Andy [Finkel] and guys like that.” The team continued working with Bob Albrecht and the ex-PCC writers at Menlo Park. “The manuscripts travelled back and forth with several iterations

and phone calls discussing various changes and various arguments about what should go where,” recalls Finkel. “It was fun trying to get a manual that was consumer oriented rather than technical oriented at that time.” Early parts of the book dealt with setting up the computer, but much of the material was programming instruction. “We were just looking for interesting and amusing ways to get people into programming,” says Harris. “I wrote a program to do various sound effects like a wolf whistle and Mike liked that, so he said, ‘Put that in the book! Show people how to write stuff like that.’” It was exciting for Finkel to work with the pioneering PCC writers. “The guys were fun people to work with. They had lots of ideas,” he says. “They made pictures and little cartoons. It was an attempt to make something correct but not dry.” As work on the manual progressed, the team began talking about creating an in-depth programming guide. The idea came about because Harris was familiar with the lack of information for his PET. “Early on in my preCommodore days, the documentation was extremely poor,” says Harris. “You really had to be tied into a network of people who were sharing information in order to really get the most out of the computer.” Once inside Commodore, Harris had discovered why documentation was so sparse for the PET. “Culturally in the company, management were not personally technical. They were more business oriented. They were much more of a closed system mentality. They wanted to keep all the knowledge inside the company so that we were the only ones who could write software and make peripherals.” From his previous experience with the PET, Harris knew the best products often originated outside Commodore. “No one company could provide all the good stuff,” he says. “As computer s, as well as technical people, we felt that was the way to build a thriving ecosystem around your computer.” Harris and the team eventually decided to make a complete programmer’s reference guide. “I was adamant that where I was going to try to make an impact was to make sure that, by God, we were going to provide as complete documentation on this system as humanly possible,” he says. “Andy felt the same way—that we should get as much information as we could get into people’s hands.”

In his zeal for open access to information, Harris even included proprietary information in the manual. “We put everything in it,” he says. “We had a foldout schematic for the motherboard, which nobody did. We made sure every call to the operating system and the entire memory map and all the hardware was fully documented.” Harris knew they would need to keep their open-strategy secret from Commodore’s established management. “We figured the other guys wouldn’t notice until it was too late. By the time everything was out there, the cat was out of the bag.” Technical information for the reference manual came directly from the VIC-20 engineers. “We did all the documentation that ended up in the programmer’s reference manual,” says Russell. “Andy Finkel and those guys pretty much took all our charts, diagrams, memory maps, and all my folders and notebooks, and that’s what became the reference guide.” The writers inadvertently created a special variation of the word kernel because of a misspelling in Russell’s documentation. “The Commodore version of it was because I misspelled it on my big banner heading on the top page,” says Russell. “So that’s how it became known as the ‘kernal.’” The authors divided the manual into sections for each team member to work on. “Several weeks later, when we started working on the Programmer’s Reference Guide project, we split up the work based on people’s backgrounds,” says Harris. “Andy has an engineering degree and was a strong assembly language programmer, so he wrote that section. I wrote the section on BASIC and that level of programming detail. Mike Tomczyk wrote pieces of it that were more of the consumer friendly oriented parts of the book.” A fourth writer, Paul Higginbottom, travelled from Commodore Canada. On the telephone, he sounded like a mature programmer, and the team was surprised when a teenager showed up. “Paul Higginbottom, who was a really young guy in Toronto, worked on a section of it,” recalls Harris. Although the writers had Russell’s notes, they often needed his help personally. “We’d go over and bother the engineers for details that they hadn’t written down or hadn’t even thought of,” says Finkel. “Then we’d sit down and try to write some programs and then write about what the programs were.”

While this was going on, the authors sometimes butted heads with the PET division. “The guys in the next section over from us in the office space were the software division for the Commodore PET business computers,” recalls Harris. The head of the PET software division, Dieter Ammon, often teased the VIC-20 team. “They used to laugh and say that our computer, the VIC-20, would be given away for free with every ‘real’ computer that would be sold.” In 1981, North American sales of the PET fell to 10 percent, with 16 percent for Tandy and Apple rising to 23 percent. Despite the low numbers, the PET group was still favored over the nascent VIC-20 group. “Those were the guys who politically were really in control of that group and so all the money was going there,” says Harris. “We would come in and requisition a set of equipment and it would be turned down.” With the release deadline approaching, the authors refused to let others get in their way. “We were absolutely ionate about what we were doing and we were working all kinds of hours,” says Harris. “We would typically stay late and be the last people in the building pretty much every night. So we just wandered over to the next department and took whatever we needed off people’s desks.” The manuscript had a firm schedule to go to the printer. As the date neared, the team began sacrificing certain quality assurances. “We had a hard deadline,” says Finkel. “We wrote the programs, tested them, documented them and kept going but you could tell near the end of the book where we actually had to stop writing out and testing the examples completely because we got kind of big. So up to a certain point in the book you can tell we had written and tested each of those things we were talking about and then at some point it becomes sort of nebulous. ‘Yeah, we think it will work like this.’” The programmers were under tremendous pressure to deliver two large manuals quickly, but they succeeded. The first book, Personal Computing on the VIC-20: a friendly computer guide came in at 164 pages, while the VIC-20 Programmer’s Reference Guide was 289 pages. Amusingly, Tomczyk uses the term “computerists” on the cover of the reference guide to refer to s. Russell credits Andy Finkel for much of the success. “[Tomczyk] had people like Andy who were clever, smart people who saved his ass by actually making sure that some things got done that he was trying to do.”

* * * After CES Las Vegas, Chuck Peddle and Jack Tramiel were ready to go to court. Peddle and his estranged wife Shirley had gotten back together in the meantime. Tramiel was determined to win the case for several reasons. First, anything that undermined Peddle would undermine Sirius Computers, which was a potential rival to the PET in the UK. Second, winning the lawsuit meant he would have approximately three million dollars more of Commodore stock, which could come in handy for expanding production. Commodore lawyers began using every legal tactic in their arsenal against Peddle. “The day we were getting ready to go into [the courts] with the jury trial, they changed the whole case,” says Peddle. All the time and money Peddle spent preparing for the trial was now for nothing. “After they bled us as much as they could bleed us, they burned us by just dropping the suit.” In place of the jury trial against Sirius, there was now only the single case against Peddle for all of his Commodore stock. A single judge would determine the outcome of Peddle’s case. “The minute they changed the case, I should have just gone back and said, ‘Okay, how do we settle now?’ They were in a mood to settle,” says Peddle. Commodore’s lawyers convinced the judge that Peddle was holding onto stock he had not earned. “Their representation was they gave me the shares but I agreed to give them back, even though there was no document to back that up,” says Peddle. The trial went poorly for Peddle. “We were just totally out-lawyered,” he recalls. “They jerked us around ten different ways and my lawyer was too naïve. We were so sure we had won. We lost.” Another judge later turned down Peddle’s appeal. “I lost everything. I had effectively two or three million dollars worth of stock. Everybody else in the company had cashed out two or three million dollars worth of stock and Jack cheated me out of everything I ever made,” says Peddle. “It was everything I own.” In the wake of the judgment, Peddle was unprepared for the devastation it would personally cause to his family. Both his ex-wife Shirley and his daughter fell into

a severe depression. “My wife was with me every step of the way when we built the computer business,” he explains. “Jack Tramiel had made promises to her just like he made promises to everybody. She knew exactly the sacrifices we made as a family and everything else to make that happen.” “For him to go into court and be able to reach in and deprive us of every penny we owned after they knew how much we had done and sacrificed, they just basically said that the world is totally unfair.” When it became clear Peddle had lost the rewards from their years of dedicated work at Commodore, Shirley left for good. “We tried to get back together, but it didn’t really work out,” he says. Today, with the tragedy far behind him, Peddle retains his sense of humor about the decision. “Do you know what the definition of ten thousand lawyers buried in chains in the bottom of the ocean is? A good beginning.” * * * In early 1981, the rest of Commodore International launched the VIC computer in their respective territories. Most countries stuck with the VIC name. “We launched it in Europe as the VIC, which it was also launched as in the USA and Canada,” says Spencer. However, in , the VIC name did not receive a warm reception. “It was the shorthand for fuck,” says Russell, referring to the German word fick. According to Russell, when engineers came to for trade shows, Harald Speyer warned them, “Don’t call it the VIC-20!” “Because VIC was not a good word in German, it was always the VC-20 in ,” explains Spencer. VC stood for Volkscomputer, or “the people’s computer.” The name was significant for Germans, since it sounded similar to the much-loved Volkswagen. Despite the initial naming concerns, the VC-20 launched with solid orders under GM Harald Speyer’s leadership. “Commodore was a big market in , and very successful,” says Spencer. For now, the Commodore Japan factory produced VIC computers for the rest of

the world, but it soon became clear that demand for the VIC-20 would overtake its production. Tramiel planned to open new manufacturing plants in and the UK, along with the US factory, provided he could scrounge enough funds to build them. Tramiel would need to be creative in order to raise funds because Gould was still not willing offer new shares while the stock price continued rising. However, Gould had no qualms about selling existing shares owned by the company. The successful verdict against Peddle gave Commodore approximately three million dollars to put towards new production. This gave Tramiel the idea of obtaining funding through litigation. In 1981, Tramiel was still attempting to recover some of his investment from the disastrous Micro Display Systems acquisition from Seiko. He did not want the case to go through court because he needed the funds immediately for production, so Commodore’s lawyers let Seiko know they could avoid a long and costly court battle by settling out of court. Surprisingly, Seiko agreed to settle and Tramiel succeeded in regaining half of the 100,000 shares. Along with regular profits from the sale of PET and VIC-20 computers, Tramiel now had the funds to expand VIC-20 production. During a meeting of general managers in 1981, he asked for the creation of the new production factories. Tramiel trusted Speyer due to his past success with the PET, but he joked that if he did not see VC-20s rolling out of a German factory in one month, he would fire him. Speyer replied, “You can’t fire me. Slaves can’t be fired!” British s affectionately called the VIC-20 the breadbox computer[1] because of the unusual shape of the case. The VIC-20 faced stiff competition from Sinclair, but the lack of color and sound in the ZX-80 made the VIC-20 a clear winner. Commodore UK soon began using the factory at 1 Hunters Road in Corby, Northamptonshire to assemble VIC-20 computers locally. Before 1980, the microcomputer industry had sold less than a million computers worldwide. By 1981, Apple was beginning to take the lead. With computers gaining popularity, it became a race to see who would be the first to sell a million. * * * With the manuals completed in March 1981, the VIC-20 group transformed into

a software house. Charles Winterble believes it was a mistake to have Tomczyk controlling a software development team. “It’s being run by a marketing guy who’s really not even a marketing guy,” he says. “He was certainly not a software manager.” When Ken Hollandsworth saw the success of the manual, he gave Tomczyk more resources for software development, including a new workspace at 681 Moore Road in King of Prussia, Pennsylvania, a ten-minute drive from MOS Technology. The group moved into an open bay with walls and shelves around their work area. In the new location, each person had a four-foot wide desk. “We were all working in this large cubicle, so it was a group effort,” says Finkel. The VIC-20 would require a software library before the product launched at the CES. They would release the software on cassette tapes or cartridges. To load and save programs, Commodore would sell a separate cassette player and recorder, called the 1530 Datasette (an amalgamation of Data and Cassette). Initial software development focused on cassette programs. “In the beginning, our lead-time on cartridges was going to be four to six months,” recalls Finkel. “So we had to write something that would fit entirely in the 3.5 K of usable memory and that meant cassette programs written in BASIC, machine language —whatever would fit.” The programmers held planning sessions before embarking on individual projects. “We had these sessions that we focused on two things: what would people want and what could we actually wedge into the VIC-20,” explains Finkel. “That was interesting and fun because we could have these great ideas but we didn’t have the RAM to actually do it.” Finkel wanted to program a slot car racing game, but quickly found the limited VIC-20 memory prevented him from realizing his idea. Instead, most of their game ideas came from slightly more primitive 1970s arcade machines, which used very little memory for the time. The team had developed their programming skills on the VIC-20 while writing the manuals. Now they began producing reasonable software titles. Neil Harris developed two tape-based games called VIC-21 (Blackjack) and Super Slither. Another programmer named Duane Later developed Space Math. Andy Finkel stood apart from the group due to his experience, going back to the

first KIM-1. “I had a lot of programming and machine language experience, even on large IBM 370s,” he says. “That was part of the stuff I did in college.” Those machines had ample memory. “The VIC-20 was new and different in of trying to stretch its capabilities.” He would later become Commodore’s programming guru, but for now, he was building his specific VIC-20 programming skills. He developed Loan-Mortgage Calculator, Car Chase, and VIC-20 Music Composer. Music Composer allowed Finkel the opportunity to pioneer some programming techniques. “I was the first one to poke at the sound chip and actually try and get it to do interesting stuff,” he recalls. “I spent a lot of time on that.” Music Composer became a developer tool when the programmers wanted to add music to their games. Car Chase was a port of the 1979 Irem game Head On. The game predated PacMan but shared many similarities. Finkel’s port used a combination of BASIC and assembly language. “The car moving routines and the input routines were in assembler and everything else was in BASIC,” he says. Many of the games were simply ports from the Commodore PET. “We took a lot of old games that were on the Commodore PET,” says Tomczyk. “We just updated them and improved the graphics and made them VIC-20 games.” With few exceptions, the programmers developed their software in BASIC, with graphics using simple PETSCII characters. They looked like PET games with color and sound. Surprisingly, none of the games used a joystick. The titles did not approach the quality of games found in arcades. “When they started the software group they didn’t do very good work,” says Winterble. “The fact of the matter is most of them, except for Andy, had no knowledge about how to do software architecture and how to put together a development plan from beginning to end. They thought they could just sit down there and start coding. What they developed was product you would expect without a knowledge of how to truly develop software at the time.” Neil Harris felt that many games were of marketable quality, but the applications suffered. “Personally, I think that Commodore was pretty much terrible at applications software,” he says.

The varied quality of the games and applications would have resulted in some tapes selling well and others barely at all. To balance the quality of the releases and improve the overall value, Tomczyk decided to release the games and applications as six-packs. To produce the tapes in quantity, Tomczyk went to a local audio tape company. “We went to Cory Sound, an audio tape manufacturer and recorder, and taught them how to make computer cassettes,” recalls Finkel. “It took a while to convince him that we wanted the audio levels saturated and not balanced. That was something that produces a really horrible screeching sound when you listen to it, but it helped the cassettes read it better. They were our supplier throughout the whole cassette tape manufacturing period.” Almost immediately after sending the duplication cassettes to Cory Sound, the cassette tapes were ready for store shelves. Tomczyk hoped the games and applications would sate demand for software until Commodore could release more advanced cartridge games. “We had about six months when all we had to the computer were the tape six-packs that I created,” says Tomczyk. * * * As 1981 progressed, it became clear the VIC-20 was enjoying success in Japan, , and especially Britain. “Harald and I had been very successful in launching the VIC in Europe,” says Spencer. The dramatic success in Britain was largely the result of 35-year-old marketing director Kit Spencer. “The VIC-20 was a pretty successful thing in the UK,” says Andy Finkel. However, in the United States, the VIC-20 was underperforming compared to overseas markets, as usual. “It wasn’t very successful in those early days in the States,” says Spencer. “In the States, Mike was pretty strongly involved, but in fairness he was perhaps limited in what he could do.” Ken Hollandsworth had been guiding the VIC-20 launch in theory, but in truth, Tomczyk did more to launch he computer than anyone else in the US up to that point. “When the VIC-20 started gaining momentum, Jack wanted to bring somebody in more senior and with longer tenure in the organization and with more experience in mass-market advertising to take it to the next stage,” explains Harris. “That probably wouldn’t have made Mike very happy.”

“Mike had some very good points, but he was probably better with an experienced manager working with him to help guide his talents,” says Spencer. “Ours was a young business and there were not that many experienced managers around. You really needed some type of consumer electronics or other marketing experience behind you and Michael had only a little of that before ing Commodore, but he did learn fast.” Although Tomczyk wanted to be the VIC-20 product manager, he was glad to work with the more experienced Spencer. “He was one of my favorite people in Commodore and during the Commodore years I would say he was my best friend in the company,” says Tomczyk. “I loved Kit Spencer. He was just a sterling character and a wonderful human being and super accomplished.” Tomczyk had modeled his VIC-20 strategy on what he saw Spencer doing in Europe. However, a successful marketing manager would require a specific strategy tailored for the US market. Due to his success in the UK, Tramiel had previously tasked Spencer to lead the European operations. “I’d been working out of Switzerland for about six months, because that had become the European head office and I was handling all of the marketing operations in Switzerland,” says Spencer. “I had been waiting until the end of the school year to move my family, and also waiting for an official work permit and things. I had a house I’d rented in Basel. I had a school for my daughter. I was one week off moving our belongings when Jack came over.” In April, Commodore held the quarterly general manager meeting in Basel, Switzerland. Spencer noticed Tramiel treating him strangely. “Funny enough he gave me a hard time for awhile one night,” recalls Spencer. “You’re used to that with Jack, but it didn’t seem to add up quite right.” Spencer soon learned the subtle ways Tramiel could manipulate his employees. “Then one of the other old managers, probably Harald Speyer, said to me, ‘Kit, you know why he’s giving you a hard time, don’t you?’ I said, ‘No.’ He said, ‘He wants you to go to the States but he wants to make you feel uncomfortable before he asks you.’” Tramiel was simply escalating the pressure on the reluctant Spencer. “That is typical Jack,” laughs Spencer. “After he’d fire people, I always used to go over to the States and help out for a while. But the joke was, we work great thousands

of miles away.” Spencer was dubious about Speyer’s explanation until Tramiel approached him. “The next day Jack said to me, ‘Kit, I’m under a lot of pressure from the board because we are not doing well in the States. You probably don’t want to come here, but I personally need you to come as our USA marketing is in a mess and you are the best person to help sort it out. I know it’s not where you want to be tax wise. If you would like to be somewhere else later, I understand.’” Spencer replied with his characteristic wit. “You’re asking me to take on a smaller job, Jack!” In truth, the USA was a larger market. “Obviously the potential in the USA was much greater as events of the next two years would prove,” says Spencer. Spencer had turned Tramiel down countless times, but this time he decided to give the offer serious consideration. “When you’ve worked for a guy for a while, you think about it,” he says. “So I said to Jack, ‘Let me just go and talk to the family, and I’ll let you know tomorrow.’” In truth, Spencer was wary of occupying a marketing job with Commodore US. “I didn’t think going to the States would be very smart because he had a habit of firing marketing guys over there quite quickly,” he explains. “There had been six marketing directors in about two years in the USA. Half had been fired and half quit.”’

Bob Gleadow (left) and Kit Spencer attending a trade show.

Neil Harris thinks Tramiel had no choice but to corner Spencer. “We churned through marketing and sales people like nobody’s business,” he says. “As the years went by, it was really difficult to hire good marketing people and management of the company because they would look at it and say, ‘You know what, why do I want to work for this kind of organization? It’s not a marketing oriented organization, I can’t have an impact on the product, and it’s not culturally a good fit for marketing people.’” That evening, Spencer went over the offer with his family, who had packed their belongings for the move to Basel. “I talked to my wife,” he says. “We sort of said, well, I’d worked for Jack for years and it’s what the company really wanted and I could see the sense of it.” Spencer’s family decided to go on a new adventure to the United States. “I went back the next day and said to Jack, ‘Okay Jack, but there are a couple of conditions.’ I didn’t discuss salary with Jack. I didn’t adjust contract or share options. I could have done so, because he probably would have given me whatever I asked for, because he really wanted me to come. But it wasn’t that sort of relationship if you understood Jack. I said, ‘The two conditions: One, I want to bring the family from day one, because I’ve just spent six months hardly at home. I don’t want to do that again. The other condition is I’ve planned to take a vacation soon. I know the day I arrive in the States and the shit hits the fan, I’m not going to be able to take a vacation for a long time because it’s going to be very hectic. I want to take a vacation before I come.’” Tramiel needed no time to consider the offer. “He said, ‘Okay, where are you thinking of going?’ I said, ‘Exeter.’ He said, ‘Can you go in the next week? It will be on the company. I’ll see you in two weeks in Valley Forge. There will be an apartment waiting for you. Bring the family.’” Spencer trusted Tramiel, and knew he would honor every part of their deal. “A lot of people would have asked for a contract,” he says. “I moved to the States on a handshake deal.” Tramiel honored the deal. “Sure enough, I arrived there two weeks later and he

had an apartment,” says Spencer. “I started work.” With Spencer’s announcement, the press took notice. “I when I moved to the States the Wall Street Journal rang me up and asked, ‘How long do you expect to last? There have been six people in this job in the last two years. Three have been fired and three have quit.’ I said, ‘I’ve already worked 10 years for Jack.’ They said, ‘Nobody’s worked that long!’ I said, ‘Yes they have if you come to Europe. I have and Harald Speyer has for about eight years.’” Without the distance from Tramiel, Spencer himself wondered how long he would last. “It did help that we were a long distance away and we were given a reasonable amount of autonomy,” he adds. At the four-day National Computer Conference held in Chicago on May 4 1981, Kit Spencer and Harald Speyer of Commodore attended. Each brought along literature and their unique cardboard box designs for the VIC-20. Tramiel was impressed. Spencer rapidly diagnosed the problem with US sales. “In the USA we had no decent packaging, limited software and limited distribution,” he says. “When I arrived we basically re-launched the VIC completely.” Surprisingly, Spencer did not immediately pursue advertising for the VIC-20. “I didn’t start a massive advertising campaign when I arrived. The first thing I did was actually cut back on advertising. It wasn’t worth spending a lot because distribution was not geared up enough to make the spend really worthwhile.” Because the VIC-20 had just launched in Europe, he was able to reuse much of the European launch plan in the US. “I said, ‘Let’s take what’s good already and don’t reinvent the wheel.’ I picked up the things we did in Europe, even bringing in some European software and packaging design.” Tomczyk recalls, “We took elements of the German and English designs and made that the American packaging.” Spencer combined the eye catching silvergray box from Commodore with the rainbow UK packaging. By the end of May, a number of VIC-20 computers had been sold by Commodore dealers. As a result, customers were frequently calling with questions. In response to this, Spencer wanted to create a helpdesk for Commodore. As Spencer had done in the UK, he attempted to make it a profit

center. “When Jack said, ‘Why are you hiring somebody?’, which is what he would do if he saw the headcount going up, it made many people scared to add software and training people. I would respond to Jack saying, ‘Because he’s making money for us Jack. This is what he’s doing’, and he’d let me get on with it.” In the US, Spencer saw that executives were more micromanaged by Tramiel. “The people in the USA at that time said, ‘Jack won’t allow us to do that.’ What that really meant was Jack wouldn’t allow them to hire other people if they were just more overhead,” says Spencer. “I’d start with one guy and say, ‘Look, I’m covering your salary for the first couple of months but then you’ve got to start making money. If you make enough money then you can hire some more people and we can do more.’ Which is a little different approach and it worked very well.” Spencer had Tomczyk hire six college students for the Commodore Hotline, the first official technical provided by Commodore. [1] In , they also called VC-20 the breadbox, or Brotkasten.

CHAPTER 23

Mass Distribution 1981

Commodore’s official magazine had its origins as the PET Club Newsletter, created by Commodore’s then VP of marketing, Dennis Barnhart, whom had since left Commodore.[1] “Dennis started it when he was at Commodore,” says Chuck Peddle. “It was just another marketing thing.” In 1981, the newsletter transformed into a magazine called Interface. Interface was too close to another early computer magazine called Interface Age. In mid-1981, Commodore changed the name to Commodore the Microcomputer Magazine. The magazine focused on productivity software and the PET computer, but also included updates on the VIC-20. Michael Tomczyk and his VIC Commandos contributed articles to the magazine. “Those were started and run by our PR department back in 1981,” says Neil Harris. “They’d come to me and need an article from time to time, so I’d write an article for them.” In May 1981, Tomczyk’s software team began work on their cartridge games, which they hoped to display in prototype form at the June CES show. Commodore had lots of experience manufacturing cartridge ROMs for Atari. Now they would develop their own games. The VIC Commandos, as Tomczyk called his developers, translated two PET games Spencer brought from the UK, Cosmic Jailbreak and Draw Poker, and enhanced them with better sound, color, and graphics. The engrossing game had a twisted sense of humor. “We were getting some decent software from the UK at that time,” says Andy Finkel. “That was actually entirely in character graphics

on the PET.” In the game, some determined aliens are trying to break their fellow aliens out of a prison by taking it apart one brick at a time. The player has to shoot them before the prisoners escape. “It was a very addictive game,” says Finkel. “We had one of those running on a PET in the VIC-20 area at all times during the early part of the VIC-20 days. It was a great game so we had to have that ported to the VIC-20. We got some other games from Commodore UK as well.” In the ensuing weeks, Tomczyk’s group grew even larger. “Within that group there were about six young guys,” recalls Charles Winterble. Tomczyk hired at least four new people, including programmers Bill Hindorff, Eric Cotton, David Street and Lee Ancier.

Michael Tomczyk, his wife and his “VIC Commandos” playing board games after hours (Andy Finkel is far right).

The programmers developed a fun atmosphere, away from management interference. “They were over in this little building separate from the engineering organizations and marketing organizations,” explains Winterble. After work, the group attended science fiction conventions and played board games. They also took long lunch breaks at the local arcade, teasing others that it was all for research. Others in the company soon noticed the group’s antics were out of proportion to the work they delivered. “They were kind of tucked away, living this life of Riley,” explains Winterble. “The idea was we should treat these guys like superstars. They really don’t have to come to work if they don’t want to. Things like that.” The VIC-20 had a head start in Japan, so it had the largest games library by May 1981. According to Harris, “Tony Tokai would come over like your favorite uncle with a bag full of gifts and say, ‘Here’s Pac-Man. Here’s Rally-X. Here’s all these games.’ We’d look at him and say, ‘But we don’t have the rights to these!’” The quality of the games from HAL Laboratory impressed Harris. “They were doing amazing, amazing work,” he says. “They didn’t get the original source code to the games obviously. They just got an arcade machine and they said, ‘Okay, copy this.’ And that’s what they did.” The quality of the games inspired Finkel, who vowed to match their quality with the next game he delivered. “HAL Labs turned out to be some very young programmers but they were very good,” says Finkel. “They could do very good copies of arcade games.” When Tomczyk saw them, he could not resist adding the dubiously legal games to his June CES software lineup, even though Commodore did not own the North American rights. Tokai also brought some original games from

Commodore Japan, such as Mole Attack and Super Alien. With the software library slowly developing, there were only a few more things to contend with before the June CES show. The original Japanese design of the VIC-20 had one major problem: the heat sink was in with the cartridge port, which made it so hot people sometimes burned their fingers inserting cartridges. Engineers solved the problem by moving the heat sink slightly. Another major problem occurred when the FCC discovered the VIC-20 emitted radio waves, sometimes enough to disrupt a neighbor’s television or radio reception. They required the VIC-20 to use metal shielding to block the radio waves. The FCC requirements surprised Russell because the heavy metal case of the PET blocked these emissions. “Nobody had to meet the FCC requirements before, which was a gigantic roadblock all of a sudden near the end of the VIC,” says Russell. The engineers modified the VIC-20 design by adding metal shielding around the video circuitry. The solution added to the cost to manufacture the computer, but Russell says there was no option. “We’ll hire or spend whatever money we have to otherwise we won’t be able to sell any computers.” * * * By June 1981, the VIC-20 was ready for launch as a retail product. Commodore hoped to take large orders for the computer from retailers at the Consumer Electronics Show in Chicago. It was the first time Commodore appeared at CES without a lineup of new calculators. The Commodore booth was the best yet, with a large central pillar and five VIC-20 computers arrayed around it. With all the game cartridges available for demonstration, the product had the potential to generate massive sales. Robert Russell was eager to unveil the computer he helped develop. “I sat by that baby 24/7 when the VIC-20 came out,” he says. “I went out with [Michael] Tomczyk to buy the big screen monitor and told the little old ladies in the booth all the bullshit stories to tell the press.” A year earlier, Commodore had released the CBM 8032 computer system for $1500—or $5000, including printer and disk drives. Now Commodore was offering a color and sound computer for only $299. The price left the

competition stunned. The biggest draw at CES was the Atari 2600. According to Russell, it was easy to convince people to buy a VIC-20 instead of the Atari game console. “We were pushing the whole computer angle, which they couldn’t.” When Russell told retailers the VIC-20 could play games and run applications software, they asked him, “What do I need a computer for? Why would I buy it? Why would I spend $300 for it?” They were good questions, since games were the VIC-20’s strength. Andy Finkel also attended the show, displaying the manuals and Programmer’s Reference Guide. “We were there trying to figure out how to launch the VIC-20,” he says. He sums up the response. “Everyone liked the VIC-20’s keyboard, they were shocked at how little memory it had, and surprised at the display. We got compared to Apple a lot.” The cartridge lineup for the VIC-20 was impressive. It included VIC Invaders (a Space Invaders clone), Star Battle (a clone of Galaxian), Super Alien, Pakkuman (known as Pac-Man in North America), Jupiter Lander (a Lunar Lander clone), Night Driver, Mole Attack, Super Slot (slot machines), Rally-X (temporarily renamed Car Race), and The Sky is Falling. The software library was large enough to satisfy the retail marketplace for months. The VIC developers had mixed feelings about the lineup. “The games from Japan were perfect copies of arcade games, and completely unsellable in this country, legally,” laughs Finkel. Pac-Man received the most attention, since Pac-Mania was sweeping North America. “Tony Tokai brought it over and said, ‘Look at this!’ We all looked at it and we were amazed because it was exactly Pac-Man. It was as good as any arcade,” recalls Tomczyk. No one was quite sure of the legality of showing Pakkuman without actually owning rights to it. “At the beginning, there was some question as to what was patentable and what was copyrighted,” recalls Finkel. Commodore decided to use Pakkuman to promote the VIC-20 at CES surreptitiously. The HAL programmed Pakkuman was identical to the official Namco Pac-Man. It had the same music, the same sounds, the same maze, the same yellow Pac-

Man, and the same ghosts that turned blue when Pac-Man ate a power pill. “On the VIC-20 I liked playing Pac-Man. That was done very well,” says Yannes. It was probably the best home-version of Pac-Man for any computer or console in the world. It was also illegal in North America. Atari had already purchased the North American rights from Namco, although they would not begin development on their 2600 version until later in the year. When Atari saw the VIC-20 games on display, they had two reactions. The first was respect for the programmers who produced such excellent copies of the originals. The second was outrage at the obvious copyright infringement. There were other unauthorized games aside from Pac-Man. VIC Invaders was a clone of Space Invaders by Taito. Night Driver was a clone of the game by Atari. Car Race was a clone of Rally-X by Bally-Midway. Star Battle was a clone of Galaxian by Bally-Midway. Jupiter Lander was a clone of Lunar Lander by Atari. Shortly after the Commodore booth opened, Atari lawyers informed Tomczyk and Tramiel of the infringements. Tomczyk felt the blood drain from his face as he realized the VIC-20 launch titles were in jeopardy. However, Atari was surprisingly conciliatory towards Commodore, perhaps owing to their longstanding relationship purchasing 6502 processors and ROM chips from MOS Technology. Atari’s lawyers explained to Tomczyk what they could and could not do. Most of the games would be fine, so long as Tomczyk modified their names so they did not infringe on Atari’s trademarks. However, the most valuable game in the launch titles, Pac-Man, was illegal and Atari said they would sue if Commodore attempted to release the game. Tomczyk thought there was now little hope to have the arcade-perfect game on the VIC-20. After the Atari lawyers departed, he turned to his boss. “I said, ‘Jack, what are we going to do?’ Jack looked at me and he said, ‘I’m going to sell it anyway.’ I said, ‘What do you mean?’ He said, ‘I’m going to sell it anyway. We’ll rename it and we’re going to sell it in England.’” It was a bold plan, considering minutes earlier, Atari lawyers stated clearly that they would sue Commodore if they attempted to release it. “I said, ‘So what are you going to do when Atari comes after us?’ He said, ‘We’ll set aside a reserve

to pay the royalties on the game. When Atari comes after us, we’ll tell them, ‘We’re going to pay you the royalties on the games we sold and then we’ll just stop selling it.’” Tomczyk had never before heard a CEO actually plan to be sued as part of a business opportunity, but it made sense. Pac-mania had the potential to raise interest in game machines, and Tramiel was not about to allow the VIC-20 to be left out. “I said, ‘Oh yeah?’ He said, ‘We’ll probably sell a million of those before any of that happens.’ I said, ‘That’s pretty ballsy.’ He said, ‘Yeah, well you know, business is war.’ I was really impressed by that.” Although the VIC-20 games impressed attendees, Commodore did not make much progress getting the VIC-20 onto retail store shelves. None of the major players—Kmart, Sears, Wal-Mart, Toys’R’Us, Target, or J.C. Penny—took orders for the computer. It was apparent Commodore would need to make a concerted effort to gain retail acceptance for the VIC-20. * * * The VIC-20 software lineup for the US did not have a good start with the lackluster titles on the six-pack tapes. With the help of some titles from Commodore Japan, and Spencer’s redefinition of the VIC-20 as a game machine, Commodore would attempt to produce quality games that might fare well in the marketplace. Their most urgent task was to get the Japanese games on store shelves—legally. Commodore was reluctant to deal directly with the companies whose games HAL Laboratory had cloned. “We decided not to deal with that,” explains Finkel. “Instead, we took the basic game and, with the help and advice of the Commodore legal team, modified, changed and reworked it so that we could sell them in this country.” The software group was able to save most of the Atari games simply by changing the names. The name VIC Invaders infringed on the name Space Invaders. This was easy to remedy because other companies had emulated Space Invaders and Taito had not properly defended the trademark, so everything but the name became public domain. Commodore could retain the same game play but changed the name to VIC Avenger. Midnight Drive was too close to the original Night Driver name and Atari

objected, so Commodore changed the name to Road Race. The game play remained the same because there were many similar 3D racing games. Jupiter Lander was fine, because the 1979 Atari Lunar Lander was a copy of the Lander game going back to shared mini-computer systems. Even the PET had a version of Lander earlier than Atari. Following CES, Kit Spencer renamed the Pac-Man clone as Jelly Monsters for release in the UK. However, Tramiel was opposed to releasing it in the United States where it would come under scrutiny by Atari. “It was never considered,” says Tomczyk. Instead, Tomczyk had his VIC commandos begin programming a Pac-Man clone for US release called Cosmic Cruncher, which featured a Commodore chickenhead logo running through a maze with satellites. Rally-X was a unique game, and Commodore’s shameless rip-off was almost identical in look and feel to the arcade game. Even the text instructions were identical. Commodore’s lawyers advised the programmers to modify the look of the game dramatically. Neil Harris recalls, “I very clearly, Andy Finkel took the graphics and turned it into Radar Rat Race where you had mice and cheese instead of racing cars and oil slicks and things.” Once the programmers finished converting the games, Tomczyk released the modified code to MOS Technology where they would replicate the code thousands of times in 8-kilobyte ROM chips for cartridges. “You’d finish something and it would take many months before there were actually cartridges in stores,” recalls Harris. * * * The long production delay troubled one of the young programmers. “One of the of the programming staff had become impatient with the pace at which our work was being released to the marketplace,” says Harris. One day in the summer of 1981, Tomczyk opened up a copy of Compute! magazine and spotted an ad near the back for a company called Blue Atlantic Software. “It had a series of VIC-20 games that really looked great,” says Harris.

“It had an address in Pennsylvania right up the road from where we were.” Tomczyk told Harris, “Fantastic. We need companies that can create games for us. Let’s call these guys and see if we can sign them up to do work for us.” Harris called the number and said, “Hi this is Neil Harris from Commodore.” The person on the other end of the line hung up immediately. “I was like, ‘That was really weird.’ And I tried it again and they hung up again.” Tomczyk drove to the address and rang the doorbell. A few moments later, one of his programmers opened the door. “This was a programmer who we clearly never let out to talk to customers,” laughs Harris. Tomczyk recalls, “He had just gotten himself a hamburger and he said, ‘Do you mind if I eat my hamburger and french-fries while we talk?’ I said, ‘Yeah, that’s fine.’ We sat on the steps outside his apartment and he ate a Burger King hamburger.” The Commodore employee was attempting to sell Commodore games. “As far as I know, he was selling exactly what he got from Japan,” says Finkel. “The one programmer thought it would be a shame and a crime not to release these wonderful games to the VIC-20 owners because they were heads and tails above what you could fit on a 3.5 K cassette. So he started a side business secretly selling these games. That obviously wasn’t legal, because I don’t think he even had an agreement with HAL Labs. He thought the software deserved to be out there.” After listening to his explanation, Tomczyk was in a difficult position. “I felt more sympathetic than angry,” he says. “I explained to him very patiently and nicely that he had two choices: he could either go to prison or he could leave Commodore and sign an agreement to cease and desist, and our lawyers would write all of that up. He chose option number two.” His coworkers never saw him again. “He was a foolish young guy who just didn’t know any better and was certainly no business guy,” says Harris. “It was a weird and unfortunate incident.” * * * Spencer had declined to renegotiate his salary when Tramiel desperately asked

him to come to the United States. “When he asked me to do a job, I told him what I needed to do the job,” says Spencer. “I didn’t take advantage of the situation personally.” When Tramiel realized Spencer could deliver, he decided to reward him anyway. “I’d been there about two months and Jack said, ‘Oh by the way, I just doubled your salary to the day you arrived here and here’s a load of more share options.’” While the programmers worked on their games, Tomczyk negotiated the first game licensing deals for Commodore. He licensed Sargon II Chess from Hayden publishing. Andy Finkel, who was rapidly becoming the software guru of the VIC Commandos, adapted the code to fit into a VIC-20 cartridge. “That one is from the Spracklens, Dan and Kathe, a husband and wife programming team,” he recalls. The Spracklens originally programmed their chess game for the TRS80, then Kathe’s brother Gary Shannon ported it to the Apple II. “So my challenge was taking a program that ran on a 16 K Apple II and making it run on a 3.5 K VIC-20. It was all machine language.” While the Apple II version resided in 16 kilobytes of memory, Finkel only had 8 kilobytes of ROM program code and 3.5 kilobytes of RAM memory to work in. “It was one of the first ROM cartridges so it had 8 K of programming code in ROM and then the rest was in RAM,” he explains. “They had 16 K of RAM to play with, and that’s huge.” Finkel had to make some sacrifices in his code, but the Spracklins were satisfied with the result. “I couldn’t use a lot of their tables and I had to generate small sections of the tables programmatically—lots of little tricks,” he recalls. “But I got it in there and it worked pretty well. Everyone was happy.” Tomczyk also licensed five Scott Adams text adventure games: Adventure Land, Pirate’s Cove, Mission Impossible, Voodoo Castle, and The Count. “I knew Scott and I asked him if we could license all of his games,” recalls Tomczyk. Andy Finkel examined Adams’ code and reprogrammed it for the VIC-20. “That was one of the parts of the job that was a lot of fun,” he recalls. “I got to go out to various developers like Scott Adams and work with them to convert the games.” The fun atmosphere of Scott Adams early 1980s game company made an impression on Finkel. “He’s a programmer type of guy, definitely a nerd, but a

lot of fun,” he recalls. “His wife was running the istrative side and he was running the technical side.” Commodore flew Finkel to Florida to work with Adams. “His office in Florida was in a geodesic dome,” he recalls. “All of the programmers had desks around the wall, and it was a big open space. The first day I got there, we unpacked everything, set everything up and then we went to his favorite place: Disneyworld. It was just to set the tone. Game companies should be fun.” Finkel’s main challenge was to trim the 24-kilobyte games to fit onto 16-kilobyte ROM chips. “He showed me a routine for the computer we were porting from, and then I’d write some small VIC-20 specific routines.” Finkel was becoming adept at shrinking the size of code, a valued skill due to memory constraints at the time. He also programmed the games to work with an add-on piece of hardware called the Type N Talk by Votrax, which gave the games speech capability. The cartridges were easy for Commodore to produce because they were already producing cartridge ROMs for Atari.[2] “Commodore was building almost all the VCS ROMs,” says Russell. “MOS Technology was the cheapest ROM producer at that point in time.” However, manufacturing times were long and it would be months after Tomczyk sent each game to production before they would actually appear. Each game required box art and a write-up describing the games. Tomczyk wrote the box text, while he contracted out the art to a local oil painter. “I stumbled onto a really talented artist locally in Philadelphia that I contracted to do these,” he recalls. “He would come in with a concept and I would just say, ‘Go do it.’” The artist had previously done the art for the six-packs. Tomczyk continued their relationship for all subsequent Commodore games and even hardware peripherals. “We were doing so much packaging at the time,” he recalls. “This artist was working constantly, almost 100% for us.” For many covers, the inspiration behind the artwork was obvious, such as an Imperial Star Destroyer from Star Wars (1977) on the cover of the games sixpack. “One time he came in with a Dracula cover that looked exactly like Frank Langella,” recalls Tomczyk, referring to the actor. “I said, ‘Are we going to get sued by Frank Langella?’ We decided probably not, so we kept it.”

Commodore paid the contractor between $500 and $1000 per cover. “He did these covers from a few days to two weeks and he was like our artistic cover machine. He was constantly cranking them out and I was constantly meeting with him,” says Tomczyk. “He was wonderful. He was another really important part of the team.” * * * In the early eighties, personal computers were starting to catch the attention of Fortune 500 executives. These companies relied on mainframes and minicomputers supplied by IBM, but they started to see the benefits of ondemand computing in applications like VisiCalc, which were impossible on shared systems. Jack Tramiel dismissed Chuck Peddle’s ideas for a business computer. After Peddle departed, an executive named Bill Robinson tried to convince Tramiel to attack the business market. Robinson came from the minicomputer industry, and as such, he knew microcomputers had the potential to displace minicomputers. However, Tramiel was still trying to win the mass-market for computers. He told Robinson he felt Commodore was not ready to attempt to break into the business market. Instead, he wanted to wait a year to get everything in place first. Robinson became frustrated and departed Commodore in December 1980. As it turned out, IBM got there first. In September 1981, IBM released the new IBM PC 5150. It was exactly what the industry was looking for at the time. It was modular and offered great expansion capabilities. As Byte magazine described, “It has a number of interesting features and a few flaws, but it is easily the best-designed microcomputer to date.” The IBM PC sold with a color graphics adapter (CGA) which produced 16 colors, including high-resolution graphics and a programmable character set like the VIC-20. The IBM PC contained Microsoft BASIC in a 40-kilobyte ROM chip, very similar to the PET and VIC-20. Like Commodore BASIC, it contained the commands NEW, LOAD, SAVE, LIST and RUN. Microsoft even included a variation of the PETSCII graphical characters. “I have

to it the IBM guys took it one step further and did an even better job,” says Seiler. “I’m sure they took our graphics when they did their IBM PC. They had double borders and things which really looked nice in the business world when you did spreadsheets and stuff.” Bill Gates made sure to specify that Microsoft would receive royalties for each copy of the operating system. “Bill Gates learnt from the Commodore BASIC deal, which is probably why he kept the rights to his operating system when he sold it to IBM,” explains Kit Spencer. “On that deal, he took royalties on every computer sold and also sold it to all the other IBM-like personal computer manufacturers as well. He certainly learnt well from Jack!” Since MS-DOS was not ROM based like Commodore computers, Microsoft could continually upgrade the operating system. Microsoft would go on to release new versions and amass a fortune. The very first IBM PCs had only 48 kilobytes of RAM. Like the VIC-20, it used cassettes to load and store programs.[3] The entire package sold for $1745. VIC-20 sales were unaffected, but the IBM PC was an Apple killer. With a floppy drive, the IBM PC sold for $2575, compared to $2788 for an Apple II Plus with an 80-column board and floppy drive. IBM sold more than 250,000 personal computers in the first year. It took Apple five years to sell 500,000 Apple II computers. The release of the IBM PC turned out to be the defining moment for personal computers. The world wanted a standard for computer software just as it had wanted a standard for VCRs and audio cassettes. Now it had one. * * * All of Spencer’s groundwork would be for nothing unless the mass consumer market could obtain VIC-20s easily. Tramiel was no longer content to sell computers from specialized computer stores and wanted to sell computers everywhere people shopped, much as he had done with calculators. “It was in the fall of 81 that we made the decision to totally change the VIC strategy in the USA,” says Spencer. “It was something Jack and I agreed on. Up until then, it had been sold primarily as a computer through office dealers, without much success.”

Atari’s domination of CES, which was due to video games, inspired Spencer’s new strategy. “My strategy was to re-orientate and market it as a consumer product aimed directly at the large video game market that Atari already had at the time.” Spencer and Tramiel began to take on the biggest challenge for the VIC-20. “We decided after a few months what we would do: we would open a consumer distribution channel.” Commodore would attempt to move their product into retail stores. It would be a difficult job. Earlier, Dick Sanford had reorganized the Commodore Dealer Network in 1980 to distribute VIC-20 computers. He loosened rules for distributors and tried to make the program more amenable to the low-cost VIC-20 computer. The changes allowed the VIC-20 to sell alongside the PET in computer specialty stores. He then tried expanding the network by g up new distributors, with no real success. “The key decision was to reposition VIC from where we had been selling it as a computer, which was mainly through the computer dealer network they had for the Commodore PET,” explains Spencer. “Jack and I discussed it and we pretty much decided to re-orientate the VIC to take on the games machine market and market it as a consumer product.” Although the VIC-20 was a home computer, they decided to compete with the Atari VCS—not with competing Atari home computers, which the VIC-20 could not match technologically. “Atari was being very successful on the games end,” recalls Spencer. “We said, ‘The VIC computer is a great game machine, and it’s also a computer. Why buy just a game machine when you can get a computer as well?’ That was the strategy we adopted.” It was one thing to talk about increasing distribution, but it was another thing entirely to convince stores to purchase your product and put it on their shelves. There were many worries and concerns from retail giants about stocking such a complex device. Spencer chose someone in Commodore who had hands-on experience in retail marketing. “Bill Wade, who was another Commodore oldtimer, got put in charge of developing that,” says Spencer. The no-nonsense, straight talking Oklahoma native was managing Commodore’s Mr. Calculator stores at the time. Spencer pulled him off the project and put him in charge of Commodore’s retail efforts.

“Commodore was kind of a weird place,” says Andy Finkel. “You sort of got to define your own role. You could take responsibility and pretty much do what you wanted to do as long as you succeeded and as long as you were willing to take responsibility. So we had people like Bill Wade, who was head of sales, because he was willing to take responsibility.” How long Wade lasted in the job depended on results. “You could build your empire and you were fine until you failed and then of course Jack would come down on you like a ton of bricks,” says Finkel. A marketing executive named David Harris (no relation to Neil Harris) brought in Commodore’s biggest retailer, Kmart. “David Harris was involved in that sale,” says Spencer. Harris had previously sold watches and calculators to Kmart. In late 1981, David Harris arranged a meeting between Jack Tramiel and Fred Shimp, the electronics buyer for Kmart. To make the deal happen, it was important to ease any worries Kmart had with the new product. Neil Harris often accompanied the VP of sales, Bill Wade, when they attempted to recruit new retailers. “I would do the product demo then they would go in the back room and work on the deal ,” says Neil Harris. “In order to open up Kmart, the buyer of Kmart said, ‘Look, I think this is a great product and it’s interesting and you’ve promised us a lot of advertising . The problem is when somebody walks into the store, nobody in my store is going to have any idea how to tell anybody about these things.’ He said, ‘No problem, I’ve got all these people in my company. We’ll send them out on the road as part of our package and we will train your store personnel.’” During the meeting, a grateful Tramiel promised Shimp he would Kmart as “our first wife” in mass-market computers. Shimp commented that some men go on to have other wives, but Tramiel promised, “We won’t forget you were first. Our first wife will always be special.”[4] Because of the promise to Kmart, Neil Harris and some of his coworkers found themselves travelling across North America. “I was literally on the road constantly travelling around for months,” recalls Harris. “You never knew what hotel you would be in next. It could be a great place and it could be something terrible. It was just part of the deal.”

Although Commodore employees expended a lot of effort, it was worth it to capture Kmart. “There were two teams. Dave Rogers and I were on one team, and there were three people on another team,” says Harris. “We would fly from city to city with our crates and our computers and our big monitors. this was in the days when we’re talking a 25 inch TV set.” In each city, the Commodore employees visited dozens of stores and unpacked their demonstration equipment. “We’d go into stores—not just Kmarts but there were others as well,” recalls Harris. “We’d get department heads from all the Kmarts in the area, and the store managers would come in, and we would show them how the machine worked and how to sell it and how to set it up.” Harris doubts their efforts actually helped Kmart very much. “In retrospect, probably six months later none of those people worked there anymore, but at least we did what we were supposed to do,” he says. Kmart soon became the world’s largest computer retailer. “Kmart sold more computers than any other company in the world during that time period. It was huge,” says Harris. Each Kmart set up a computer center with demonstration VIC-20s, software, and peripherals contained in glass display cases. The new displays attracted customers who stopped to try out the new devices. Retail distribution was much different from specialized computer stores. At ComputerLand, a store employee would approach new customers, test them to see how much they would be willing to spend, and then steer them towards an Apple II if at all possible. After all, ComputerLand stood to make more money if each customer bought a $1200 Apple II rather than a $299 VIC-20.[5] However, at retail giants like Kmart, there was no one to steer customers. They just researched the product as best they could and bought one. It was the perfect venue for Commodore. Tramiel sold VIC-20 computers cheaper to Kmart than they did to regular computer dealers and Kmart soon began selling the computers for under $200. Suddenly, Commodore dealers were stuck with overpriced VIC-20 computers they could only sell at a loss. “Jack burned ComputerLand,” says Yannes. “When he introduced the VIC-20 it was $299. Then he started selling it in Kmart for one hundred some bucks.”

The sudden price drop alienated every single Commodore dealer. Yannes recalls, “ComputerLand said, ‘How could you do this to us?’ Jack replied, ‘That’s business.’” When asked if Tramiel offered the computer chains a rebate to make up for the difference, Yannes laughs. “You’ve got to be kidding me! Jack Tramiel? No. No he would not.” Surprisingly, Tramiel saw no long-term value in maintaining good relationships with his Commodore dealers. “Long term value is not something Jack understands,” says Yannes. “It’s, ‘What can you do for me now?’ It’s all driven by greed.” Christmas 1981 was a turning point for Commodore. Kmart sold VIC-20s all across the nation and manufacturing plants turned out thousands of VIC-20s each day. Once Kmart began selling computers, Kmart’s competitors began approaching Commodore. “Bill Wade did an excellent job of setting up a network of consumer sales distribution,” says Spencer. Retail giants such as Sears, Montgomery Ward, Toys’R’Us, Musicland, and J.C. Penny signed up to distribute Commodore computers. The VIC-20 was now a guaranteed success. Commodore was suddenly a force in North America. Tramiel had lost the early momentum of the PET due to a lack of distribution. Now, Commodore had more distribution channels than any other computer company, including the mighty Radio Shack. The new retail strategy would help them eclipse all other computer makers. [1] Dennis Barnhart later worked for Chuck Peddle, then became a Silicon Valley legend when he started a company called Eagle Computer. On the same day as the IPO, he purchased a Yacht, went for lunch and celebrated over drinks, then smashed his $70,000 Ferrari into a guardrail. He died from his injuries. [2] Commodore employees were the first to see Atari games. “We had all the early stuff before it was released, and we played all these Atari games,” says Russell. [3] The cassette commands remained in the operating system until MS-DOS 5. [4] Michael Tomczyk, The Home Computer Wars (Compute! Publications, 1984), p. 235.

[5] According to author Jim Butterfield, ComputerLand also routinely steered customers away from the less profitable PET computer. Later, when word traveled back to Commodore, they severed relations with ComputerLand.

CHAPTER 24

Entertainment 1981

Michael Tomczyk’s initial licensing deals for adventure games and chess would prove profitable, even if they did not really capture the same gaming space as the Atari 2600, which was where Kit Spencer wanted to compete. An investor relations consultant named Steven Greenberg represented both Commodore and Bally-Midway. He was aware of both companies’ goals and spotted an opportunity for licensing. Jack Tramiel, Irving Gould, Steven Greenberg, and Michael Tomczyk flew to Chicago to meet with Bally-Midway executives. The company got its start in the gambling industry making slot machines and running casinos and hotels. There were numerous rumors that management was actually in the mafia. In the 1970s, CEO Bill O’Donnell famously attempted to deny the rumors after several government investigations uncovered links to organized crime. “Some people claim that there were some tough characters running Bally-Midway, and I can that,” says Tomczyk. The Bally-Midway executives were amicable towards Commodore, despite treading on their trademarked games earlier. Tramiel purchased the rights to port several of their coin-operated games, including Omega Race, Wizard of Wor, Seawolf, Clowns, and Gorf. Under the deal, Commodore in-house programmers would develop the games and Bally-Midway would collect royalties for each game sold. “They had a list of games that they wanted us to license and not all of their games were on it,” says Andy Finkel. Commodore would also get right of first refusal for any games they released in the future. Surprisingly, Bally-Midway did not insist on Commodore licensing Rally-X, the

Namco developed game on which Bally-Midway owned US rights. “Radar Rat Race was over and done with by the time we did the Midway deal,” says Finkel. “I think they had a bias towards the ones that they developed in the US themselves, as opposed to the ones they got from other sources.” Bally-Midway’s part of the deal was straightforward. “They would provide us with the actual physical upright arcade game and the right to do whatever we wanted to make it on our VIC-20, but they would not give us source code,” says Finkel. “They’d give us the manuals, the arcade machine, the schematic diagram and that was it.” The Bally-Midway executives accepted Tramiel’s initial offer quickly. “We walked into the room, they agreed to it and we walked out,” says Tomczyk. “Jack turned to me and said, ‘I wish all of our business negotiations would be handled like this.’” The contract with Bally-Midway was even simpler. “Jack asked me to write a letter,” says Tomczyk. “I wrote a letter to Bally-Midway documenting the royalty and indicating that if the economic circumstances changed, and we are forced by competition to lower the price, that we can come back and renegotiate the royalty. That was all the letter said.” The executives signed the agreement and Commodore was on its way to releasing quality arcade games for the home. * * * Commodore programmers soon began their VIC-20 arcade conversions. This time, instead of BASIC, they programmed in machine code, using a CBM computer to compile the code. Weeks after g the agreement, several arcade games arrived at Commodore’s King of Prussia location and the programmers began planning their code. “They shipped over a couple of arcade machines,” recalls Neil Harris. “Those guys could play them, try to understand exactly what was going on and reproduce them through software as best they could. Andy Finkel and Bill Hindorff really did the lion’s share on those ports.” Finkel wanted to create the most accurate duplication of the arcade game as possible. “We tried a couple of times to get the source code but they said it was written in a proprietary programming language and that it wouldn’t help us anyway,” he recalls. Eventually the programmers got in with the Bally-

Midway game designers. “It turns out the proprietary programming language was pretty much Forth. It was a Forth variant with assembler extensions.” Without access to the source code, it was difficult for the programmers to determine exactly how the game was supposed to play. “We did those games based on playing them for hours and hours and hours. And that was rough I tell you,” he says sarcastically. Arcade games require a high skill level, which made it hard for the programmers to study the game completely. “We weren’t good enough to get to those final levels so we would ask the programmers and they would usually tell us if we knew the question to ask,” says Finkel. “But I always think the source code helps when doing a port.” Finkel was the most skilled programmer of the VIC Commandos, followed closely by Bill Hindorff. “I was probably the one who had the most experience in software engineering because that’s what I studied in college,” he says. “I had more of the basic knowledge of what software engineering is all about.” Finkel’s interest in computers started in high school when he met the slightly older Neil Harris. “We were in different high schools but we were both in computer programs,” says Finkel. When a Philadelphia teacher’s strike occurred in 1972, Finkel found himself locked away from the school’s PDP computers. “There was nothing else going on for a couple of months, but we couldn’t stand being away from computers for that long. So we got permission and went into school to use the computers. That’s where I got to know Neil and we’ve been friends ever since.” Later, Finkel attempted to build his own home computer. “I built a TV Typewriter way back in the ’70s from Don Lancaster’s article,” he says, referring to the September 1973 article in Radio-Electronics magazine. The TV Typewriter was limited, and Finkel gravitated towards the 6502 microprocessor. “I was basically just hooked on computers. When the KIM-1 came out, I went to Norristown to the factory and made them sell me one,” he recalls. “They weren’t really set up to sell directly to high school kids who drove to the factory, but I just couldn’t wait at that point. I met Chuck Peddle back then.”

In college, Finkel split his engineering courses between hardware and programming. “When I went to college, it was basically half and half. I got a bachelors of electrical engineering from Cornell in 1979. I took a lot of computer courses at the same time. Even though I was there for electrical engineering, I made it as much programming as I could. When I got out, I decided I liked the programming stuff better.” Finkel took on the job of porting Omega Race, a relatively new title released by Bally-Midway in 1981. When he was done, Tomczyk promised he could take the arcade machine home. “Each of the programmers was allowed to buy the machine from Commodore for a nominal price,” says Finkel. “I think it was 100 bucks after you finished the game.” Finkel’s junior programmer on Omega Race was Eric Cotton. “When we started doing our own cartridge games, we grouped up into teams of two, a senior programmer and a junior programmer,” he recalls. The arcade game used vector graphics, in which the game drew objects using thin glowing lines on the picture tube. Bob Yannes believes the VIC-20 was more than adequate for the task. “The VIC-20 had pretty much the resolution and graphics capability of the arcade games at that time,” he says. The VIC-20 used raster graphics, meaning it drew pictures out of small dots. “The VIC-20 has a programmable character display but you could turn that into a bitmap by putting up all the possible characters on the screen and set the pointer for the characters. Instead of pointing to ROM, you point it at RAM, and then you write in the character definition areas.” Once he had his bitmap display, he implemented methods to draw lines. “It’s a bitmap graphics program on the VIC-20 and I just wrote a set of fast line drawing routines with anti-aliasing and used that instead of their vector display,” explains Finkel. “Anti-aliasing for grey-scale lines is really easy. It sounds fancy, but it’s trivial.” Finkel incorporated every detail of the arcade game into his VIC-20 port, including sound effects and a storyline that appeared during the attraction screen. It was an almost exact clone of the arcade game. When done, Finkel opted to purchase the original arcade machine and brought it home for his basement. While Finkel programmed Omega Race, another team led by Bill Hindorff[1]

created GORF. The name stood for Galactic Orbiting Robot Force. GORF was an elaborate version of Space Invaders with diverse game play throughout five distinct levels. One level consists of dive-bombing aliens in the same vein as Galaxian, while another level has a large space ship, which a player must shoot full of holes until the alien inside is destroyed. GORF was an ambitious game for the 8-kilobytes allowed in the VIC-20 cartridge. “GORF graphically is more complicated,” says Finkel. “That was the big factor in trying to get it to fit and work well.” The demands of the project meant it took longer to complete than Omega Race. All the programmers, including Jeff Bruette, Eric Cotton, Mike Scott and Jim Snyder ended up contributing something to the game. “We actually had several programming teams at the time,” says Finkel. “After my team finished Omega Race, we moved over and helped the GORF people finish theirs.” Finkel saved time by recycling some of his code from Omega Race. Surprisingly, VIC-20 GORF was an almost exact translation of the arcade game. The main difference was that the third level, named Galaxians, had to be removed from the port due to copyright infringement on the Galaxian name. The third game of the Bally-Midway deal, Wizard of Wor, would never appear on the VIC-20. Although Commodore announced and d the title, development eventually fizzled away. After months of programming effort on Wizard of Wor, the game was cancelled during a meeting. “I talking about it,” recalls Finkel. “We had meetings on it with the head of the game department, and we just decided it wasn’t quite there.” Finkel believes the game was as good as it could have been on the VIC-20. “The people who were working on it did a really good job but we decided it just didn’t have enough of the Wizard of Wor thing that we were hoping for. It was a limitation of the VIC-20, not the programmers. The guys did a great job but the hardware just wasn’t enough for it.” Neil Harris is proud of the first three games in the Bally-Midway deal. “Wizard of War, Gorf, Omega Race—those were all amazing ports,” he says. The VIC-20 soon began competing in the same market as the Atari 2600. “It

turned into a cartridge machine,” says Bill Seiler. “People bought a lot of games in cartridge form. It was a pretty good game machine.” * * * Stagnant economic growth and inflation characterized much of the 1970s. In the early eighties, the Federal Reserve raised interest rates to incredible levels to combat inflation. Jack Tramiel saw an opportunity. Rather than pay his suppliers the money Commodore owed them, he decided to withhold payment so he could keep the money in the bank and earn interest. “Interest rates were 21% in the Bahamas,” says Michael Tomczyk. “We made a million dollars one year on interest income alone.” As a side effect, observers believed Commodore was having financial difficulties. “The whole world reported, ‘Commodore is going bankrupt because they are not paying their suppliers. They have a cash flow problem.’ And Jack just laughed like hell,” says Tomczyk. It was another one of Tramiel’s unorthodox business strategies to maximize Commodore’s profits. After Spencer’s arrival, Commodore began to undergo a fundamental change, culminating in mass-market distribution throughout Kmart stores. The company began a period of intense growth faster than Commodore had during the calculator or PET days. Unfortunately, rapid growth in corporations often produces casualties. “The decision to move VIC to a consumer strategy probably contributed to Dick’s decision to leave,” says Spencer. “Dick felt that the way Commodore was moving was eroding his own position in Commodore from that of being, in many ways, the closest to Jack.” In November 1981, Sanford decided to leave the company. “I left Commodore because I wanted to build my own company,” he says. “I loved Jack. I bled for the company—as did Chuck—but it was time. I left because I wanted to do my own thing.” Tramiel was not ready to part with one of his most trustworthy executives. “Jack gave me an offer in one of our companies,” recalls Sanford. Tramiel asked him, “Do you just want to tune out and run around the woods? I’ll send you to Switzerland. Go to Switzerland, hang around and run the mountains or something. Just cool down.”

Sanford smiled and shook his head. Tramiel asked, “What do you want to do?” Sanford replied, “I want to build a company. I want to take what you’ve taught me and I want to do it on my own.” Tramiel thought for a moment and then made another offer. “We’ve got this LCD company down in Newport Beach, California. What young guy wouldn’t want to go to Newport Beach? It’s a great place. Take the company public for us.” Sanford replied, “Jack I love you man, but I just want to go out and do it on my own. I don’t want it given to me.” “And I did. So I left.” * * * Now that the VIC-20 was in wide distribution in Kmart and other retail stores, Kit Spencer knew it would make sense to spend money on advertising. “We made that decision about six to eight weeks before the January CES,” he says. “I put together an advertising campaign for the VIC.” Tramiel respected Spencer’s judgment because of his wild success in the UK and authorized the advertising campaign. “He could see the potential and I think he trusted me to put it together,” says Spencer. Through television and magazines, advertising would bring the VIC-20 to the minds of customers across North America. Tramiel suggested a company spokesman months before to Michael Tomczyk and now Spencer picked up the idea again. Like most manufacturing companies, Commodore would not produce the ments. Instead, they hired an ad agency called Kornha & Calene. “We had a really, really great advertising team comprised of a New York vice president of advertising and a red haired woman who were just genius,” says Tomczyk. “They were a very tough, wonderfully creative, strong team. We were really lucky to have them.”

The ad agency was responsible for many high-profile brands including Arm & Hammer baking soda (with the boxes wearing scarves and earmuffs and chatting with each other inside the freezer), Head & Shoulders (the man with two different dandruff shampoos on each side of his head saying, ‘This side tingles’), and Black Flag insecticide. The agency would film the commercials and shoot the photographs in New York. In the early eighties, most computer companies had a spokesman named Bill. Radio Shack had Bill Bixby and later Bill Gates in a stunning pink-collared shirt and white sweater. Texas Instruments had Bill Cosby. Commodore would also have a Bill. The ad agency made an inspired choice for company spokesman. Almost every computer nerd in the 1980s was a Star Trek fanatic. A sure sign of this was the number of early computer games that focused on Star Trek. The characters from Star Trek even routinely appeared on the cover of early Byte magazines. The two lead roles of Star Trek belonged to William Shatner and Leonard Nimoy. The ad executives seriously considered Nimoy, who portrayed Spock, the authority on science and technology aboard the Enterprise. However, he was already representing a Magnavox computer, so they approached William Shatner. “It came from input originally from Kornha & Calene,” says Spencer. “They were the people who suggested him.” Shatner had lived an uneven life since the cancellation of Star Trek in 1969. He released an album of offbeat renditions of popular, psychedelic songs called The Transformed Man. In the seventies, between guest spots on network television shows, he performed in stage plays across the United States, touring in a small pickup truck and sleeping in the back under a fiberglass shell. His low point came in 1978 when he released a two-album set of songs, poetry, and book readings gathered from Star Trek conventions titled Captain of the Starship. Shatner came into favor again in 1979 with the release of the Star Trek: the Motion Picture, which was moderately successful. Tramiel was prepared to spend 6% of revenue from the VIC-20 on advertising, and the down-and-out Shatner fit his budget. “We decided we had to have enough to kick it off, to show it to the consumer dealers and get it going, but it wasn’t that high initially,” says Spencer

Tramiel did not have the final say on such a large expenditure, however. The ultimate decision lay with Irving Gould and his board of directors. “Obviously, to make the spend on TV advertising was a big decision for Commodore,” says Spencer. “I would explain to Irving what was happening there. But Jack and I were in agreement and he pushed it through the board okay.” In 1981, when Commodore ed him, he was in the middle of filming Star Trek II: the Wrath of Khan, which would elevate him to super-star status. It was fortunate timing for Commodore, provided Shatner could find some time away from the set to film the commercials. Principal photography for Star Trek II began on November 9, 1981, and ended on January 29, 1982. Kornha & Calene hoped to film the commercials before the January CES. “The Star Trek movie was also being filmed at that time,” says Spencer. When the director, Nicolas Meyer, was filming scenes without Shatner, his agent was able to schedule a slot for filming the commercials. Decades later, Spencer recalls the message. “We created the Shatner VIC commercials with a Star Trek type feel and the strategy of, ‘Why buy just a game machine when you can play great games and learn computing too with a Commodore VIC-20.’” Kit Spencer, Michael Tomczyk, and scriptwriter David Rogers monitored the shoot. “I was there when we first set up the set and got it going and introduced him,” says Spencer. “He was a very professional actor and he put it across well.” Kornha & Calene hoped to use the Star Trek brand surreptitiously. “The first thing they did was provide him with wardrobe for the commercial shoot that was kind of a sweatshirt that sort of looked like a Starfleet uniform,” recalls Neil Harris. “He said, ‘No way. I’m not doing your commercials as Captain Kirk.’ So they went out and got him a normal looking wardrobe.” “He grudgingly accepted a sweater that was kind of reminiscent of Star Trek,” says Tomczyk. After things got going, Spencer left to do other business in New York with Irving Gould. “I ed it onto Michael Tomczyk to do most of it after that,” he says. “I was just too busy with other things. It might have been nice to hang around with William Shatner there for a while, but it was pretty hectic in those days.”

Trekkies Neil Harris and Andy Finkel provided technical . “I actually worked the computers in the background when Shatner was doing the commercials,” says Harris. Star Trek was an obsession for Harris and Finkel, and it took effort to remain calm and normal around the object of their obsession. “I’m a big Star Trek fan so I liked it a lot,” says Finkel. “I got to go to the set once while the commercials were being filmed and that was fun meeting Shatner. He seemed like a nice guy and a professional actor.” The two were mildly curious to see how the captain of the Starship Enterprise would react to the latest in technology. Unfortunately, the former starship captain knew little of computers “He was hired to be the spokesman but he was not a computer guy at all,” says Harris. Shatner sat down in front of the demonstration VIC-20, which Harris had connected to a color television, and began slowly typing. As the Commodore crew stood around, they heard a click, then another click, then a couple of clicks, then a few seconds later another click. The captain of the Enterprise had just typed hello using the index fingers of both hands. “He thought it was wonderful and said, ‘This is obviously the future,’” recalls Tomczyk. He quickly filled up the screen with random gibberish, amazed that a television could display things other than television shows. It was obvious it was the first time he had used a computer. Their spokesman was not very knowledgeable, but he was enthusiastic. “He was keenly interested and wanted to know how it worked,” says Tomczyk. “Basically it looked like a typewriter hooked up to a television set. His reaction was the same as anyone who sees a computer for the first time.” After filming different sequences most of the morning, the crew went for lunch. “I sat next to him and he was charming, intelligent and interesting,” says Tomczyk. While eating, Shatner reminisced about his early spokesman work with Promise Margarine in 1970. “He said the only thing he did wrong in his career was he did Promise Margarine commercials after Star Trek, and he should have done technology commercials instead. That’s the only regret he had,” says Tomczyk. With the new Commodore commercials, he was out to rectify his earlier misstep.

The crew spent the afternoon taking pictures for print ads. As a smiling Shatner posed in a grey business suit from Bloomingdales, the photographer took hundreds of photographs in a variety of poses. Kornha & Calene would later incorporate his image into print ads. At the end of the day, Harris and Finkel made sure Shatner did not leave empty handed. “We gave him a bunch of Commodore stuff, VIC-20s and PETs and so on,” says Finkel. Shatner later required hands-on help from Commodore employees to figure out what to do with his PET computer. “He was exceptionally intrigued to the point that we ended up sending someone over to his hotel or house to help show him how to use it,” says Tomczyk. It would take months for Kornha & Calene to edit, print and show their ments—too late for the Christmas season. Nevertheless, the advertising had the potential to increase their brand awareness dramatically. * * * In 1981, Tramiel decided he wanted to move Commodore closer to New York. Robert Russell believes necessity motivated his decision. “He couldn’t hire people on the west coast because of his reputation,” he explains. “People didn’t want to work for Jack.” Intense competition for engineering talent made California unattractive. “In those days, if you were pissed off, you could walk across the street and get $5,000 more,” says Russell. “At Commodore, you could earn $18,000, but once you had been there a year and people knew you, it was easy to get more money.” Tramiel resorted to another tactic to hold his employees captive. “A trick of Jack’s was to give people stock options that he knew they couldn’t afford to exercise,” explains Yannes. “Then he lent them money to exercise the options, so they became indebted to the company until they were permitted to sell the stock.” Tramiel’s shortcuts could only last for so long. “In 1981, he said, ‘I want to move the engineers out [to Pennsylvania],’” recalls Russell. “It was one of those things where I didn’t really want to move but I was asked.”

Russell tried his best to remain on the west coast. “People came and went in those two years, but they were always trying to consolidate people on the east coast,” recalls Russell. “I was saying, ‘No, no, no,’ and the movers were sitting outside my house for two days saying, ‘Yes, yes, yes. The company says you’re moving and we can’t leave until we move you.’” Finally, Russell relented and moved to the east coast. The few engineers who made the transition moved in with the semiconductor engineers. “We spent a couple of years at MOS Technology in that building,” says Russell. Although Russell migrated, many engineers resisted and stayed on the west coast for a few more years. “There were some, like John Feagans, who didn’t want to move and wouldn’t move,” explains Russell. “There were some good engineers that just didn’t move and we lost them.” The Japanese engineers, Nobuo Aoji and Mr. Fujiyama, remained with Tramiel at 3330 Scott Boulevard. Charles Winterble, an east coast engineer, felt it was a mistake to lose their California presence. “You needed to have a California connection,” he says. “There are a lot of creative people in California working on exciting things. The cross-fertilization in California when you are sitting down for lunch with other people is an important interaction.” * * * Prior to their departure, Scott Patterson and Glen Stark had developed a single floppy disk drive for the TOI computer. Now that circumstances had ended TOI development, it made sense to adapt their work into a disk drive for the VIC-20 —even if the small memory space of the VIC-20 barely justified the need for one. The disk drive, called the VIC-1540, was virtually identical to the PET 2031LP disk drive (LP stood for Low Profile), except it used the new Commodore serial bus, rather than the expensive IEEE-488 interface. A company called Alps Electronics manufactured the drive heads. The new drive required less processor power than the over engineered dual drives for the PET. “The original [PET] drives were written for two 6502 processors,” explains Russell. “One did the GCR encoding and a second dealt with the IEEE interface. Basically the 1540 drive was getting one processor to

do all that stuff.” The disk drive would be a severe test of the new serial interface design. One of Russell’s co-workers at Commodore had been responsible for developing the disk drive code. “Bob Fairbairn was supposed to write the drive interface,” says Russell. Fairbairn was a competent programmer and delivered good code, but the hardware contained a problem. “He had done all his work, but when it came down to the final thing, it didn’t work. The 6522 chip had a glitch.”

Bob Fairbairn prior to departing Commodore.

The VIC-20 used the 6522 Peripheral Interface Adapter chip to attain fast disk drive access. “We were originally supposed to use the 6522 high speed lines, but they didn’t work right,” explains Russell. “Unfortunately, Bob Fairbairn wrote all this code and spent all this time, and it wasn’t reliable and would lock up the computer and the peripheral. When it was all ‘done,’ none of it really worked.” At the time, Tramiel was pushing his engineers to move east, and Fairbairn opted not to make the move. “Bob Fairbairn was leaving town, going back to the Midwest,” explains Russell. Fairbairn quit Commodore and returned to Illinois, leaving the code unfinished and nonfunctional. The job of getting the disk drive to work fell to Russell, who was now working out of MOS Technology on the east coast. He began studying the problem in detail. “We were sitting there looking at the 1540 drive, and the units were locking up,” he recalls. “Neither one was doing anything.” Russell struggled with the chip problem, but eventually time ran out. “Jack came into my office and said, ‘We’re shipping on time in ten days,’ and he wanted to close the code,” he says. “It was like, ‘Yes sir!’” Russell now had no option but to implement a quick fix, regardless of performance. “They didn’t care how fast it was, so I just wrote code that worked,” he recalls. “I did the modifications to make the serial stuff work inside the 1540.” Every time Russell revised his code, he had to burn the code into a chip and place it into the drive to test it. “At the time, we were [burning] EPROMs to meet ship dates, which were costing a fortune,” he says. “Luckily we had MOS Technology. We could do ROMs as fast as we could get the mask layer to them.” Russell needed a quick solution. Since the timing of the I/O chip was off, the only way for him to make the disk drive work was to slow down the transfer speed by four times. “That’s when I wrote the really slow code,” he explains. “I just did the slowest, surest thing that I thought would work.” It was not an elegant solution. Although the PET 2031 was a fast disk drive, there was no way

to make the 1540 perform as fast on the VIC-20. Russell’s engineers were impressed with his hurried solution. “I attribute the serial bus to him as a heroic effort when the [PIA chip] was found to have a glitch,” says engineer Bil Herd. Russell would display the 1540 drive at the CES show in January 1982. * * * The field of computer telecommunications was building momentum in the early eighties. Michael Tomczyk was aware of the trend and wanted the VIC-20 to have the ability to connect to on-line services. Completely on his own, he began seeking manufacturers to develop a low-cost modem. “Mike understood quite a few things about where our business was headed,” says Spencer. “He was very gung ho on that one and key to its implementation. That was a good decision.” An engineer named Dennis Hayes invented the personal computer modem in 1977. It was a critical piece of technology that allowed telecommunications industries to grow. “He had all the patents on the original modem stuff,” says Russell. Hayes compatible modems soon became the industry standard. “Tomczyk got Hayes involved with us to get the modem going,” recalls Russell. “I worked with Hayes on the modem to work with my serial bus design.” At the time, modems were selling for $400. Tomczyk knew this was too high, considering the cost of the VIC-20 was under $300. He wanted Commodore to sell a modem for under $100. In order to remain profitable, the manufacturing cost had to be one third of the retail price, or $33 in parts. “We wanted the simplest, cheapest thing,” says Russell. The low-cost requirement would push Hayes’ creativity to the limit. “We were basically building a prototype card hanging off the back of a VIC-20 prototype in Santa Clara,” says Russell. Occasionally, Tomczyk dropped by to check on the progress of the modem. “He might have come by and bothered us in that little cubicle.” The modem was the first big test of Russell’s serial interface. “I was sweating whether we could figure out how to make it work with my serial

implementation,” says Russell. “That was the first time anybody tried to do something other than just running a stupid serial port off it. That required some serious hand shaking.” Russell was impressed with Hayes’ engineering ability. “That guy was a genius,” he says. “He was putting stuff together that blew me away.” Most modems at the time had a special device that clipped to the phone receiver, called an acoustic coupler. The Commodore modem omitted this device to save money. “It was straight to the phone jack,” says Russell. “I don’t know if it was the first one, but I know that he was trying stuff out that hadn’t been tried before.” Russell and Hayes were able to shrink the circuitry to fit into a three by six inch case. To give s something to do with the modems, Tomczyk approached the leading on-line services and requested free subscriptions and trial offers. It was a brilliant marketing strategy, which other on-line services such as AOL would later emulate. “We did a deal with Compve to include free hours on their service with our modem,” says Spencer. Each who purchased a modem received a free subscription to Compve and an hour worth of on-line time. The Source and the Dow Jones Information Service also gave one free hour, but they stubbornly would not waive the subscription fee.[2] The VIC modem came packaged with dramatic cover art from the same artist who created the game covers. It was a spectacular package. “I would give both Michael Tomczyk and some of the technical people credit for envisaging where potentially the computer business could go and where it’s at now in of the Internet and things like that,” says Spencer. [1] Bill Hindorff went on to have lucrative gaming career working for Atari, Midway, and 3DO producing the Army Men series of games. [2] After Commodore released the VIC-20, Compve quickly outpaced The Source.

CHAPTER 25

The VIC-40 1981

Al Charpentier had been hard at work deg his new video chip since early 1981. His previous failures with the TOI chips made him more determined to succeed. The biggest limitation to the design, which he dubbed the VIC-II, was the number of transistors that could fit onto a microchip. At the time, designers created their chips using components and pathways that were 5-micrometers in width.[1] “Back then, you’re probably talking under 10,000 transistors, with maybe 100 capacitors on a chip,” he explains. Charpentier had the freedom to pack as many features as he could fit on the surface of the chip. To accommodate this limitation, he developed a prioritized wish list of features. He would lay out the most important features first and when he ran out of room on the chip, his design was complete. “The decisions that were needed at various points became fairly automatic,” says Charles Winterble. Because of memory limitations, Charpentier decided to use a screen size of 320 pixels wide by 200 pixels tall, or 64,000 pixels total. With each pixel represented by a single on-off bit, this meant it would only require 64,000 bits, or 8000 bytes of memory to represent a screen. Since North American televisions refreshed the screen 60 times each second, the chip had to draw an incredible 3,840,000 pixels per second. Charpentier designed the video chip to use 16 kilobytes of memory—8 kilobytes for displaying the background and the rest of the memory for character sets and sprite data. He allowed several different modes of display, allowing programmers more freedom. At 320 x 200 pixels, a game programmer could display images with only two colors on the screen. If the programmer wanted

four colors at once, then they would use a resolution of 160 x 200 pixels. Charpentier was most excited about generating color. “I put in a pretty complex color generation scheme that was way ahead of its time,” he says. The young designer knew scrolling games would be popular, so he included functions for shifting the screen by one or more pixels in each direction, just like the original VIC chip. Video chips can display bitmapped graphics pixel by pixel. Charpentier allowed bitmapped graphics, but not much beyond plotting individual dots. He included no hardware routines for drawing lines, curves, or filling in polygons. If a programmer wanted to create wire frame objects or polygons, the 6502 processor would have to perform the calculations, resulting in slower 3D games. Charpentier dedicated two-thirds of the chip surface to sprite functions. With the architecture complete, it was time to begin the layout. The tools available to the engineers had improved since the days of the VIC chip. “They were using computers to do the layout, but they were still checking plots by hand,” says Bob Yannes. “They were still drawing the chip layouts on Mylar and then digitizing them.” Yannes witnessed the evolution of semiconductor design tools. “Originally it was a real drafting area with big tables, pencils, and T-squares,” he recalls. “They had only one station for the computer, so that’s why they would draw them by hand and then the guy working at the computer station would digitize them. Then they would go in and edit them if they needed to. Eventually they all had their own computer stations.” Charpentier worked with two draftsmen and a CAD operator (Computer Aided Design). The CAD operator sat in front of a special-purpose Applicon system and created a blueprint for the chip. The two draftsmen would then use his schematics to lay out the surface of the chip, hunched over large pieces of vellum paper. “The chip work he did by hand with his team was phenomenal,” says Winterble. “It was all hand layout work.” The most difficult part of the process was checking the layout, which most engineers found mentally draining. For hours at a time, Charpentier hovered over his diagrams, bellowing smoke and measuring distances. “He always had a cigarette in his mouth,” says Yannes.

Charpentier was able to progress quicker than other semiconductor designhouses because he had one tool no other had: an on-site fabrication facility. “Having the fab area was such a key,” says Winterble. “We had people coming in at all hours to make sure that this stuff could be ed from hand to hand to hand to get a very fast turnaround in our chip area.” Normally it took a few weeks to manufacture a chip, but if the engineers needed a chip for CES, the fabrication facility could produce a new batch of chips in just four days. For debugging purposes, Charpentier sometimes created a small section of the chip to test. “We didn’t have emulation tools to speak of for the time,” recalls Winterble. “What we would do is we would build certain parts of the circuit on a special test chip.” The system worked wonders for development. “Once it’s in the whole chip system, you can’t get at stuff,” explains Winterble. “You take five or ten of the different key circuit areas, put them on a test chip, and run it through fab real fast. Then you bring out the [test] pads so we could probe and make sure things worked properly.” When time allowed, Winterble and Charpentier left the offices to visit the local Valley Forge Golf Club. “We’d try to sneak out in the late afternoon sometimes and catch nine holes,” says Charpentier. “We were just hacking around and having some fun.” As the project progressed, the engineers obsessed over the project and barely left the premises at MOS Technology. “We were working just ungodly 16-hour days. A couple of times I put in 24 and 36 hour days,” says Charpentier. “When you are cutting edge, it takes a ion.” Without management or marketing interference, the engineers focused everything on the chip. “We were just kind of doing what we thought was the right thing to do,” says Charpentier. “We were having fun. It was during that time I call the Wild West era in that we were cutting new ground and no one knew where it was going.” Once his team finished the layout, it was time for the first of the complete chip. It was a nervous time because so much could go wrong. “The one thing you don’t want to have happen when you first come out with a chip is you don’t want a data bit to be dead on a data bus or an address line to be dead inside the

chip because then you can’t do anything,” says Yannes. Near the end of November, it was clear it would be difficult to have chips out in time for the January 1982 CES. Then Jack Tramiel stepped in and applied pressure. “Jack was constantly at odds with the chip designers who told him it would take six months to do a mask and the technical implementation of a new chip,” recalls Michael Tomczyk. “He kept beating on them and I was present in some of the meetings where finally they said, ‘Well, if we close down the whole factory and did one chip we could probably do it in two weeks.’ And he said, ‘That’s exactly what I want.’ That was a big breakthrough with the VIC-II chip.” MOS Technology now devoted all their resources to making the VIC-II happen. Winterble had to modify the NMOS process to produce a workable chip. “We played around with our processing just to get the speeds up,” says Winterble. “We kept making our oxides thinner and thinner so it would make it.” Normally, chips are full of bugs on the first and rarely work properly. Due to the incremental testing of smaller parts of the chip, the complete VIC-II chip worked on the first —almost. “I’ve heard Charlie say everything worked except one sprite,” says Yannes. The engineers wasted little time in connecting the chip to a makeshift computer to see the display. “Our first of silicon, enough of that video chip worked that we were able to jury rig a computer that day,” says Winterble. “It was a phenomenal effort to get that much working on that .” * * * Bob Yannes had been deg his sound chip since early 1981. Charles Winterble re the dedication the rookie engineer had for the project. “The most important thing in Bob’s life at the time was the sound chip. He always wanted to do a sound chip.” The engineers called the chip SID, which stood for Sound Interface Device. The friendly three-letter name was in the same tradition as TIM, KIM, PET, and VIC. They also gave it the product number 6581. Winterble wanted to make certain Yannes could meet the aggressive January 1982 CES deadline. “The most common problem is the engineers just get too many creative ideas and lose track of the priorities, and the result is you get nothing done,” he explains.

The experienced engineer told Yannes, “I’m not going to tell you what you can and can’t put in your sound chip, but you know we’ve got a date, and we’ve got a cost issue. Within those constraints, you be as creative as you want to be.” Yannes and Winterble created a list of features. “He wanted something very elaborate,” recalls Winterble. “I put together a die size for Bob, and made a prioritized list of must-haves and might-dos.” Yannes had the freedom to include as many features as he could fit onto the surface of the chip. “As a little team, we worked out pretty damn good methods,” says Winterble. “That’s why we got so much done in so little time.” Unlike Charpentier, Yannes did not look to the competition for inspiration. “I thought the sound chips on the market, including those in the Atari computers, were primitive and obviously had been designed by people who knew nothing about music,” he says. The biggest difference between Yannes’ chip and other sound chips was the precise frequency control. “With most of the sound effects in games, there is either full volume or no volume at all,” he explains. “That really makes music impossible. There’s no way to simulate the sound of any instrument even vaguely with that kind of envelope, except maybe an organ.” A sound synthesizer was the right strategy for producing sound on early computers with limited memory. Sounds in modern games can be prerecorded and digitally stored, resulting in files larger than several megabytes each. With only 64 kilobytes to work with, the SID had to be capable of creating great sound from very little data. Most people look at a tiny fingernail sized chip and wonder how it could possibly generate sound. The secret was in a tiny mechanism on the chip, built from transistors, called an oscillator. The oscillator produces an electronic signal, which travels to a signal amplifier and finally the speaker. The number of voices a chip can produce is important for the quality of music and sound. Each voice can produce a different instrument, so a chip with only two voices will produce dull music compared to a chip with eight voices. Yannes initially planned to an astounding 32 voices on his chip. He wanted to build special circuitry to share the oscillator, called a multiplexer.

No one had ever attempted a multiplexing sound chip before. “I wish I had done it because it would have been the first chip implementation of a wave table oscillator,” he says. Unfortunately, multiplexing was too low on the list of priorities. “We just didn’t have the time to get the design and layout done, so once we got one voice working we just replicated it three times in the layout,” says Yannes. “If you look at a SID chip under a microscope you’ll see three chunks that look the same.” The chip would have three voices. The chip could produce four base sounds he called saw tooth, noise, pulse, and triangle because of the shape of the waveforms. His fellow engineers were under the impression he was deg a sound chip exclusively for computers and video game consoles. “That’s what Al and Charlie thought!” reveals Yannes. “I really wanted to do a multi-track, polyphonic music synthesizer.” Yannes even snuck in some features that had no relevance to computers. One of the input pins on the SID chip could receive signals from another source, such as an electric guitar. “We were working so fast on the VIC chip and the SID chip that no one was looking over my shoulder to see what I was doing,” he recalls. Yannes even included a lookup table in the chip to convert data from musical notes into the equivalent frequencies. However, Winterble objected to the extra silicon it would use. “Charlie axed that, and he was totally right,” says Yannes. “It would have been a big waste of silicon just for a minor convenience.” After only four or five months, the design was finished and ready for production. Yannes was able to include most of the features he wanted. “He got about twothirds or even three-quarters of the list done,” says Winterble. Yannes was not fully satisfied with the SID chip, but happy to release it on schedule. “We were so close to the thing we were worrying about every little detail,” he says. “If I had taken the time to do the multiplex oscillator and timing design, I don’t think we ever would have gotten the product out in the timeframe.” However, Winterble put Yannes’ achievement in perspective, telling him, “This thing is already 10 times better than anything out there and 20 times better than

it needs to be.”[2] * * * Nine months after Winterble spoke with Tramiel about a new project, Charpentier and Yannes had their chip designs ready. “I look back on that and I’m still amazed,” says Charpentier. Compared to other semiconductor development cycles, the VIC-II and SID chips took about half the time to complete. “I don’t think anyone had ever done that before and I don’t think anyone at Commodore could have done it again,” says Yannes. Charpentier credits the speed of the project to the freedom given to the engineers. “Part of it was we were just left alone,” he explains. “There weren’t a lot of people in our face during the chip design process. We were in control of the design. Bob, myself, and Charlie knew what we wanted to accomplish and just did it without much fanfare or debate.” Like true artists, the engineers signed their work, along with the MOS Technology logo. “Hidden in the layout were the people’s initials,” says Winterble. “It was something I didn’t know about until it was done.” Yannes recalls attempts to market the chips to other companies. “We were seeing if anyone was interested. In the early development phase, we actually went to a couple of set-top box companies to promote these chips in their systems. You had people making boxes for cable TV companies that needed graphics capability to put menus up on the screen.” The companies turned them down. Yannes’ goal of selling the SID chip to synthesizer manufacturers also went unrealized. As he recalls, “At one point, Sequential Circuits, which was a well known synthesizer company at the time, had called with some interest in the chip. They never really did anything with it.” Efforts to market the chips ended when Tramiel gave new orders. “Once this thing started going on, I that Jack sealed it up,” says Winterble. “We’re not going to tell anybody, we’re not going to show it to anybody, and it’s going to be strictly a Commodore product.” Although Yannes hoped to see his sound chip adopted by other companies, he

took the news in stride. “You pretty much knew up front that Jack was in charge and he called the shots,” he explains. “Therefore don’t get flustered when the totally unexpected happens because that’s the nature of how he runs his business.” * * * Jack Tramiel had resisted Chuck Peddle’s efforts to establish a new line of business computers for Commodore. The previous year, he told executive Bill Robinson that Commodore was not ready to launch a new business computer in North America, instead preferring to concentrate on releasing a consumer level device. However, now that the VIC-20 had launched, Tramiel was ready to try again. He proposed a line of high-end business and personal computers to compete with IBM and Apple, which initially went under the name PET-II. Development of the new computers occurred in parallel with MOS Technology’s chip design. “We were deg the chips in tandem,” explains Yannes. “We were working hand in hand to try to come up with something that would work in these systems they were deg.” Rather than one computer, Commodore developed several different computers around the new chips. “There was going to be a whole line of computers based on that whole product,” says Charpentier. The new systems would replace the aging PET line. “We had plans for taking the same architecture, putting in additional memory through banking, and creating a whole series of computers essentially to take the place of the PET,” explains Winterble. Tramiel’s IBM killer was a black and white business machine. It used the 6545 CRTC chip for video display and the SID chip for sound. The machine would have 256 kilobytes of memory, which required a 6509 processor to access the additional memory. Marketing dubbed the machine the B256, which stood for business. The engineers included a slot for an optional Intel 8088 processor running at 5 MHz, or customers could purchase a BX256, which included the 8088. Since 1979, Tramiel had wanted an Apple killer. When Tramiel saw the color VIC-II chip, he saw the potential for an up-market personal computer for the home, similar to the TOI. The 128-kilobyte machine used both a SID chip and a

40-column VIC-II chip, making it appealing to home s. Marketing dubbed it the P128, which stood for personal. For the moment, Yannes had no input into the computer designs. “I wasn’t in the systems design group,” he says. “I was still in the chip design group so I really wasn’t that involved in deg the computers, other than whenever they needed to know about how the chips worked.” Tramiel chose Charlie Winterble to lead the new systems development. “By that time, Charlie was the head of engineering for Commodore,” recalls Yannes. “Al was the head of Systems Design and Chip Design.” Winterble and his MOS Technology engineers were inexperienced with systems design. It was an entirely new field for them. “Al and I were chip people, and we only really became systems people during this project,” says Winterble. In the past, Tramiel had surprised his engineers with the PET case, the TOI case, and the VIC-20 case. He felt a tangible case could inspire his engineers and visualize the computer as a complete product. For his Apple killer, he wanted something memorable. “We wanted to do a really exotic case design for the personal one,” recalls Winterble. “His instructions were for me to use Porsche. I him saying, ‘You can get the best design center in the world.’” Winterble traveled to the Porsche Design Center in Stuttgart, . The grandson of Professor Ferdinand Porsche started the design firm in 1972. Porsche prided itself on its “disciplined aesthetic that refines objects down to their basic function.” Over the years, Porsche has designed everything from pens to auto-bodies.[3] “I went to the Porsche design facilities over there in Stuttgart and worked with them and they designed a beautiful case that, really when you looked at it, was reminiscent of the 928 Porsche,” recalls Winterble. “I met all the Porsche people, and that was really great for me because I happened to own a 911 at the time.” The computer earned a new name because of the anticipated Porsche design. “At one point they were being referred to as the Porsche Pets,” says Yannes. The case design was exotic, but it was also impractical. “The Porsche PET was just a blue-line concept sketch that looked like a space helmet. I it had a CRT monitor that was suspended on a swivel and would have been ridiculously expensive to produce.”

Commodore dropped the concept and sought a US firm. “The packaging had been designed by a house in Boston, headed by Ira Velinsky, who was later hired by Commodore,” says Yannes. The engineers were more enthusiastic about the practical Boston designs. “The guys in Boston had done a really nice job making them. They were very curved.”[4] Although Commodore abandoned the Porsche concept, they surreptitiously continued using the Porsche name to promote the Boston designs. “People assumed that these new cases were the Porsche Pets and there were those at Commodore who tried to promote that false impression,” says Yannes.

Ira Velinsky’s case styling for the Commodore P128.

For the small team of engineers, it was an immense task to design the computers before CES. “We tried to do them simultaneously with a handful of people. It was quite a job,” recalls Winterble. “Along with each of these were the peripherals to go with them. You had to have a disk drive. You had to have a modem. You had to have all these things, and they had to be ready at time-zero.” Robert Russell began creating code for the new computers. “I was starting up with what became the P and B series,” he recalls. “We worked on BASIC 4.0 which was the extended BASIC. We called it Business BASIC.” Commodore’s Moorpark center also developed two external hard drives, which used the IEEE-488 interface on the new computers. The first drive, the CBM D9060, was capable of storing five megabytes of data, a respectable amount back in the early eighties. They also developed the CBM D9090 with 7.5 megabytes of storage. The engineers would display their new products at the January CES. * * * After completing the final design on the VIC-20, Japanese engineer Yash Terakura decided he liked his North American engineers so much he would them. The charismatic design engineer moved to Pennsylvania in late 1981 with his wife and family. “I came to the States when my kids were three and five,” he recalls. “When they told me to come to the US, I thought I was going to be in Santa Clara but actually they moved everything to Pennsylvania so I started in Pennsylvania.” At the time of his arrival, the MOS Technology engineers were struggling with the P and B computers. “Al Charpentier and Charles Winterble were heading up MOS Technology’s design and engineering side,” Terakura recalls. “Jack wanted to start doing something and we had the idea of developing bigger machines, but we didn’t have good enough parts. At that time, the dynamic memory wasn’t fast enough.” When Charpentier and Yannes proposed their chipset, they intended to use the

chips for video games. “Originally, we first started out deg those chips to build the next generation of video game [consoles] and we were also going to use them in these two computers,” says Yannes. Sometime during 1981, they lost the idea of using the chips for video games. At the time, Yash Terakura had recently arrived from Japan. After Tramiel closed the Moorpark R & D lab, Terakura became Commodore’s unofficial R & D lab. “I was involved in this or that, but at that time I was doing my own stuff deg something new,” he says. “I was pretty much the maverick inventor engineer over there and nobody said anything about what I had to do.” When Terakura heard about the next generation chips, he knew what he wanted to build. “He wanted to do a game machine,” recalls Russell. He called his machine the Ultimax. Tramiel gave Terakura total freedom to design whatever he wanted. “Jack always told me, ‘Yeah, don’t worry about it. Just do what you want to do.’ He kind of expected me to come up with new ideas rather than being told to do something.” Although Terakura worked in Pennsylvania at MOS Technology, he did not report to either Winterble or Charpentier. “Actually, I did not have any boss,” he says. “I was with the group of people but my orders always came from Jack Tramiel.” It seemed obvious for Commodore to build a game machine using the new chipset, but the small engineering staff was busy with the larger computers. “There was no intention at that point of us producing something like the Commodore 64 or even anything like the Ultimax,” says Yannes. Things quickly changed with Terakura’s arrival. Soon after he began work on his project, the machine underwent a strategic name change. “It was originally called Ultimax until somebody noted it sounded like a feminine hygiene product,” explains Yannes. Soon, engineers began calling it the Max Machine instead. Although the other Commodore engineers aided Terakura with the SID and VICII chips, Terakura drove the project himself. “He was coming to us and asking us what to do to try and get the Max thing done,” says Russell. With the engineers still having problems obtaining fast enough DRAM speeds, Terakura decided to

use SRAM. The decision meant Commodore would have to create a special version of the VIC-II chip, designated the 6566. Bob Yannes fondly re his time spent with Terakura. “He was a great guy. Great English. He was always very humorous and always poking fun at people. He was a great buddy of mine. I really loved Yash.” During development, the engineers discussed how much memory to include in the Max Machine and could not agree on the proper amount. “When we were developing the Max architecture, we only put 4K of RAM in it to keep costs down,” says Yannes. However, Robert Russell knew it would not be enough. “The software group wanted 8K of RAM so that they could produce a full bitmapped display.” Tramiel learned of the disagreement and decided to help his engineers resolve the argument. “It was a big fight until Jack stepped in,” recalls Yannes. Acting like a modern day King Solomon, Tramiel listened to their positions and gave his judgment. “He decreed that we should put 6K of RAM in as a compromise and then left.” The engineers stood in shocked silence, since the solution helped no one. “6K cost more than 4K and it still wasn’t enough to do a bit map display,” says Yannes. “We all just stared at each other.” * * * MOS Technology still had their usual customer, Atari, clamoring for ROM chips during their design of the new computers. While Yannes waited for the VIC-II and SID chips to come off the fabrication line, he worked on a special project. Atari began to realize its four-kilobyte games were losing their edge. Commodore came through with a technological breakthrough that gave their games new life. “The Atari VCS could only address 4K of memory in a game ROM, and they wanted to be able to address more than that,” recalls Yannes. “I actually ended up deg a custom chip for Atari. We custom designed them an 8K bank switch ROM. It looked like a 4K ROM [to the VCS] but it would trigger into another bank when you accessed a particular address. We jumped through hoops to get that done on time for their Christmas season.” Atari used the new ROM chips in their hit game for the 1981 season, Asteroids. Atari also considered using the new ROM chip in their port of Pac-Man, but

ended up using the standard four-kilobyte chip for the game in order to save money on the project. The game did not fare well. * * * As 1981 progressed, Commodore’s share price rose to the equivalent of 138 ¼ (not including a 3-for-1 split). It had only been 5½ in early 1979. The attitudes of Commodore employees towards the VIC-20 soon changed from scorn to respect, partly because the VIC-20 was now covering a significant portion of everyone’s paychecks. “When the chips were working, it was clear that the VIC-20 was becoming a successful product,” says Charpentier. Low-end computers began to look more promising. The success of the VIC-20 threatened to kill innovation within Commodore. “The VIC-20 is now being sold. It’s a pretty good success story making lots of money, and it’s got a hugely established software base of games and applications,” says Charles Winterble. “So now all of a sudden we have an establishment for the VIC-20.” Winterble wanted to develop a VIC-20 sequel with 40-characters per line. They provisionally dubbed this idea the VIC-40. Unfortunately, some executives so adored the VIC-20 they were hostile to anything else. “You had to be careful where you mentioned the VIC-40 because somebody would shoot you,” he says. In the middle of launching the VIC-20 in the US, the sales managers wanted nothing else. “There was a real controversy about the idea of coming out with a product that would displace the VIC-20,” says Winterble. “Talk about crummy strategy! We’re hearing from the marketing guys that we can’t have a VIC-40. ‘Don’t even talk about it because you’ll kill our business.’” Although some Kit Spencer’s own marketing people were protective of the VIC20, he agreed with Winterble. He recalls, “When we’d come along with a new product, somebody would say, ‘Well we can’t launch that at that price, we’re going to kill our own product.’” Spencer replied, “In which case we better do it because we’re in control of it. If we can do it, someone else will do it soon. If we don’t do it, we’ll lose control of the marketplace. We’ll introduce something and we’ll be our own competitors.” The philosophy of internal competition originated with Tramiel. “That was very

much our marketing philosophy at Commodore,” says Spencer. “We often used to say it at meetings, and it’s something I learned from Jack. I was one of those that would often repeat the same mantra.” Commodore marketing was intent on duplicating the longevity of the Atari 2600, which sold well for over half a decade without improvements. “It was typical shortsighted, sales oriented [strategy],” says Winterble. The marketing people had little input into the actual design of computers, however. “At Commodore, that was not part of the process at all,” says Neil Harris. “The spec of the product was purely engineering driven. The only person outside of engineering that had input into that was Jack Tramiel himself.” Meanwhile, the engineers continued development of the high-end computers. “We were supposed to be doing the P series and B series,” says Russell. However, the engineer’s faith in the new line of computers was beginning to falter. “Bob Yannes and I went to Al, and I think we all approached Charlie and said, “We really don’t like this P series machine. We want to do just the simplest thing possible.” He added, “We can even make it what Yash wants to do with the Max Machine.” The engineers discussed their concept in an impromptu meeting with Tramiel. “Jack would hold court, if you will, at various times,” explains Charpentier. “He was the emperor. Sometimes it was the marketing people, sometimes it was the engineering people. He would sift the information and make a decision.” The engineers proposed a true sequel to the VIC-20. According to Charpentier, “We basically had a meeting with Jack and said, ‘Gee, why don’t we put this into a VIC-20 case and do a VIC-40.’ That was the original name we had used.” The biggest decision with the VIC-40 was the memory size. “When you are laying out the groundwork, you know that you want to be able to have [lots of] RAM, but the fact is RAM was still very expensive,” explains Charpentier. It was obvious to the engineers that the machine would need more memory than the VIC-20. “The original game concept was to put 16K into it.” Yannes led the project, which also went by the alternate development name of the Commodore 16. “Bob Yannes was a young engineer. He was a junior engineer,” says Terakura. “He was involved a lot with the SID and VIC both, and he did the smaller newer machine, the Commodore 16.”

Winterble assumed Tramiel would want to show the new computer for the June CES, only seven months away. Normally, engineers require at least a year to design a computer, but Winterble knew if he put his engineers on high alert, they might get it done. Instead, Tramiel wanted the new computer sooner. “Jack decided he wanted to show a machine with 64K of RAM at the January 1982 CES,” says Yannes. It was already November, and CES was less than two months away. “I don’t even know that he intended to put it into production.” Tramiel expected miracles from his engineers. “Every time before a show, he would come to someone and say, ‘I want something for the show.’ You had to jump through hoops to have something,” says Yannes. Faced with this goal, the engineers could either throw their hands up in disgust or do their best to meet his seemingly unrealistic target. The aggressive deadline took all his engineers by surprise. “We didn’t start working on that until Thanksgiving of ’81,” recalls Yannes. Winterble assigned two junior engineers and a technician to complete the project. “Charlie approved us to go ahead and pursue that with our own little team,” says Russell. “It was Bob Yannes, me, and Dave Ziembicki the technician who really went off and did the Commodore 64. Luckily we had a guy like Charlie Winterble who let us go off and do that when we were supposed to be making the P and B stuff work.” With less than two months to build a complete computer system, the engineers rarely left the MOS Technology building. “In the middle of the building lab, we took over one corner of the room and worked 20 hours a day, 7 days a week to get the prototypes running,” says Russell Deg a full computer system was a new challenge for Yannes. “I was still in the chip group so I wasn’t really supposed to be working on systems,” he says. “The only reason I ended up doing the C64 was because I was the only one who knew enough about the chips and how to put them together in a timely fashion.” With such a tight schedule, Yannes and Russell began laying out the architecture of the computer. “Bob [Yannes] and I sat down and came up with the hardware architecture,” recalls Russell.

Yannes was an assiduous engineer by nature. For two short days, he worked in his office and the drafting area to design the architecture for the VIC-40. “It was a pretty easy architecture,” he says. “I just designed the most minimal system I could with the fewest number of components. There’s not a whole lot of stuff in there. There’s the VIC chip, the SID chip, and there’s 64K of DRAM.” Almost none of the design came from the VIC-20. Only the serial port, cassette port, and port remained the same. It also used the same joystick connecter, except there were two. Internally, the computer was unique. “There were very few chips that were used in the C64 that had ever been used before,” says Yannes. Rather than use the same bulky cartridge system of the VIC-20, Yannes decided to borrow technology from the Max Machine. “Since the Max Machine was already in progress, I decided to make one of the C64 memory configurations match the Max so that it would be able to use Max game cartridges,” he explains. “When you plug the game cartridge in, it would automatically collapse the memory map of the Commodore 64 to look like the Max Machine.” The VIC-40 was essentially a computer with a game console built into memory. The engineers wondered how they could create such a complex memory layout before CES. They found their salvation in the Programmable Logic Array chip (PLA). According to Russell, “I finding that chip and saying, ‘Oh, that will do exactly what we want!’” The PLA chip acted like glue to hold the different parts of the system together. Yannes could simply insert the PLA chip and program it later. “I didn’t have time to design all the logic before they laid the PC board out, so I just took a PLA and named the signals I needed and told them to lay that out,” recalls Yannes. “While they were laying it out, I could figure out the coding for the PLA. That got us to the show.” When engineers need to build a circuit quickly, they use thin wires and a special wire wrap tool to connect the chips together. However, Yannes believed it would be inadequate. “You really couldn’t do a wire wrap with dynamic RAM because the timing was too tight,” he says. Instead, the engineers would fabricate a printed circuit board. To allow time to develop software, Yannes left nothing to the last minute. “We

had to have a working circuit board practically a month before that to get the software working because we wanted to show it running with BASIC,” says Yannes. “It was going to be perfect,” says Russell. “We made it simple and clean. We cut boards and everything in one month.” [1] In contrast, the chipset built four years earlier for the Atari 2600 VCS used 6micrometer technology. [2] In September 1995, as part of their 20th anniversary edition, Byte magazine named the SID as one of the top-20 most important chips ever. [3] The Porsche Design Center has continued to design computer cases, but Commodore was the first. [4] Velinsky later won the German iF Product Design award in 1983 for his cases.

CHAPTER 26

The Unveiling 1981-1982

Since Robert Russell’s move to the east coast, he and Bob Yannes developed a close relationship. “We worked hand in hand on the C64. We spent all that time in the lab,” says Russell. “We were best friends.” Yannes also looks back fondly on the friendship. “We hung out a lot together back then,” he recalls. “He was transplanted from Iowa to California to Pennsylvania so he was probably [alone]; at least I was always in my environment.” According to Yannes, in between work, the two engineers took in films from the early eighties, such as Raiders of the Lost Ark and Superman II. “We were going to movies together and all kinds of stuff,” he says. “We were both single geeks without any social life.” The VIC-40 schedule destroyed two major holidays for the friends. “It had us jumping through hoops to try to get some totally new thing that we hadn’t even intended to work on and try to get it done between Thanksgiving and early January with Christmas break in the middle,” says Yannes. “I checking the PC board layout over Thanksgiving weekend. That’s how tight the timeframe was on that.” Although the new assignment would cause Yannes to miss the holiday season completely, he did not think of Jack Tramiel as a Scrooge. “It didn’t matter,” he says. “I was living at home and I wasn’t married or anything. I thought this was great.”

Throughout the development of the project, the engineers kept the project a secret from others at Commodore. “We didn’t talk to marketing,” recalls Charles Winterble. “We bounced ideas off Jack, but he didn’t really care about the specifics of it.” Not even John Feagans knew about the project, even though the computer used his kernel code. “He didn’t do code work on the C64 at all because he never even knew it existed until it came out,” says Russell. “It’s his architecture and it’s me building on what I did on the VIC-20.” Out of all the engineers at Commodore, Yannes’ philosophy of low-cost computers was the closest to Tramiel’s. “I tried to design the cheapest possible thing I could because that was just my nature,” he explains. “I didn’t like expensive things, I didn’t have very much money, and I didn’t see any reason why this stuff needed to be expensive.” Although the P and B computers had plastic case designs, no one tried to design a new case for the VIC-40. “If you’ve ever wondered why the C64 has the same case as the VIC-20, it’s because we didn’t have any time to tool anything up,” says Yannes. “We just put it in a VIC-20 case and spray painted it. Everything about the Commodore 64 is the way it is because of just an unbelievably tight time constraint on the product.” In retrospect, Charles Winterble believes the decision to use the VIC-20 case ended up costing time. “One of the design criteria which we chose, which was a mistake, was we said, ‘Gee, let’s put this all into VIC-20 plastic.’ That was wrong because we didn’t have enough room,” he explains. “We spent so much time and resources trying to make the motherboard fit inside that stupid VIC-20 case.” The engineers also had to choose colors for the text and background. “Blue and white is what we used, because that gave you the best color contrast, other than black and white, which was too boring,” says Al Charpentier. “We wanted the people to see those colors.” Problems with the chips remained well into December. According to Charpentier, “We literally had gotten the video chips a couple of weeks or a weekend before the [CES] show.” On the week of December 9, SID chips started coming off the fabrication line at

MOS Technology. The sound chip used 7-micrometer technology, compared to 5-micrometer technology for the VIC-II chip. This ensured each silicon wafer produced a high yield, which reduced costs. The technicians produced low quantities initially because it was certain the first batch would require revisions before they went into full production. Yannes nervously waited for his first chips. He knew what the SID would look like, but he really did not know how it would sound. “The SID chip came out pretty well the first time,” he says. “It made sound. Everything we needed for the show was working after the second .” One of his disappointments was the poor signal-to-noise ratio, which produced an audible hum when the SID was quiet.[1] “So many things in the SID chip don’t work as well as they could have because there just wasn’t time to make them right,” says Yannes. Commodore was never successful producing reliable RAM, making Tramiel’s vertical integration strategy incomplete. The inherent cost of 16 kilobytes would make the machine too costly to be a true VIC-20 sequel. With such a small amount of memory, the engineers knew the computer would not directly compete with the new 48-kilobyte Apple II Plus. However, when 64-kilobyte chips began to appear on the market, Commodore made an abrupt change with the VIC-40 design. Surprisingly, Tramiel insisted on more memory. “Jack felt we should design it with 64K,” says Charpentier. Tramiel predicted the price of memory would fall by the time the computer was ready for mass production. Yannes felt Tramiel’s decision was correct. “It was really a brilliant move because 64K RAMs had just come out and it meant that you could introduce a machine that had 64K of memory when everyone else had at most 48K,” he says. Soon, the design was complete. By late December, the team had multiple VIC40s. “We had built two or three at that point in time that were running pretty good,” says Russell. Commodore now had a computer they could show to the world at CES. Yannes was understandably proud. For CES, the computer required demonstration software to show off the sound and graphics. Winterble pulled in many of the VIC Commandos to expedite the

software development. “I was involved in the prototypes from the time it was called the VIC-40,” says Andy Finkel. “We were documenting it, figuring out how things worked, testing it. Somewhere I still have my prototype in a VIC-20 case.” The programmers were busy with VIC-20 games, but excited to work with a new piece of technology. “The C64 was most everything we wanted the VIC-20 to be,” says Finkel. “Obviously we liked it. It had the SID chip, to get some real sound out of it, and it had 40 columns. It had enough memory to do a whole lot. So we were very pleased.” Michael Tomczyk chose only a few elite developers to work on VIC-40 software development. He chose his best machine language programmers—Andy Finkel, Eric Cotton, and Bill Hindorff—for the project. Finkel became the software manager for the new machine. “When I got my first prototype, I was still in Moore Road, and Tomczyk was still deciding who was going to be the software product manager of the C64 and who was going to continue being the software product manager of the VIC-20,” says Finkel. “Some of us had to continue to do VIC-20 stuff and others of us got to move onto the VIC-40.” Robert Russell began developing demonstrations of the VIC-40 computer. “All that stuff was originally BASIC with just a million poke statements,” says Russell. “You didn’t write assembly language for those early demos, you poked in assembly language. We’re talking some ugly old BASIC code for the original demos.” A Commodore engineer named Fred Bowen helped create a playful demo of a small man who walked out onto the screen and showed off all the features of the VIC-II and SID chips. “Freddie Bowen wrote a lot of the stuff,” says Winterble. “We had a sprite guy with some music playing,” says Russell. “We had the transparency and stuff like that to show the sprites. … I can doing the coding with Yannes trying to get stuff to work.” John Feagans flew in from Moorpark to assist with the C64 programming. “I flew commercial back to Pennsylvania where I developed the SID music program and other demos for the CES intro,” he says. Feagans followed the previous strategy of converting PET software to the new computer. “John Feagans had done some music stuff to demonstrate some of the

sound,” says Yannes. “They had different music programs for various PETs along the way. I think he just converted them to play whatever library of songs they had.” “I sitting in the lab with a prototype and Yannes is there and it’s the day before Christmas,” recalls Russell. “We couldn’t get good radio reception in there, so we were creating music with the SID to listen to.” Surprisingly, Russell had a disk drive functioning with the VIC-40 in time for CES to load their demos. “We loaded them in from a 1540 drive,” he says. It was a remarkable achievement to have working computers for CES, especially considering the engineers had not even started the VIC-II and SID chips until April of 1981. “Nine months later we had enough working prototypes and we were able to go to the show,” says Winterble. “So much of the Commodore 64 was just the way it was because of the constraints of time, and I think it actually made it a better product to be honest,” says Yannes. “We didn’t have time to fiddle with things and change it around too much.” The final verdict on the computer would come at CES. * * * The Las Vegas CES began on the first weekend of January. “When you worked for Commodore, you always had to have something for the Winter CES,” says Yannes. The night before departing, the engineers prepared to transport their delicate prototypes. “We were putting the things together and packing them up to ship them,” recalls Yannes. Tramiel allowed the engineers to fly aboard the PET Jet to the show. “We loaded up the plane in Philadelphia with C64 prototypes and flew to Teterboro [airport in New Jersey] where we picked up Irving Gould and [Commodore president] Jim Finke,” says Feagans. “I might add that I always flew in the jump seats on the plane.” Everyone aboard the jet realized there was something unusual with the flight when they stopped midway. “Unknown to us, we hit an extreme jet stream and only made it as far as Lincoln Nebraska before needing a refueling stop,” says

Feagans. “The pilot had maxed out his Commodore credit card. Same with Bernie, the controller. Fortunately, Jack was there or I would have been next in line to max my credit card for the fuel.” While on the ground, Feagans found out the weather was going bad in California, but arriving late was not an option. “We were headed into a big Pacific storm,” says Feagans. “I called my wife in the Santa Cruz Mountains. She said it was raining, then later she said the power was out, next there was no water, and last the phone was dead.” Fortunately, the pilot persevered and the prototypes made it. Once in Las Vegas, the engineers hauled their prototypes to Tramiel’s suite, where he would determine whether the devices were worth showing to the crowds. The tired engineers took up a corner and readied their prototypes. “We were pretty burned out just from getting this stuff ready for the show,” says Yannes. While setting up, Robert Russell overheard Tramiel and his inner circle discussing their plans. “I was in Jack’s hotel suite preparing a demonstration,” he recalls. “There was some strange stuff I saw. He would be going over things with the European and Asian cronies about them getting things in and out of countries. I don’t know how legal a lot of it was. I didn’t want to know some of that stuff because it sounded like you might end up in a barrel someplace.” It was clear to Russell that Tramiel and his inner circle were prepared to do almost anything to help Commodore. “They were so bad,” he says, laughing so hard he can barely breathe. “There were times they were pissed off at certain people. Customs and duties were a lot more complicated in those days than they are these days. Every country in Europe was different. They made certain threats because they were having problems with some of the European countries as far as how they were handling product and dealing with all the issues.” Although the young engineers worked hard on the VIC-40, there was no plan to display the computer publicly. “It wasn’t even planned to be on the floor or anything,” says Russell. No one except for Tramiel and his small group of engineers were aware of the project. “We actually took it and showed it to him and some of his cronies in his suite in the hotel. Those guys didn’t know about it at all until we showed it.”

The computer still had obvious chip problems, but Tramiel would determine if it was ready to show. “I these heat problems of video sparkling back then on those early chips,” recalls Finkel. “I think it might have been, ‘Is it ready enough to show?’ And he made that decision there.” The nervous engineers displayed the result of almost two months of compressed labor. “We told him what it was, how simple it was, and what it could cost,” recalls Russell. “He said, ‘Put it on the floor.’” As Winterble predicted, marketing was not happy to learn of the project so late in development. “When these guys found out about it, and found out that they were not involved in it, then right away you can imagine: it hit the fan,” says Winterble. “It was turmoil.” With no advanced warning, Kit Spencer had to work non-stop to prepare print material for the prototype. “The marketing guys ended up claiming it was going to do everything under the sun on the charts,” says Russell. For the most part, the engineers dictated the content of the ments. “We told the marketing guys what to write down and made up signs.” The VIC-40 name lasted through most of the production design. However, Spencer wanted to change the name to match the other computers in the Commodore lineup. They already had the P128, which was a personal computer with 128 kilobytes. They also had the B256, which was a business computer with 256 kilobytes. Now they had a consumer computer with 64 kilobytes, so naturally it became the C64. Most people just called it the Commodore 64. “I decided to name our new product the Commodore 64 and not give it a specific name like the VIC and PET,” explains Spencer. “This was not only because the name flowed quite nicely, but because it’s biggest feature over the Apple was its 64K memory, which was much larger than its competitors had at the time and I wanted to emphasize that right up front in the name.” When the CES doors opened the next day, the main attraction was the VIC-20. “This was the really big re-launch of the VIC-20 at CES,” explains Spencer. Bill Wade’s newly recruited sales team manned the offices of the booth, g up new distribution deals by the hour. Spencer brought VCR tapes of the partially edited William Shatner commercials and played them in a non-stop loop. “I had the commercials running all around

our booth together with the new look of the VIC packaging, software and other marketing materials.” Kornha & Calene also produced life sized cardboard cutouts for Commodore retailers to use for in-store promotions. In these, Shatner is wearing a blue pinstripe business suit while holding a PET computer. A few of these sat on the show floor around the booth. Shatner was an enthusiastic spokesman and even attended the show.[2] “I met him when he came by some shows,” says Russell. “Shatner was certainly a hoot.” The image of the Enterprise captain caused a lot of interest around the Commodore booth. “It was certainly a highlight at that point that got us a lot of attention,” says Neil Harris.

William Shatner advertising the VIC-20.

Commodore unveiled the 1600 VICModem at the show. Retailers were enthusiastic about their potential. Later, Byte magazine wrote a three-page product description of the modem in their March 1982 issue. In it, Michael Tomczyk said, “We think this modem may sell as many VICs as the game cartridges.” Compute!’s Gazette also gave the VIC Modem a favorable review, noting it was the least expensive method for telecommunications on the market. Commodore would release the VICModem in March 1982 for a price of $109.95. The product information, written by Tomczyk, fancifully proclaimed, “Science fiction is now reality. Science fiction writers used to speculate that one day we’d be able to use ‘electronic libraries’ in our home or office...Your Commodore computer gives you tomorrow’s world of electronic communications today.....You’ve stepped into the next era of technology, the Computer Information Age!” Compared to most prototype demonstrations, the Commodore 64 was remarkably complete. “Almost all the stuff that was put together quickly for the show was not anything that was real,” says Yannes. “They were smoke and mirrors. Part of it was just to get some press at the show and to gauge people’s reaction to things. The Commodore 64 was probably one of the most real things that showed at a show.” Yannes was too inexperienced, too quiet, and too much of an engineer to demonstrate his prototype. “The only reason I was there was in case it broke and needed to be fixed,” he explains. “I was pretty much off to the side, but I think there were a few times when I was called upon to explain some of the features and capabilities, particularly when it came to the sound.” Yannes also spied on potential competitors. “One of the things I was supposed to do at that show was to case the competition and check what was going on at the other places. Charlie or Al asked me to do it,” he recalls. There was not much competition at the 1982 CES. Commodore’s main rival, Atari, was still showing their Atari 400 and 800 computers. Mattel introduced the Aquarius computer, and a company called Spectravideo introduced the SV 318. Both of these machines were similar to the VIC-20 in specifications, but both were doomed

largely because of their calculator-style keyboards. The competition was weak. “There wasn’t really anything out there,” says Yannes. “It really was a coup because Apple and Atari and everybody else were just pretty much showing what they already had with a few little additions here and there.” The Commodore 64 was also able to demonstrate a full line of peripherals, including the 1540 disk drive. “It used the VIC-20 disk drive and the VIC-20 printer and all the peripherals that had been designed for the VIC-20,” says Yannes. “We didn’t have time to design new peripherals.” Bill Seiler attended CES and saw that his former coworkers at MOS Technology had taken the dream of the TOI and improved upon it. “In a lot of ways it was better than the TOI,” he says. “Those guys had integrated a lot of the functions into the chips. It had a pretty nice audio generator and it did sprites and other fundamental graphics things.” The minimalism of the machine impressed the former employee. “It didn’t have hardly anything around it,” says Seiler. “Just the memory and a couple of MOS chips and they were done. There was nothing to it. When we first saw it I said, ‘Wow these guys are doing good work. This is good. This is really good.’” Although he was a video enthusiast, it was the SID chip that amazed Seiler the most. “Yannes went crazy and did all kinds of nice things in the audio stuff and sound generation but they couldn’t get it in the chip so he had to strip it way back,” he recalls. “His thing was way over-built. It was kind of like a predecessor of the Sound Blaster.” It was a satisfying closure to the TOI project, which failed due to memory speed. “The 64 was nice because the chipset hooked right up to the DRAM chips,” says Seiler. “There was no static RAM anymore, it was DRAM.” The engineers also brought prototypes of the P and B series computers, as well as the Max Machine. Commodore hid these in a small office within the booth and showed them to journalists, software developers, dealers and sometimes even trusted acquaintances from competing companies. “We had a couple of backroom things going on for special customers,” says Winterble. Terakura worked the backroom, demonstrating the Max Machine. “I was one of

the engineers who could go anywhere I wanted, including all the electronics shows,” he says. Although marketing created the Commodore 64 presentation at the last moment, it was an impressive demonstration. “It was a rather fancy booth,” recalls Winterble. “We had a bunch of stations showing different aspects and different parts with a skilled person at each space. One guy was showing a game demo, one guy was showing something else. We had them scattered all around this booth.” A prototype demonstrated the capabilities of the SID chip using John Feagans music program. The beautifully strange music filled the air, acting as a siren call for technophiles. “They were lined up,” says Winterble. Mostly, the presenters just let Fred Bowen’s demonstration program speak for itself. “It was really impressive for only having the machine in his hands for a short period of time,” says Winterble. “He wrote a ‘Welcome to Commodore 64,’ that came out with a big splash. Then this little man walked across the screen, turned around, and started doing things. It was a great little demo using the sprites.” As Robert Russell recalls, the burgeoning press helped fuel interest. “There was starting to be a whole press industry around personal computing,” he says. As word got out about the new computer, the lines to enter the booth grew longer. “It was a huge sensation. Everybody and their brother were stopping by.” Yannes was also thrilled at the positive reception of his computer. “The C64 just kind of blew everybody out of the water because it came out of nowhere,” he says. “There was no expectation of it, it was very reasonably priced, and it had 64K of RAM which was a magic number at that point in time because nobody else had 64K of RAM.” The computer even impressed Chuck Peddle, who never missed a CES show. “It was a great game machine; not because of the 6502 and not because of the memory that was in it. It was a great game machine because of the work Charpentier did,” he says. Perhaps the most impressive aspect of the computer was the proposed price. Before the show, Tramiel decided on a retail price of $595. Competitors reacted to the announcement with skepticism and shock. “We got their attention,” says

Winterble. “The guys from Atari came by to look at it and said, ‘They can’t do that. It’s impossible for the price.’” From the reactions, Winterble felt the new computer might even outsell the VIC20. “When we left CES, we knew we had a fantastic product.” Press was good but not as prominent as Commodore received in earlier years. David Thornburg of Compute! magazine mentioned the Commodore 64 and Max Machine, reporting, “For sheer impact, Commodore stole the show with the announcement of two new color computers!” However, Byte magazine reported nothing on the Commodore 64. By now, Commodore had a reputation of announcing products such as the ColorPET and TOI and not releasing them, so Byte was wary. The magazine seemed far more interested in the debut of the low cost VIC modem. The meaning of the Commodore 64 debut was obvious to Tramiel: he finally had his Apple II killer. Now all he had to do was deliver the crushing blow. He wanted to get the machine into production quickly. * * * The success of the VIC-20 surprised even Tramiel, who did not anticipate the growth it would cause when he created the employee stock sharing plan. “When they designed that plan, they didn’t expect Commodore was going to go from a $50-million company to a $500-million company in a couple of months,” says Russell. “The stock shot through the roof.” Employees in the stock sharing plan were ecstatic. “The stock plan made so much money that every person would have received double their salary as a bonus,” recalls Russell. Unfortunately, Tramiel did not offer the plan to the other engineers who were crucial to the VIC-20 and C64. He had been so impressed with MOS Technology, whom he credited with starting the VIC-20, that he only offered the plan to them and Tomczyk’s VIC-20 group, including Neil Harris and Andy Finkel. Until the migration to the east coast, none of the west coast engineers knew of the plan. “When I moved out there, everyone was saying, ‘We’re going to get a

mint because the VIC-20 has driven up our stock price so much.’ We were like, ‘What the hell are you talking about? How come we didn’t get offered that?’ The guys from the west coast were hugely pissed off.” The bonus plan seemed unjust to Russell, who had designed most of the VIC-20 with Bill Seiler. “That was one of the issues that hurt,” he says. “They did double their salary, but it was all because of the VIC-20 success. You know, I thought I deserved a hell of a lot more.” Tramiel routinely gave 100% salary to his trusted inner family, but 200% was too much for him. When it became apparent he underestimated the bonus plan, he began to back out of the agreement. “I don’t think Jack could deal with giving away that much money,” says Yannes. “They started playing some shenanigans about the stock plan, ‘No, no, no. You misunderstood us. This is what we really meant.’” Russell believes Tramiel had no option. “The inequality between them and the rest of [Commodore] would have torn the whole company apart,” he says. “Those guys maybe didn’t realize that, but it would have.” Tramiel called a meeting at MOS Technology to explain the revised bonus plan. “MOS had a little cafeteria at the back alongside the production line,” says Russell. “I going to the meeting where it was rescinded. They had all the guys down in the cafeteria room at MOS.” Tramiel and MOS Technology president Jerone Guinn delivered their version of the bonus plan. “These guys had posters and charts on the wall that said what they were going to get,” recalls Russell. “There were lots of unhappy people there.” Tramiel had made a promise to his engineers and it seemed inconceivable to take it back in the midst of wild success. To keep his promise to the engineers, he could have retroactively offered the plan to the few remaining west coast engineers to satisfy everyone. Instead, he chose to hold back most of the promised bonus. The VIC Commandos were the only group who happily accepted the revised plan. “I was happy to get anything at all because we never really counted on it,” says Finkel. “Other people may have counted on it more and then been upset when it was cut in half.”

After delivering a convoluted explanation about the revised plan, Guinn asked his employees to sign a new agreement. According to Yannes, “They had a complicated explanation of what they ‘really meant.’ They handed out these forms where they told you how much money you would get, which was nothing at all like twice your salary,” says Yannes. “In fact, it was a fraction of your salary.” Tramiel applied pressure to persuade his employees. “You had to sign this form that you understood and agreed to it, and if you didn’t sign the form then you didn’t get anything,” says Russell. Under duress, the engineers felt they had no option but to sign. Yannes and his fellow engineers at MOS Technology felt outrage. “Every MOS employee got cheated out of what they had been promised,” says Yannes. “It was still a sizable amount, but when you’ve been promised a bonus of twice your salary and receive less than half your salary, it led to hard feelings.” Yannes was surprisingly accepting of the new agreement. “He wasn’t one to make a big fuss,” says Russell. “He was upset, but he wasn’t an outspoken guy.” “Getting into a lawsuit against your employer is a bad thing to do for your career,” explains Yannes. “I just signed the form and said, ‘Fine, I’ll take whatever you want to give me.’” After the meeting, morale at MOS Technology was at a new low. “There were a lot of people that were really bent out of shape and there were people talking [about a] class action suit,” recalls Yannes. “I some people being very upset,” says Finkel. “My particular group wasn’t all that upset.” Yannes had some consolation. “I was fortunate in that Jack actually gave me a separate bonus for the VIC-20,” he says, referring to the $5,000 gift. However, the meeting left him weary of Commodore. “It was pretty clear that unless you were among Jack’s elite circle of friends and trusted people, you weren’t going to make much money at Commodore, no matter how hard you worked and no matter what you contributed to the success of the company.” The revised agreement failed to satisfy west coast engineers like Robert Russell. “They did do something different, but those guys got a bonus nobody else got,”

he laments. “I always got screwed on Commodore bonuses. Believe me, it was not the first or last time that it happened.” [1] Software programmers dealt with the hum by stopping the oscillator when the note was completed, making it difficult for the listener to detect the noise. [2] In 2003, Commodore/Shatner fan Robert Bernardo asked the movie star if he ed Commodore. He replied, “Are they still around?”

CHAPTER 27

Commodore 64 1982

After CES, Jack Tramiel wanted the Commodore 64 sent to the assembly lines. Unfortunately, the prototypes could not be mass-produced. “We weren’t ready to go into production,” says Charles Winterble. “After the show, we had a great deal of work to do to really turn it into a production product.” Although the MOS Technology engineers had no experience in production engineering, it would be up to them to design a production model in record time. “We wanted to be in production in three months,” says Robert Russell. To launch the product, the marketing team also had plenty of work to do, since they were unaware of the project until CES. “It wasn’t even handed to marketing until after that show,” says Russell. Winterble suspected there would be some resentment within Commodore due to the secrecy of the project. “Commodore was a company that had friction a lot of times,” he explains. “One of the big friction areas was between marketing and engineering. There was a long standing... animosity is probably too strong of a word, but marketing and engineering at Commodore did not get along. And from our point of view, why did we care?” Michael Tomczyk, like most people in Commodore marketing, still clung to the success of the VIC-20. “Tomczyk was fighting the C64,” says Chuck Peddle. “I’ve written him a note two or three times saying, ‘I don’t know why you did that at the time. The VIC-20 did exactly what it was supposed to do, which is to pave the way for the C64.’”

Part of the animosity was probably due to the favored position of the engineers in Tramiel’s eyes, and the powerlessness of marketing to influence engineering. “The engineering side had an easier life because everybody is an expert in marketing, but if you’re going to be an expert in engineering you have to know the technology,” says Winterble. Kit Spencer wanted the engineers to modify the C64. “When we met with the marketing people and went through it all, there were a number of issues that they wanted to change right away,” says Winterble. Spencer raised some legitimate issues with the Commodore 64. First, the operating system was horribly antiquated. It was no different from the PET released almost five years earlier. Second, the computer had no backward compatibility with VIC-20 software. The VIC-20 was a huge market and Spencer did not want to lose the established software base. Third, the BASIC language had no extensions for sprites, sounds, or graphics. This forced programmers to use difficult POKE statements. Winterble opposed Spencer on every change. “One of my jobs was to fight the battles,” says Winterble. “I had relatively strong opinions about things and I was willing to fight.” Spencer recognized the importance of DOS to the new IBM PC computers. He wanted Commodore computers to have the same high-level operating system, so he tried to push the engineers into adopting the /M operating system. /M was the standard operating system at the time, similar to Windows years later. “Microsoft has proved that, through the years with an operating system, people get used to it and it’s the standard,” says Spencer. “They just kept building up and building up on it.” A key feature of /M was the ability to create directories. Without them, disks with more than one program often contained a disorganized jumble of files. One of Spencer’s main objectives with /M was to allow Commodore to claim an already large software base for the C64. “We needed to silence the /M people,” he says. “Sometimes you’ve got to do things to kill the opposition’s criticisms.” Winterble disagreed with Spencer over /M. “One of the battles that came up

was /M,” says Winterble. “They were pushing for, ‘We’ve got to have a computer that will run /M.’” Winterble felt the /M operating system was not worth including. As their discussions progressed, Spencer was happy to allow /M as a separate product, rather than built right into the system ROMs. “A /M cartridge was probably needed as an add-on,” he says. “I didn’t think we could sell it in volume, but the absence would hurt us.” The engineers began work on a /M cartridge using the Z80 microprocessor. “We designed it right from the beginning [of the production design] to take a Z80 module,” says Russell. Since the engineers believed the computer would use /M, Russell made no effort to improve the native operating system. Memory limitations also prevented the Commodore 64 from using the new BASIC 4.0, intended for the P and B computers. “Why did we have [BASIC 2.0]? That’s how much room the ROM could fit,” says Russell. Time was the overriding factor for Russell and the engineers. “We were trying to get it out the door,” he says. “We could see the computers coming that had better BASIC and real operating systems but we wanted to do the simplest, most straight forward, cheapest and quickest system possible.” Bob Yannes acknowledges BASIC 2.0 was antiquated by 1982. “It was a very primitive BASIC,” he recalls. “It had no graphics expansions or sound extensions or anything.” To delete a file, s had to type:

OPEN 15,8,15:PRINT#15,“S0:filename”:CLOSE15

In contrast, BASIC 4.0 s merely typed:

SCRATCH filename

Another quirk of the Commodore BASIC operating system was the memory designation. Commodore engineers chose to identify the size of files and disk space using blocks as opposed to bytes or kilobytes. Few s understood what a block was. They knew their computer had 64 kilobytes, but it was hard to say how much memory each program occupied on disk. Spencer appealed to Tramiel, but he was unive. “Jack didn’t care what kind of software his machines had,” says Yannes. “He was putting the same BASIC in every machine, even though it was obsolete and had very few features.” Tramiel’s views were rooted in mechanical devices like typewriters and calculators, so he placed an emphasis on hardware. He had a blind spot when it came to the invisible software in the ROM chips. “Commodore was an extension of Jack Tramiel, and to him software wasn’t tangible,” says Al Charpentier. “You couldn’t hold it, feel it, or touch it, so it wasn’t worth spending money on.” Spencer also lobbied for backward compatibility with VIC-20 software. Backward compatibility was on the drawing board ever since John Feagans implemented the kernel. Now that Commodore was deg their next computer, they had the option of following Feagans' plan and making their next computer backward compatible. However, Russell had an insurmountable obstacle preventing him from achieving backward compatibility. The VIC-II video chip was not capable of displaying the same resolutions as the original VIC-I chip. As a result, it could not display 22 characters per line. Without this native capability, there was no point in even attempting backward compatibility. Robert Russell re discussing backward compatibility in the VIC-II chip. “We wanted a compatibility mode, but that would have taken up too much of the chip,” he explains. “We would rather have sprites than compatibility mode.” Winterble also took part in the discussions. “Al originally had compatibility built into the chip, but realistically you are talking two totally different machines,” he says. “When you are that price sensitive, every penny counts. You really don’t

want to put a lot of hardware in there for backward compatibility. It’s better to rewrite the software and use the new features.”[1] Spencer also found little from Tramiel. “He couldn’t care less if anything was compatible at that point in time,” says Russell. “When you would talk to Jack, he was like, ‘Don’t worry about compatibility. We’ll sell them a new computer for cheap, so who cares about compatibility.’ As far as we were concerned, if it ran BASIC it was compatible.” In the end, Spencer had no option but to defer to Tramiel’s decision. “The marketing people obviously really wanted backward compatibility because that was big to them,” says Russell. “Kit was definitely marketing at that point in time. We got beat up [by Spencer] but Jack silenced them when he said he didn’t care about compatibility. The engineers won that battle.” One of the biggest omissions from BASIC was the lack of programming commands for sound and graphics. Under the original Microsoft deal, Commodore could have added commands to BASIC for the SID and VIC-II chip, but they chose not to because of Tramiel’s reluctance to invest in software. As a result, BASIC programmers found themselves faced with complicated POKE and PEEK commands, which required a comprehensive knowledge of memory locations. This dampened the enthusiasm of many beginners who hoped to develop their own games. The extra commands would have required an extra ROM chip, which cost three dollars at the time. “It would have been quite a bit more money because there would have been a second chip,” says Russell. MOS Technology was also unable to create larger ROM chips. “At that time, our process only allowed us to make a certain sized chip,” explains Russell. A larger ROM space would also require a major redesign of the C64 memory space. Winterble blames the lack of features on the scarcity of engineers. “We talked about putting in macros and the ability to do some of these things, but we just didn’t have the resources to do it,” he says. “We didn’t have the people. It was such a small software group. The idea was we would try to go back and do some of that stuff later on.” Commodore later released a special cartridge with more commands, but it found little success in the marketplace. “The Super Expander cartridge added a bunch

of extensions to BASIC to let you access the features of the Commodore 64,” says Yannes. With the limited ROM space, Russell barely had room to insert his own initials. “In those days we didn’t even dream of putting in Easter eggs because in the C64 I had 5 bytes left when I was done,” he says. “I was going to put in, ‘BYRSR’, but Bob Yannes didn’t think that was a good idea so we just put RRBY in it.” Winterble began to feel annoyed when the marketing people went straight to his engineers to request features. “We had things like the marketing guys trying to call the engineers directly and lobby for changes,” he says. The relationship between Winterble and Spencer continued to deteriorate until the two had an explosive argument. “I one time we bumped heads,” says Winterble. “I tried to keep the guys isolated so they could do their job.” In one heated exchange, Winterble chided Spencer, saying, “You don’t talk to these guys. Leave them alone.” After that, the engineers were free to complete the design without interference. Although the Commodore 64 was not truly software compatible, Commodore paid lip service to backward compatibility. Their marketers promoted limited backwards compatibility, meaning VIC-20 hardware like joysticks, modems, disk drives, and printers would still work on the C64. Additionally, the BASIC 2.0 ROM meant very simple BASIC programs that did not use POKE commands would technically run on both systems. Commodore hoped the ambiguous statement might entice VIC-20 s to buy the new C64. It was probably wise to drop backward compatibility, given the limitations of the VIC-20. Unfortunately, it seemed to set the pattern for computer design in later years. * * * John Feagans continued working at the Moorpark R&D office after the C64 debut at CES. “I developed the music player for the launch of the C64 there and stayed to develop the Commodore network for the C64 and PET,” he says. Inspired by Robert Metcalfe’s visit in 1980, Feagans created a computer network. His network required a master computer, which served up to 10 slaves. “I created a plug-in for the expansion port on the PET and also the C64 which

brought out the serial port from the 6522 to an RS-422 interface,” says Feagans. “The ‘teacher’s PET’ was the master and the C64s were the slaves. I also networked a printer in the configuration.” Feagans envisioned the network for use in the classroom. “The C64s each had a cartridge with the software,” he explains. “The teacher’s PET could software simultaneously to each C64, take control of the keyboard and screen, and even monitor any screen. It was like a combination of PC Anywhere and a print server, as any unit with permission could print on the teacher’s printer.” Unfortunately, Feagans’ prototype system never made it past the trial stage. “I did a field trial at Watsonville, California middle school and also at New Brighton Middle School in Capitola, California,” he recalls. With Commodore centralizing their operations, Moorpark ended. “Moorpark was closed around March 1982 and I moved back to Scott Boulevard,” says Feagans. The VCR data drive, developed for the past year at Moorpark, also did not make it into production. * * * In March 1982, the 30-second VIC-20 commercials began airing during programs like Police Squad, the Dukes of Hazard, and William Shatner’s new show, T.J. Hooker. In the commercials, Shatner stands in front of a star-field asking, “Why buy just a video game from Atari or Intellivision? Invest in the wonder computer of the 1980s for under $300, the Commodore VIC-20.” David Rogers, the scriptwriter of the commercial, humorously confused a game console with a game, saying, “Unlike games, it has a real computer keyboard. With the Commodore VIC-20, the whole family can learn computing at home. Plays great games too.” An addendum to the commercial previews the games Omega Race and Gorf, calling the latter the “wonder arcade game.” It was clear Rogers was struggling with computer terminology, but none of that mattered. The only thing that really mattered was that the captain of the Enterprise was telling people to go out and buy a VIC-20 computer.

Using Shatner as the spokesman was genius. Now people would associate the high-tech image of Star Trek with Commodore and the VIC-20. “Shatner was helpful at that time in giving us credibility, as Commodore did not have much reputation and brand image in the USA at that time,” says Spencer. Print ads came out in magazines at the same time. The first ad showed Shatner presenting the VIC-20 with a sweep of his arm, with a quote saying, “The first honest-to-goodness full color computer you can buy for only $299.95.” An army of life-sized cardboard cutouts of Shatner also began appearing in retail and specialty stores around North America. “We put him on point of sales displays to go in stores and all sorts of things,” says Spencer. Spencer also created videotapes for mass-market retail stores like Kmart. “We filmed a whole presentation which showed all the aspects of the VIC,” he says. “I realized that when we went to the consumer market, at the price of the VIC-20 you couldn’t afford to demonstrate it. You’re selling boxes in large numbers through consumer outlets.” Neil Harris and his group hand delivered copies of the tape to every store they visited. “Nearly all the stores we would sell to had videos and TVs,” he explains. “They do presentations of kitchen knives and other things on VCR tapes. So we did a presentation on a continuous VCR loop that sort of fitted in with theme of the Shatner commercials, but also took you right through what the product could do, showing the games and computer aspects. It was about a 10-minute film, because you couldn’t keep people’s attention any longer than that.” Brand name recognition increased dramatically during the spring. Before Christmas, the average North American did not recognize the Commodore name. Now, everyone knew Commodore. In a few short months, Spencer had turned Commodore’s situation around. For the first time, Commodore sured Apple’s marketing, which lacked highprofile television advertising. In the space of a few months, Spencer gave Commodore the most high-profile campaign out of any computer company. Sales of hardware and software began rising. The Scott Adams games became the most popular VIC-20 games from Commodore up to that time. “We definitely sold over a million Scott Adams’ games total across the six or seven games that he licensed to us,” recalls Tomczyk.

Each 16-kilobyte game cartridge sold for $39.95, with approximately $5.00 going to Adams for each game sold. “Sales were very high and he made a lot of money,” recalls Tomczyk. “I know he built a castle in Florida.” Tramiel preferred paying for licenses outright for a fixed sum, such as his BASIC deal with Microsoft. He disliked royalties. According to Tomczyk, “Jack Tramiel came to me once and said, ‘Why are we giving more than $100,000 to Scott Adams? That’s horrible.’ I said, ‘That’s not horrible, that’s a high class problem. That means we sold so many of his games that we had to pay him all that money.’” * * * Commodore had put together some highly profitable licensing deals with Scott Adams and Bally-Midway. Now, Tomczyk was about to put together their biggest deal of all. In the spring of 1982, Donkey Kong was popular in the arcades. “I loved Donkey Kong and some of the Nintendo games,” he recalls. “I played them in arcades and said, ‘We need to have these on the Commodore computers.’” Tomczyk arranged a meeting with Nintendo to purchase the North American home computer rights to their arcade games. “I went to Nintendo and made friends with a top Vice President there, and negotiated a deal to put Nintendo games on Commodore computers,” he says. “I took that all the way to the contract.”

Scott Adams working in his offices.

Tramiel ed Tomczyk and allowed him to make the deal. It would have been a highly profitable game on the VIC-20 and C64, easily selling millions of copies. “That was a sweet deal,” says Tomczyk. “I had negotiated very good economic . Nintendo was so enthusiastic about putting their games on our computers.” Unfortunately, Tramiel preferred handshake deals. When the contract was ready for Tramiel’s signature, Tomczyk had the surprise of his life. “I took it into Jack for him to sign, and Jack said to me, ‘I changed my mind. We’re not going to do that deal.’” All the work Tomczyk had put into making the deal was now for nothing, and worse, he had to go back to Nintendo and tell them the deal was off. “I was horrified, just horrified,” he recalls. “Not only would I lose face, but it was my personal integrity on the line. I had negotiated in good faith, I had the contract ready to go, Jack had already approved all the and now he just suddenly— for whatever reason—backed out.” Tomczyk argued ionately with Tramiel for the contract. “I’m just so frustrated when I think about that, because it would have doubled our sales on top of what we were already selling,” he says. Nintendo had a stable of popular titles, including Popeye and Donkey Kong Junior, all of which Commodore could have developed and published in North America on its computers. Tomczyk argued for days, but Tramiel merely replied, “That’s just my decision.” “That’s the kind of guy he was. He would make a decision off the cuff and you had to live with it,” says Tomczyk. “I argued hotly to keep that deal and I just totally couldn’t change his mind.” Decades later, Tomczyk is still unsure of Tramiel’s motives. “I have racked my brain ever since then trying to figure out why he did that. The only thing I can think of is that he didn’t want to insult or offend Bally.” It is also possible Tramiel was reluctant to help Japanese companies gain market share in North America.

After Commodore pulled out of the deal, Nintendo went to Atari, which ported Donkey Kong and numerous other Nintendo games to the VIC-20 and C64. * * * In the spring of 1982, Commodore’s east coast employees resided at either MOS Technology in Norristown or the King of Prussia location. The recent migration of west coast employees quickly filled the remaining office space. In response, Tramiel leased a new building at 487 Devon Park Drive in Wayne, Pennsylvania to act as the new headquarters for Commodore US. Tomczyk’s software group was the first to move into the new location. He wanted to take the Commodore games division in a new direction by creating games based on original ideas. To further this strategy, he wanted to give his programmers royalties from the games they designed. The deal would give the programmers more incentive to create hit games. Tomczyk prepared a royalty structure and scheduled a meeting with Tramiel. Unfortunately, Tramiel was not happy with Tomczyk’s handling of the software group. So far, only the Japanese games were hits, and Commodore had not released the Bally-Midway games. “They’re not putting anything out and Jack’s getting a little ticked off about the whole thing,” recalls Winterble. Tomczyk’s latest request for royalties only added to his unhappiness. When profits began rolling in from games, the games group became a desirable division for many executives. “We had become a plum because we were a definite profit center in that each cartridge was making a significant amount of money,” explains Andy Finkel. “So there were two factors going on. Some of the VIC-20 team wanted to be in engineering and engineering wanted the profit center to be where they thought it made sense: in engineering.” Tramiel decided the game group should no longer reside under the marketing umbrella. Office politics was about to deal Tomczyk another blow. “I don’t know how it is at other companies, but Commodore politics were pretty brutal and straightforward, and they all ended with Jack,” says Finkel. Tomczyk was aware of the politics in Commodore. “People were constantly trying to be in charge of the VIC-20 or the big money makers so that it would raise their prestige,” he says. “I constantly had to fight off grey-haired 50-yearold Vice Presidents who thought that I should be working for them. It was a

constant struggle and I never took it personally, it was just part of the political game. It happens in many large corporations.” Charles Winterble paid a visit to the group. “I wanted to see the work they’re doing,” he says. “A guy is working on a piece of software for a rocket game and he demonstrates the thing for me. He’s been working on it for four months and on the screen there are a couple of lines. If you look very carefully, it might look like the stick figure of a rocket with some little dots coming out the front of it. He’s explaining to me what’s going to be there. It’s like selling land in Arizona. ‘Here’s where there’s going to be the swimming pool.’” The only person in the group who seemed able to produce exceptional results was Andy Finkel. “Andy was a very sharp guy,” says Winterble. “If he was managed properly at the time, I’m sure he could do anything.” Winterble, an inner family member, felt the group would perform better within his engineering group. “The game group was taken away from me because one of the executives played some shady politics behind the scenes and convinced Jack that he could do a better job running the game group,” says Tomczyk. For the second time in a year, Tramiel turned Tomczyk’s world upside down. The decision seemed incredible to Tomczyk, who had hired and mentored most of his staff. Tramiel reassigned him to the role of public relations director for Commodore, a function he found unsatisfying. Tramiel asked Winterble to bring the division under control. “So I inherited the group,” he says. Although the group now fell under the engineering umbrella, it did not have a manager to report to immediately. “I came into work one day and found out I was not in marketing anymore in Commodore US but instead I was in Commodore engineering,” says Finkel. “For a while, there was no one directly in charge of us. They really didn’t have anyone to be in charge of the software people they had just moved over.” Later, Winterble found a new manager for the group. “They took a couple of weeks and hired Joe McEnerney, who used to work at Commodore as a programmer for the calculators,” says Finkel. Winterble moved the entire group to MOS Technology in Norristown, Pennsylvania.

Initially, Andy Finkel liked the move. “It was fun being in engineering. We got to have the ID cards for engineering labs. We could wander through the labs. It was nice working a lot closer with people like Bob Russell, Charlie [Winterble], Bob Yannes and Al [Charpentier]. It was really convenient because we’d be able to talk to them. As the new chipsets were being thought about, we would be right there to add our two cents in.” * * * The VIC Commandos had completed Omega Race and Gorf for the VIC-20 by the time Tomczyk lost the games group, but it would be months before MOS Technology released the cartridges. Other games, such as Seawolf and Clowns, were the oldest and simplest to program. Even in 1982, Seawolf was an older title, first released to arcades in 1975. Clowns was an aging but entertaining game originally released by Bally in 1978. Commodore later touted both of these games as arcade classics in their advertising material. By now, the VIC-20 had a large base and Commodore’s games were a huge success. Each game sold for $29.95, with Gorf and Omega Race selling for $39.95. Omega Race became an instant favorite. Jack Tramiel and Irving Gould were pleasantly surprised when each game went on to make Commodore millions of dollars. Since most games had only small groups of one or two programmers, it was one of the most profitable divisions. The Programmer’s Reference Guide sold particularly well. “It was a huge number because everybody was eager to get all the materials they could about it,” says Tomczyk. “We had a very high rate of peripherals and software going out with the first wave of computers.” Harris was particularly pleased to see the book embraced by the programmer community. “We threw it out there and the book sold millions,” he says. “But the key thing was it got into the hands of the right people so that all the information people needed was available to them to write stuff for the machine. So software and hardware was tremendous thanks to that.” Finkel and Harris had not told anyone they had included schematics of the VIC20, and they started feeling backlash. “There were a lot of people inside the company who would rather that hadn’t happened,” says Harris. “But it was much healthier for the computer and helped the C64 and VIC-20 sell in the millions

because you could get whatever you wanted for it.” Because of Commodore’s openness with VIC-20 information, and the rapidly growing base, the VIC-20 became irresistible to game publishers. In total, the VIC-20 attracted over 100 publishers, including Avalon Hill, Broderbund, Cosmi, Creative Software, Epyx, HES, Imagic, Parker Brothers, Sega, Sierra On-Line, and Sirius. From Britain, the VIC-20 lured developers like Llamasoft, Anirog, Romik, Mastertronic, Ocean, Firebird, Thorn EMI, Virgin Games, and Rare. In , a developer named Kingsoft dominated the games market. Most European publishers cheaply distributed their games on tape. Popular games from other computer systems were also ported to the VIC-20, including Choplifter, Lode Runner, Miner 2049’er, and Shamus. Even Atari eventually released official ports of their best arcade games to the VIC-20, including an official version of Pac-Man. Unfortunately, many of their ports were substandard and Atari took a lot of abuse over these lackluster games. Atari took particular abuse over their Pac-Man conversion for the Atari 2600 console, released in March 1982. “The Atari version was horrible,” says Tomczyk. “It was pixilated and crude and really bad. It was not a good implementation of the game.”

Comparing Jelly Monsters (top) with the official Atari version of Pac-Man

Atari’s lawyers fared better. In 1982, Atari won a lawsuit against Magnavox, which had produced a variation of Pac-Man for their Odyssey game console called K.C. Munchkin. Commodore had been selling HAL Laboratory’s version of Pac-Man for close to a year in Europe, and amassed a fortune doing so. “Pac-Man was sold in Europe as Jelly Monsters,” recalls Tomczyk. “It easily sold a million units and was probably responsible for selling at least a million Commodore computers.” At the conclusion of the Atari lawsuit with Magnavox, Atari began going after other companies with Pac-Man clones, hoping to settle out of court. Tramiel had planned for this eventuality, and had set aside a portion of Jelly Monsters’ profits to give to Atari as royalties, even though no agreement existed. “When Atari finally came after us, it’s my understanding that we settled quickly and efficiently so we didn’t suffer any problems,” says Tomczyk. Tramiel’s idea to sell Jelly Monsters, despite threats by Atari lawyers, had worked in every respect. “The reality was, if we sell a million of this game, which is technically a rip-off, and we have enough royalties to make Atari stand up and take notice, Atari will just smile, take the royalties and walk away,” explains Tomczyk. “It sounds unethical but in practice it works.” * * * As the engineers refined the C64 design, Kit Spencer prepared materials for the product launch, including manuals, packaging, advertising, and a software library. “I had to make sure we got all that right because this was going to be a very big operation that we were committing to,” he says. When Neil Harris first ed Commodore, his group was responsible for manuals, software, and advertising literature for the VIC-20. By the time the C64 came along, things began to change. “It wasn’t until the VIC-20 really started to take off and get big that they started splitting things out into separate departments,” he recalls. “This was at a time when Mike Tomczyk’s original team had been broken up and scattered.”

In the software group, programmers badly wanted to the new Commodore 64 development team. “Within the software department, the C64, from the first time we saw it, was the hot new machine,” says Andy Finkel. “So everybody wanted to work on it. It was just a much better, more powerful machine than the VIC-20.” Andy Finkel remained in the software division for the C64. His friend, Neil Harris, was still primarily on the road educating retail employees about the VIC20. A third division would produce the manuals and packaging. According to Spencer, “I’ve seen it screwed up in earlier days when we had no manuals, no programmer’s information, bugs, no software, no literature, and lousy packaging.” Spencer needed access to technical information about the Commodore 64. However, the engineers in Charlie Winterble’s group were busy with the C64. Disturbing them would only slow down the product release. “Charlie had to develop a product and get it out quick because time was important so he couldn’t spend a lot of time talking to anybody else,” explains Spencer. “That I understand and accept. On the other hand, I was charged with getting this product launched internationally.” Spencer and Winterble came up with an idea. “I going to Charlie and discussing the problem that I need to know everything about the machine to get my marketing launch right,” says Spencer. Spencer came up with a radical solution of hiring an engineer to help with the C64, but who would not report to Winterble. “What we did in the end was agree that I would hire one technical guy who would be working for me, and that guy would have 100% access to Charlie’s team. He would work like one of Charlie’s team, knowing all the technical stuff. In return, nobody anywhere in Commodore in any marketing organization in the world was going to Charlie.” While the engineer worked on the C64, he fed information to the manual writers. “One of the things I did with the launch of the Commodore 64 was to work hard to have a comprehensive technical manual as well as the basic product manual available at the launch,” says Spencer. Andy Finkel knew the importance of the Programmer’s Reference Guide to the eventual success of the VIC-20. “Part of the success of the VIC-20 was we

actually got all the technical information out into the hands of anybody who wanted to buy it for $20,” he says. “Having all those people with the actual information really helped make it a popular computer.” Now that Finkel was in the software development group, he had nothing to do with the production of manuals. “At that point, Commodore had its early documentation department set up, so they were no longer being written directly by engineers or programmers,” he says. A technical writer created the first draft of the manual. He did not have hands-on time with an actual C64. Instead, he wrote the manual using information supplied by Kit Spencer’s engineer. “I had one guy who was just writing the technical manual and the instruction manual for the product at the same time as Charlie was developing it,” explains Spencer. When programmer Andy Finkel saw the first draft of the technical manual, he felt instant disappointment. “I thought we’d established a certain level of quality and usefulness for the VIC-20 Programmer’s Reference Guide and I was really upset that we had gone off the rails,” he says. When Finkel tried typing in some of the examples, the code failed to work. “Giving people a bunch of things that really didn’t work and having chip manuals translated by non-technical people into smooth, pleasant, incorrect material was a step backwards and I thought it would really hurt the C64.” Finkel was firmly in the software development group, but he decided to take charge of the faltering project from the documentation group. “This being Commodore where you can actually make a difference if you are willing to do the work and take responsibility for it, I was able to—even though that wasn’t my area anymore—get in on the C64 Programmer’s Reference Guide.” One day while Neil Harris was home from his frequent road trips, he received a call from his old friend. “He called me up at home and said, ‘I just got the first draft of the Commodore 64 programmer’s reference guide that some other guy in the company had been working on. There’s a problem.’ I said, ‘What’s the problem?’ He said, ‘It’s fiction. I just read though their manuscript and it makes no sense. It’s not accurate in any way and it’s a complete disaster. We’re going to have to throw it away and start again. I need your help.’ And I said, ‘Of course I’ll help!’”

Finkel and Harris also recruited Paul Higginbottom from Commodore Canada in Toronto. “We did this at home in our spare time because it needed to be done,” says Harris. “We were ionate about it and it was the only way it was going to get done right.” As Harris worked through his material, he saw the problems with the first draft. “Somebody thought he knew what he was talking about,” he recalls. “He had no freaking clue. In the course of the next three weeks— from start to finish—we wrote that book.” Because the programmers were intimately familiar with the VIC-20, if was easier for them to develop the new C64 reference guide. In the end, the manual was not as playful as the original VIC-20 manual, perhaps due to Tomczyk not being quite as involved, but it was a solid guide. * * * Tramiel had noticed the work Harris had done on the manuals and his subsequent presentations to retailers. “Jack would see me in operation and he knew I was loyal to the company and loved what we were doing,” recalls Harris. When it came time for a massive shareholder meeting, Tramiel recruited Harris to demonstrate the Commodore 64 prototype in front of hundreds of critical shareholders. “I was always one of those guys who was technical, but I could also speak to the general public without too much fear,” he says. “They would bring me up to New York to work shareholder meetings, to work the computers and talk to the nice shareholders.” After the meeting, Harris received a vote of confidence from Tramiel. “Jack took me personally aside after a shareholder meeting and said, ‘Look, we should have done this for you long ago, but we’re going to put you on the executive bonus program. The first year you have your choice of stock options or a cash bonus. Which would you like?’” It would have been easy to overlook the dedication Harris had for Commodore products, but Tramiel’s gesture made it clear he was paying attention. “It was great to get that personally from the founder and CEO of the company,” recalls Harris. “It really meant a lot to me. I thought that was fantastic.” [1] Al Charpentier does not recall considering a VIC-I mode for the VIC-II chip.

CHAPTER 28

Delay of Game 1982

Jack Tramiel expected his engineers to have Commodore 64 computers rolling off the assembly line three months after CES, but it soon became clear the target was overly optimistic. The second revision of the SID chip, which included bug fixes, began rolling out of MOS Technology the week of March 14. The revised VIC-II chips were yet to come. Bob Yannes blames the unrealistic goal on the seemingly refined state of the Commodore 64 at CES. “After the CES show everyone was thinking that it was a real product because it looked real, because we used the VIC-20 case and just painted it,” he says. “It looked like it was ready to go into production.” With more engineers, Commodore would have been able to divide the tasks. “I think maybe there were 10 or 12 engineers in the whole company that were true engineers, including software, hardware, and chip design,” says Robert Russell. “Including people in Japan, the total number of engineers in the company was 15 when the Commodore 64 was built.” Most of the engineers were working on other diverse projects for Commodore. “That includes people working on the PET, and a couple of guys down in Texas doing the whole cash business,” says Russell. With the final design underway, the determination to meet a low-cost target grew stronger. According to Al Charpentier, “We agonized over every transistor.” For his part, Charles Winterble vetoed ideas for added features. Although the case would become as distinctive and loved as the Volkswagen

Beetle, Charpentier himself was never a big fan. “I always thought the VIC-20 case looked clunky,” he says. Winterble also had his own reasons for disliking the case. “If we had made the case bigger to begin with, we would have made the design much more manufacturable,” he says. “It was hell for those poor guys in manufacturing. We squeezed things to the limit to fit all this crap on the board. This was a product that was manufacturing hostile.” Yannes claims they had plans to substitute the case with a new design in eight to ten months. “We always wanted to design a new case for the Commodore 64,” he says. “It was never our intention that it would go into production with the VIC-20 type case. I would have liked to have seen much better styling on the case.” Charpentier recalls the proposed design. “It was thinner in front and had more of a wedge shape to it,” he says.[1] The production design seemed to be going smoothly until the engineers attempted to improve the video chip. “We took the time to make a design change in the video chip, which I think both Al and Charlie would agree that we probably shouldn’t have done,” says Yannes. The engineers wanted to improve the color on the chip. “It was the same thing with the Apple and Atari computers,” explains Yannes. “You would get interaction between the luminance signal, which is the black and white information, and the color signal. So you would end up with these various colors on the screen that weren’t really what you wanted, but it was just the nature of the NTSC video standard that the luminance and chrominance signals interact with each other.” To purify the colors, Charpentier made a risky last minute change. “Al had the idea that if the two clocks were independent from each other, then that interaction wouldn’t happen,” says Yannes. “We separated the clock generators on the VIC chip so that the color crystal was a separate clock from the video shift rate.” The engineers noticed the default white on dark blue screen no longer looked appealing. “Even though it had good contrast, the transition from blue to white produced kind of an ugly edge,” explains Yannes. “We ended up having to make

it light blue on dark blue.” Charpentier’s chip modifications had an unintended side effect. “We didn’t get the false colors anymore, but unfortunately because the two clocks weren’t phase synchronized at all, so there was this waviness in the screen which was really objectionable,” says Yannes. The engineers required a Texas Instruments chip to fix the problem. “We ended up having to throw in this phase-lock loop circuit to lock the two clocks together, which added cost to it.” According to Yannes, Tramiel was not happy about the extra chip. “It was particularly messy because the only [company] who made this chip was Texas Instruments, and we were going up against their Ti-99/4A computer with the Commodore 64,” says Yannes. “So here we are dependent on one of our major competitors for a critical component.” Commodore regained control over their chip supply by using MOS Technology to clone their competitor’s chip. Commodore could now manufacture a computer better than Apple or Atari at a cheaper price. “The reason we could is because of vertical integration,” says Winterble. “We put so much in just a couple of chips.” Without vertical integration, the custom SID and VIC-II chips would for a much more significant portion of the costs. Atari and Apple lacked the in-house expertise to manufacture their own custom chips, which meant they had to hire costly outside companies (often MOS Technology) to come up with designs. As a result, Apple and Atari computers were both costlier and not as technologically sophisticated. The low production cost meant Commodore could sell their computer for under $600. Now that Commodore was working with mass retailers like Kmart, it was crucial to cut costs. “In the consumer business, on a standard product you’re probably working with 15 to 20 percent markup for the dealers,” says Kit Spencer. “With a hot product, they go down even less on their own margins. Some were almost only selling at their costs at one stage.” * * * In April 1982, Kit Spencer received approximately six prototype C64 computers to use for development. “We had some early prototypes to debug and try and get software written,” says Spencer. “We laid down a list of the key software we felt we’d need at launch. We said, ‘We need some games, we need some educational

software, we could do with a spreadsheet and a word processor.’ These were all part of a basic launch package.” For games, Spencer could rely on the US software group and Commodore Japan. He sent one prototype to Commodore Japan and asked Tony Tokai to put one of his programmers in touch with HAL Laboratory so they could avoid another copyright fiasco. Andy Finkel handled the talks with the HAL Laboratory kids. “At that point we were able to talk to HAL and say, ‘Hey, this is what we need,’ rather than them doing stuff on their own. So they actually gave us some decent usable stuff that we could actually sell.” Finkel was impressed with the Japanese games but shocked by their ages. “The HAL Lab programmers were very bright and very young. The first time they said, ‘Well I can’t finish this because I’ve got after school work for the next three weeks.’ It was that kind of thing. They were young kids!” For debugging and educational titles, Spencer went to Commodore and Commodore Canada. “I going around the Commodore companies and saying to the general managers, Harald Speyer and Jim Dionne, ‘We need to get these software products. Do you know somebody who could do them? We’ve got a prototype we can put out, but there’s only a few of them and they will have bugs.’” The PET computer had many educational titles already, since the PET was a big seller in North America for schools. It would be a simple matter to convert the titles to the Commodore C64, since they both used the same BASIC and both had 40-column displays. Jim Dionne arranged for a Canadian university to port the educational titles. “One prototype went to the computer department at Waterloo University in Canada because they had a very strong education emphasis and had a large PET computer base there,” recalls Spencer. “They took all the free Commodore PET educational software, and the professor, during the summer break, hired all his best students, who worked in shifts 24 hours around the clock on one prototype to convert all the software they had over to the Commodore 64.” Spencer planned to release the educational titles at no cost to schools. A third prototype went to Commodore UK. “In the UK, they developed a spreadsheet and a word processor on another prototype,” recalls Spencer.

A fourth prototype went to Commodore , while the remainder stayed with the software group in Commodore US for game development. The developers reported any bugs they found to Spencer, who ed them to Winterble. “They were also helping to debug it at same time,” says Spencer. “They would not talk to Charlie’s team directly; they would talk to the technical guy I had hired and the man doing our manuals who would then feed the bugs and anything else relevant into Charlie so his people could get on with it.” Spencer’s efforts coordinating the launch would ensure that the C64 did not face the same delays the VIC-20 had with software. For the most part, Spencer’s launch team coordinated well with Winterble’s development team. “We may have had one or two ‘discussions’ early on that may have appeared like differences of opinion,” laughs Spencer. “That is probably because Jack had charged us both with large tasks and we both had our different priorities to look after. But I think in the end, we worked quite well on that and I’d like to think we respected each other’s talent.” * * * The stressful work environment sometimes produced casualties, even within Tramiel’s inner family. “Jack drove a lot of people to tears,” recalls Michael Tomczyk. “He could sit in a meeting and drive a vice president to tears. He would fire vice presidents in the room. People would Commodore, sell their house and come and move there, and a few months later they would be fired because they didn’t perform or they didn’t do what they promised.” At the spring Hanover Fair in 1982, Tramiel and his engineers brought several C64 prototypes for demonstration. Prior to the show, a member of Tramiel’s inner family experienced a dangerous mental breakdown. “We were at a trade show in , and we were all going to head someplace as a group and we wondered, ‘Where’s Jack?’ Nobody could find Jack or figure out what’s going on,” says a former Commodore employee.[2] “Then a couple of people took us to the side and we went up to his room. Jack was in his room with his wife Helen, John Calton, and a woman who was a manager, and they were being held at knifepoint by an employee.” “One of our employees, who was a manager, went insane, and had a nervous breakdown,” recalls Tomczyk. “He was running around naked and threatening

people.” The tense standoff lasted hours. “We were sitting in a room, trying to get updates as to what was going on,” says the Commodore employee. “It went on all day long. Eventually, he was occupied in one room, and they all ran out of the room. They came down the stairs, and he chased them down into the lobby, and they went running through the lobby.” Hotel security quickly apprehended the employee. “When this guy was running through the lobby, he was noticed fairly soon, because he was nude,” he recalls. “He was subdued.” For the third time in his life, Tramiel had survived a near death experience. As it turned out, Tramiel and John Calton were not the only executives the manager wanted to kill. “Apparently he had a list, and on the list were the five or ten people in the company he intended to kill.” Tramiel made sure his manager received proper care at a local hospital. “In , if you are convicted of a mental health issue, you could have very serious consequences at that time. You wouldn’t be able to get out of the country, and you would be locked up. So before he got out too fast, Jack arranged to pick this guy up out of the hospital and took him out of the country on a plane.” Despite the personal attack, Tramiel felt comion for the man. “Jack made sure he took care of this person. He went out of his way to do everything he could do to make sure this person was taken care of. There was no grudge or anything like that. This was an inner family kind of guy, and Jack took care of him.” * * * In 1982, Commodore was on the verge of releasing a technologically superior computer for the masses. The timing could not have been better. Computers were about to go mainstream and the marketplace was ready for a breakthrough product. Unfortunately, the design was not yet ready by April 1982, when Tramiel wanted to enter production. Many problems remained and the computer had not yet entered the lengthy FCC testing phase. The intense pressure to release the Commodore 64 ended up causing a debilitating defect with the serial port. Robert Russell was disappointed with the

slow speed of the VIC-20 serial bus. “The slowness has to do with the serial bus,” he says. “It has nothing to do with the disk drive.” He planned to improve the speed dramatically with the C64. Russell modified the design of the serial bus by adding high-speed lines and replacing the 6522 I/O chip. “I had a high speed method already set up in hardware,” he explains. “On the Commodore 64 we had the 6526 chip, and that had working high speed lines on it. On the schematics that I did, I had those lines connected to the serial port.” Russell also wanted to release an improved 1540 drive called the 1541, which also used the 6526 chip. “The original plan was to use the 6526 in what was going to be the 1541 drive and the Commodore 64,” he explains. His changes would have given the C64 the fastest disk access on the market. “It would have been 20 or 30 times faster,” he claims. “It would have run at the limit of the drive rather than the limit of the serial bus.” Russell had not yet upgraded the ROM code as the release date neared. “The 1541 would have been high speed even on the early C64s, but I couldn’t get the code done in time because we were busy fighting other problems with the chipset,” he says. He planned to change the ROM code once he received production samples of the C64. Commodore Japan would actually fabricate the printed circuit boards (PCBs) once the design was complete. Yash Terakura created the PCB layout, working alongside Russell, and then sent his work to Fujiyama and Aoji in Santa Clara. “Our final schematics and all final production drawings were being done in Santa Clara because we had no people that did them on the east coast,” says Russell. Once the Santa Clara engineers completed the final production drawings, they would send them to Japan for production of circuit boards. Terakura worked with the engineers in Japan when it came time for mass fabrication. “On the Commodore 64, I worked on the board layout,” he says. “I sent it to Japan and we told them to a make a PCB. Like before on the VIC-20, the pre-production engineering was done in Japan.” According to Charles Winterble, the cramped VIC-20 case left very little area to work with. “We didn’t have the experience at the time to design

manufacturability and we made a lot of mistakes,” he says. “We squeezed things to the limit to fit all this crap on the board.” Commodore began fabricating the boards, amassing them for the assembly line. Unfortunately, no one at Commodore inspected and tested the final production units thoroughly before mass-producing them in the thousands. “When we released something, they built as much as they could of it,” explains Russell. Bob Yannes offers some insight into the immense pressure to deliver the new computers. “The bottom line is, look at the rate the VIC-20s were selling. They were doing thousands per day at one point. They would say, ‘If this takes you two extra weeks, how many millions of dollars is that?’ We were like, ‘Yep, you can’t argue with that.’” “Timing was everything in our business,” explains Kit Spencer. “Time is money. Products are always coming down in price, and always doing more, so if you can do it six months quicker than the next company, that’s tremendous. That’s why we were always squeezing to get things out.” Russell finally received a C64 circuit board and began examining the finished product. “I’m doing the tests and everything is working fine because we hadn’t written the high-speed code yet,” he recalls. “Then I looked at the board and said, ‘Where are the high-speed lines?’” Someone on the west coast had changed Russell’s schematic. “The production guys took them off when they did the production boards. I put high-speed lines on and they deleted them,” he says. When Russell realized what happened, he was livid. “I threw a hell of a fit,” he recalls. He was determined to find out why someone had neutered his high-speed serial bus. “I tracked it down and it was the production engineers in California who cut it off.” To make the board fit in the cramped VIC-20 case, the engineers removed the traces for the high-speed lines. “The guys that actually did the production board layout cut off the signals to save some money,” explains Russell. “They thought, ‘Why are these extra lines running to these signal pins?’ So they chopped them off and screwed us.” It was like building an eight-cylinder engine with eight fuel lines and cutting off seven of them.

Russell was determined to rescue the disastrous situation. “I ran down to Charlie, throwing a total fit. He says, ‘Well is it still functional?’ I said, ‘Yeah, it still works as a 1540.’” Winterble looked into the situation and found out the production facility had manufactured too many circuit boards already. “We couldn’t change it after hundreds of thousands of PCBs were in production,” laments Russell. Stopping production on the C64 and restarting it with a new design was out of the question. “Technically it would have been possible, but you’ve got to realize, they were already moving their production and going to ship,” says Russell. “If I had done that, it would have been several weeks until I got a finished unit for evaluation.” If Russell attempted to make changes now, early customers would be extremely unhappy. “There would have been a bunch of machines out there that would have been incompatible,” he explains. It was now pointless to design a faster 1541 drive. “We never bothered spinning another drive,” says Russell. “The 1541 became just a 1540 with minor software changes.” The deletion of a few metal circuit traces ultimately resulted in millions of wasted hours for C64 owners. Incredibly, the drive became even slower when they attempted to make the 1541 compatible with the VIC-20. “The biggest compatibility pain in the butt was that stupid VIC-20 disk drive,” says Charpentier. “We didn’t want to do it but marketing really forced us into it.” Charpentier believes the decision to remain backward compatible was ultimately shortsighted. “They had inventory on the shelf,” he says. “If we make a change to the C64 and bring out a new disk drive, we’ve got unsold inventory. Issues like that cloud people’s judgment.” Russell originally intended the C64 to interface with an improved 1541 disk drive that used the 6526 chip. Instead, he had to make the 6522 in the VIC-20 talk with the 6526 chip in the 1541. “We thought it was a rather straight forward redo using the new I/O chips rather than the 6522, and it wasn’t,” says Winterble. “He had a lot of difficulty getting the disk drive to work using the 6526.” Russell struggled with the problem, but soon began feeling intense pressure from

Tramiel. According to Russell’s friend Bil Herd, “Jack said, ‘It’s going to be working Monday and you’re going to set it right here.’ He showed him where on the desk it would be sitting and working. Bob [Russell] wrote the serial bus over a weekend in software.” Russell had almost no time to address the problem properly. “It started out slow and we made it even slower,” says Bob Yannes. “So much of the processor’s time was being interrupted by the video chip and the drive couldn’t keep up with it, so they had to slow it down even more.” To maintain backward compatibility, Russell intentionally slowed the 1541 drive speed by four times to work with the VIC-20, even though the C64 used an improved 6526 controller. The new 1541 took over two minutes to load a 64-kilobyte program into memory. Compared to the competition, the C64 now had the slowest drive by far. “I wasted millions of people’s hours I was later told,” says Russell. The solution made every engineer at Commodore shudder. “Of course it’s slow because it makes up for broken hardware,” says Herd. “Things like this were done to make the date, to keep their jobs, and most of all to keep Jack happy.” Charpentier expresses his frustration at the marketing group. “The unfortunate problem was, the marketing and sales people really didn’t understand what computers were about,” he says. “To them it was just another thing and the world says make it compatible, regardless of whether or not you are going to cripple the machine, which they did.” Yannes, a perfectionist by nature, felt the mentality to ship at all costs hurt the final product. “It was like, ‘Well, you have something that works. Ship it.’ We were like, ‘But it doesn’t work very well.’ They would go, ‘It doesn’t matter because it doesn’t need to work well. It just needs to work so we can ship it.’” Al Charpentier also believes it was a mistake to use the antiquated 1540 design. “It was an interesting problem because Kit was there at the same time as we were trying to convince the marketing crew not to use that terrible disk drive,” he recalls. “We wanted to come out with a new disk drive with higher speed, and we didn’t win that argument. It was never clear to me if the marketing guys believed that or that was what Jack was telling them.”[3]

* * * In April 1982, the reckless production schedule resulted in a costly error. Robert Russell sent his ROM code to MOS Technology before validating the code, in an attempt to give them a head start on producing the actual ROM chips. Once he validated the code, he would sign off on it and the facility could begin mass production. However, the temptation to begin production proved too great. Russell continued testing the ROM code and realized there was a serious bug. “I made a mistake and a shifted character wasn’t correct on the keyboard. If you did shift-some character it was still lower case,” he says. When Russell ed MOS Technology, he was horrified. “They went ahead and started manufacturing and made a million of them,” he recalls. “I came in and said, ‘Whoops! Stop everything. Don’t release that part.’ They said, ‘It’s too late.’” As a result, Commodore had hundreds of thousands of worthless, unsalvageable ROM chips that cost over a million dollars to manufacture. Russell expected retribution in the form of a Jack Attack, but management treated the situation with humor. “The president of the company [Jim Finke] shows up in my office,” recalls Russell. “He said, ‘That’s going to be a buck a day for the rest of your life that you are here at Commodore.’” To make matters worse, the C64 seemed to have a congenital defect with the video chip. “You would see light blue sparkles on dark blue, especially after it heated up,” says Bil Herd. The bug became so pervasive s began calling it the sparkle bug. Though it sounded like a friendly insect, it was far from harmless. With the hardware unchangeable for the immediate future, it was up to Robert Russell to attempt to fix the problem in software. “They rewrote the code so that the unused characters were changed to dark blue. It was sparkling dark blue on top of dark blue so you couldn’t see it,” says Herd.[4] Russell also blames the Federal Communications Commission for adding to their difficulties. “They also ran into the whole FCC requirement which really made things a lot more complicated,” he explains. “The FCC is a whole other nightmare with prima donnas.”

In June 1982, completed computers finally began rolling off the assembly lines in Santa Clara. Tramiel was livid over the delays. “It took us six months, and we all got crap over that,” says Russell. With the production design complete, it was now a matter of building up enough VIC-II chips, SID chips, circuit boards, ROM chips, power supplies, cases and other parts needed to build mass quantities of computers. * * * Just days before the June CES show, Paramount Pictures released Star Trek II: the Wrath of Khan on June 6, 1982. Andy Finkel and Neil Harris, the two devoted Trekkies who gave Shatner his Commodore computers, had wondered how he would use them. When they sat in the theater on the first day, they received an unexpected surprise when their computers showed up in an early scene with James T. Kirk and Doctor McCoy. “In the Star Trek movie he has a Commodore PET in his apartment along with the other antiques. It’s even on showing bytes free,” says Finkel. “It was definitely a kick. I couldn’t have planned it any better if I tried.” A month later, on July 6, 1982, Disney gave the image of computers a face-lift when they released a groundbreaking film about the world inside a computer called Tron. It was the perfect climate for Commodore to promote its new computer. Commodore’s own Michael Tomczyk expressed his thoughts on Tron and videogames to the press. “The concept of Tron, when you are really the computer—and the computer is you—is definitely going to happen. That’s the ultimate. The trends of the pricing and power and technology indicate this is really going to happen. The physics of the fantasy expressed in that movie are probably impossible, but we are moving in that direction. We are moving toward the ultimate Tron.”[5] Bally-Midway released an arcade game of Tron. It was a huge success, earning more than the film release. Unfortunately, the game was off-limits to Commodore because Mattel purchased exclusive home rights for the Intellivision. “That wasn’t one of the ones they offered to us,” says Finkel. “We would have grabbed it otherwise. They didn’t have the right to license the home computer game rights, since they did the game under a license from [Disney].”

* * * As the June 1982 Consumer Electronics Show rolled around once again, Commodore focused on promoting the VIC-20 computer. Spencer used his Shatner-themed 10-minute promotional film to explain the VIC-20. “We’d run it at trade shows, and dealers could run it to really demonstrate what the product could do,” he says. Tramiel announced a new price for the VIC-20, dropping the retail price below $200. As a result, Commodore began shipping 70,000 units per month for the remainder of 1982. Many were sold at Toys’R’Us, which recently opened computer departments in their stores. In retaliation, Texas Instruments offered a $100 mail in rebate on the $300 TI99/4A in August 1982. Texas Instruments was attempting to follow the same path they used to dominate the calculator industry. Although the TI-99/4A was a more sophisticated machine than the VIC-20, the VIC-20 had a much larger software library, which made it a more valuable machine for s. Commodore also brought their line of computers to the show, including the C64, P128, B256, BX256, and the Max Machine. As with the previous CES, the C64 attracted the most attention. This was the first CES show where Commodore unveiled the true specification of the C64. With the IBM PC already on the market with DOS, many people noticed the antiquated operating system of the C64. “A number of people came to me and said, ‘Why didn’t you do that with the operating system?’ You could maybe agree with them, but you have got to say time and money were constraints that were important and the product worked,” says Spencer. This time, Commodore had more than simple demonstration programs for the computer. “The Japanese converted some of their VIC-20 games to the Commodore 64,” says Yannes. Commodore was able to demonstrate Avenger, Jupiter Lander, Night Drive, Super Alien, and Radar Rat Race at the show. He adds, “They were not particularly well done.” Yannes realized the Japanese programmers rigidly adhered to the preliminary specifications for the SID chip. “That was one of my definite peeves with them,” he says. “They were so hung up on the specs of things that they would write stuff according to the specs, even if it didn’t work.”

Entering the front doors to CES in Chicago.

As a result, the games shown at CES did little to show off the true potential of the SID chip. “They had games where you couldn’t even hear any of the sounds on them,” says a dejected Yannes. Bill Seiler was surprised at Russell’s appearance at the show. Since arriving in California, the hefty former Iowa native had slowly adapted to the west coast lifestyle of outdoor activity and healthy eating. “He was a big chubby guy and then he lost all his weight,” says Seiler. “I hardly recognized him.” Seiler, the original designer of the IEEE interface, immediately noticed the slow speed of the disk drive at CES. “When the serial bus got to the C64, they had to slow it way down. It was painfully slow,” he says. “It couldn’t run at the speeds we originally designed it for. Compared to the Apple II, ours was like five times slower than theirs.” Yash Terakura brought his Max Machine to the show. “There were different variations of it,” says Robert Russell. “One was just a stripped C64 without a keyboard and with a cartridge slot in it that pointed upwards.” Another variation included a membrane keyboard. Although Kit Spencer did not strongly the Max Machine, he produced a color pamphlet that read, “Commodore announces the third-generation game machine: a true computer and music synthesizer that will out-zonk, out-zap, outsing, out-think, out-program, out-teach and out-sell the competition.” The Max Machine met with little enthusiasm. Tramiel hoped it could compete against the Intellivision, Atari VCS, and ColecoVision for the console market. As a console, the Max Machine had the best sound and graphics capability of any competing console at the time. However, it seemed uncomfortably positioned somewhere between the computer market and the console market. Commodore developed a BASIC cartridge for the system, which seemed to indicate it was a computer. If it was a computer, it was glaringly inadequate, with only four kilobytes of RAM and a membrane keyboard. The future of Max was in question.

During CES, one event caused a shakeup at the top of Commodore’s North American management. “The president made a decision at a press conference and announced it,” says Yannes. “Jack called him later and said, ‘What are you doing making decisions?’” It soon became clear to Finke that even the president of Commodore held little power with Tramiel around. In June 1982, after serving for 18 months as president, Jim Finke resigned. After his departure, he made a cryptic statement about his cigar smoking CEO to the Philadelphia Enquirer. “[Jack] comes in like a lighted flare in a darkened room. He illuminates the scene with such brilliance that you’re almost blinded. But his vapor trails take a lot of the oxygen out of the air. And when he leaves the room there’s no more light.” Although Finke’s departure at the time seemed to have little effect on Commodore, it would ultimately lead to Kit Spencer’s departure as the mastermind behind Commodore’s marketing. At the same time, the lackluster orders for Commodore’s PET-based cash at CES triggered the cancellation of the project. John Feagans was one of the last employees to visit the Dallas-based subsidiary. “I made a trip to Mexico City to visit the PET distributor there,” he recalls. “We flew from San Jose International Airport to Mexico City. On the return, the PET Jet dropped me and Jack off in Dallas after a visit to Commodore Cash .” Tramiel ordered the remaining engineers to relocate to the new company headquarters in Wayne, Pennsylvania. “They ended up with us on the C64 in later days, because they were some of the few resources that existed within the company,” says Russell. While in Dallas, Tramiel had his PET Jet privileges usurped by Irving Gould. “We flew commercial back to San Jose International Airport as Irving had dibs on the plane,” says Feagans. * * * Jack Tramiel accomplished his goal of “computers for the masses and not the classes.” In 1982, the VIC-20 became the most popular computer for computer s.[6] By late 1982, Apple had sold 700,000 computers.[7] As they watched their sales slowly creeping near the magical one-million mark, the VIC-20 overtook them. “The VIC-20 was the first computer to sell more than a million units,” says

Yannes. “The VIC-20 hit a million units after not even a year of production.” The milestone was impressive, given the short time it was on the market. “It did that months before Apple sold their millionth Apple, even though the Apple II had been on the market since 1977,” says Yannes. “That gives you an idea of really how slow their sales were. That’s not exactly a fantastic sales rate.”[8] Sales of the VIC-20 were impressive compared to the other big entry in 1981, the IBM PC. In fact, IBM only shipped 13,000 computers between August and December 1981. This compares to Commodore International shipping up to 9,000 VIC-20s per day at its peak. In the midst of success, Robert Russell felt sick for turning down Tramiel’s offer of a dollar per VIC-20. “They sold a million of them! I’m an idiot,” he says. The release of the VIC-20 disk drive in early 1982 was less successful. A senior executive named Alan Fink grossly underestimated demand. Kit Spencer compiled a sales forecast, which estimated 15% of VIC-20 owners would purchase a drive. Fink ignored Spencer’s figures and produced his own forecast, which predicted only 1.5%. Unfortunately, due to Fink’s poorly constructed estimates, Commodore manufactured very few drives. As it turned out, close to 15% of VIC-20 owners eventually purchased 1540 disk drives. The demand caught Commodore by surprise, and they were in short supply throughout 1982. Alan Fink survived Commodore until after the summer CES. [1] Many years later, Commodore adopted this design in the revised Commodore 64c. [2] The employee wished to remain anonymous. [3] In the end, the 1541 drive worked with the VIC-20, but the 1540 drive was not fully compatible with the C64. A bug in the VIC-II chip caused the C64 to freeze when using the 1540. [4] According to Herd, “Russell was later rewarded with a $20,000 bonus for coming up with this solution.” [5]Compute! Magazine, October 1982.

[6] One of these s was Linus Torvalds. In 1981, the nine-year-old received a VIC-20 from his grandfather and used it to learn BASIC programming. He was taking the first steps which eventually led him to create a revolution with Linux. [7] Time magazine, “The Hottest-Selling Hardware” (January 3, 1983), p. 37. [8] Steve Wozniak has attempted to claim the Apple II was the first to a million. On BBC World’s Most Powerful, aired December 2003, Wozniak claimed, “Sales shot sky high. Apple was the first company to sell a hundred thousand computers—a million computers.”

CHAPTER 29

Selling the Revolution 1982

With the engineering job complete (or at least good enough), Charlie Winterble’s team reluctantly stepped away from the Commodore 64. “We accepted the fact that our job was done for a long time,” says Winterble. “I don’t in any way fault the fact that the marketing teams and the sales teams and those people had to come in and do their jobs. It’s like a relay race, and it was time to hand off the baton.” Kit Spencer had proven himself by making the PET the most popular business computer in Britain and then he spectacularly re-launched the VIC-20 in North America. “That strategy gave a base that helped the Commodore 64 launch as well,” he says. Jack Tramiel planned to launch the Commodore 64 worldwide. “At that time, virtually every computer manufacturer from Apple to IBM to Texas Instruments made computers for the US markets and then one or two or three years later, when it suited them and it was convenient, they would move those into foreign markets,” explains Michael Tomczyk. “Jack cut his eyeteeth selling calculators in Czechoslovakia, Poland, and England and he knew how to simultaneously introduce computers to Europe and Asia, as well as the US.” Judging by the response to the C64 at CES, Tramiel and Spencer felt Commodore should produce large numbers of the C64. “We looked at the volume we thought we could sell and what Jack was prepared to commit to production,” says Spencer. “It was a pretty large number for computers at that time. We were considering up to one million units in a year, which was more than the whole market size to date at that time. I personally told Jack I thought

we could sell that many.” Spencer told Tramiel, “If you are going to try and sell that many, we need do large advertising.” From his earlier days in electronics, Spencer used the 6% rule, meaning he tried to spend 6% of product revenue on advertising. At retail prices of $595, Commodore would earn a $200 profit on each C64 sold. According to Spencer, “When you get a big volume it becomes a big number. We said, ‘If this is what we are committing to in production, this is potentially what we need to spend.’” Tramiel was not in immediate agreement with Spencer on the advertising budget. “Jack fought tooth and nail because he thought it was too high,” recalls Tomczyk. “They tried to convince him that we should do 12 million dollars because 6% is the normal advertising budget.” Spencer knew it would be worthwhile to convince Tramiel. “Often you would have quite an argument with Jack about something and he was just testing out how you really felt about it,” he says. “You would make your point with him and he would be the devil’s advocate, but very strongly.” Al Charpentier witnessed the pains endured by marketing people at Commodore. “One of the things that was difficult for anybody who claimed to be a marketing guy at Commodore was that Jack felt he was the premier marketing guy,” he recalls. “I think Kit knew more than Jack would let him do. It was tough, because Jack had a sense of what he wanted to accomplish and the people around him just had to execute that rather than being able to leave their own mark on things.” To some people, it almost looked like Spencer was on the receiving end of a Jack Attack. “It might look like you were having quite an argument,” says Spencer. “We all got into shouting matches with Jack. The worst thing was if you didn’t disagree with him. A lot of people never understood Jack.” Spencer decided to start conservatively with advertising and increase the percentage as sales increased. “We started off at 2% to 3% of revenue and crept up to 5%, which is not that high in the consumer electronics business,” he says. The next day, Spencer called Tramiel and made sure he approved of the advertising plan. “I always found all I had to do the next day was call up Jack

and say, ‘Jack, I still feel what I said yesterday is correct.’ Nine times out of ten he said, ‘You go ahead then.’ He just wanted to know two things: one, you believed it, and two: he had to reasonably trust your judgment as being genuine, not political.” He proposed an advertising campaign that would cost the company millions of dollars in initial expenses, and grow from there as sales responded. It would be the most expensive advertising campaign in Commodore history. “It was one of the very major expenditures of the company at the time, probably the biggest for any one manager, so we had to get it right,” he says. William Shatner’s contract with Commodore ended in September 1982 and Spencer chose a different approach. “We moved to making the product the star of the advertising rather than endorsement advertising with William Shatner,” says Spencer. With the success of Star Trek II: Wrath of Khan and the popularity of T.J. Hooker, Shatner would command much higher fees than in late 1981. Spencer decided to go with a higher-caliber New York ad agency known for their bold commercials for Volvo and Hertz. “Ally & Gargano were extremely good. Very creative,” says Spencer. “I’ve worked with a lot of advertising agencies through my years in the markets and they were probably the best I worked with creatively.” Spencer worked closely with the creative people at Ally & Gargano to develop a strategy. “While the VIC-20 was ‘a great game machine you can also learn computing on,’ the Commodore 64 was, ‘a great computer that also plays great games.’” In order to build brand recognition, Ally & Gargano suggested music people would learn to associate with Commodore. The medley that would become so strongly associated with Commodore was Two-Part Invention #13 by Johan Sebastian Bach. Spencer knew the Commodore 64 measured favorably against the competition in nearly every aspect. He decided to go head to head with each of the most popular computers of the day: Apple II, IBM PC, Atari 800, and TRS-80. Because the engineers gave the C64 so much power, he could remain truthful and straightforward. To narrate the commercials, Ally & Gargano secured the sardonic voice of

Henry Morgan. “Henry Morgan was the guy on the old What’s My Line, a game show in New York,” recalls Tomczyk. “He had this incredibly ironic voice that just was so cool. He was the first sarcastic voice used in product advertising on television, and he really sold the hell out of our product.” Spencer also focused on the VIC-20, which was still a hot product in 1982. With the VIC-20 going into its second Christmas season, Ally & Gargano produced several VIC-20 commercials to keep sales moving. The agency devised a series of diverse and wickedly funny commercials around the concept that you could buy a real computer for the price of a video game machine. The target was obviously the Atari VCS, with commercials pointing out that it was not a real computer (despite the name Video Computer System). Each of the ads stressed the educational aspects of the VIC-20. * * * By the summer of 1982, Commodore had released 18 bi-monthly issues of Commodore the Microcomputer Magazine. However, the magazine focused on productivity and the PET computer. Kit Spencer wanted something to target video game entertainment, so he started a separate magazine called Power/Play. On July 1, 1982, the first quarterly issue of Power/Play was available for subscription or on newsstands, at the same time as rival Atari released their house magazine, Atari Age. Neil Harris supplied articles for the fledging Power/Play magazine. However, he spent most of his time away from his young family. “I was working in my second year at Commodore,” he recalls. “I spent probably half of my time on the road either working with our sales VP and doing demos for the head buyers at places like Kmart, Montgomery Ward, or Sears. Or I was traveling around doing training.” The lifestyle was hard for the young father. “I was a young guy with a young family; I had kids and was away all the time. I thought, ‘This is the time in my career when I should think about being a manager with people reporting to me and not just a guy.’” Around the same time, the magazine editor, Paul Flemming, departed Commodore. Harris saw an opportunity to work closer to home. “The magazine

was mostly full of rah-rah stories about why Commodore was great,” he recalls. Commodore initially used the quarterly magazine to promote in-house games. They also accepted advertising and promoted third party games, though articles favored their own creations. Minor Commodore games received cover status and eight page treatments while the magazine overlooked popular titles like Frogger and Q-Bert. Harris disliked the favoritism of the articles. “People won’t pay for that,” he says. “It’s nice that we want people to think that, but the opportunity here is that there is a big market for games and software of all kinds and computer peripherals.” At investor meetings, Harris often heard negative comments about the magazine. “The magazine was a very small circulation magazine,” he says. “The shareholders were like, ‘What are we wasting our money on this thing for?’ I said, ‘If we actually add content and make it worthwhile, it will be a great thing.’” Harris wanted to refine the magazines and make them more profitable. “So I approached Kit Spencer and basically said, ‘I think we can take the company’s magazines and make something out of them—make them a real profit center; make them something we can be proud of instead of just being a house organ.’” It seemed like a brash suggestion, considering Harris had little editorial experience. “I had done some amateur publishing in my earlier life that frankly wasn’t anything like preparation for doing a commercial magazine, but it was better than anybody else around had,” he says. Despite his weak resume, Spencer knew that a dedicated employee like Harris could be the best thing for Commodore’s magazines, and he thoroughly ed making every division of Commodore a profit center. “Kit ended up hiring me to be the department head for that group,” he says. “I worked with Kit and found him to be really smart.” Harris was something or a protégé of Michael Tomczyk, who was happy to see him rise in Commodore’s ranks. “He was a genius level marketer,” says Tomczyk. “He had an incredibly easy management style that allowed him to work with me and the product team. He was so charming and personable that he got along with everybody and yet he was incredibly intelligent, savvy and

creative.” The newly minted department head had an idea to increase magazine subscriptions substantially. “Compared to third-party magazines, we had a unique competitive advantage, which was that we controlled what went in the box with every computer,” explains Harris. “I said, ‘If we just put a subscription card in every box and offer people a reasonably priced subscription to a magazine that was actually decent, and if we worked with various distribution companies and got this magazine out on the newsstand, then this would be something we could be proud of.’” The inspiration for his subscription-in-every-box idea came from his handling of Commodore’s warranty registration cards. “I had been put in charge of processing data from surveys you put on warranty registration cards in the box,” he explains. “Believe it or not, people would send in these warranty cards, and they would be put in boxes and shipped on an airplane to the Philippines. People in the Philippines would key them in and the magnetic tapes with all the data would be shipped over to the United States for processing so we could get the data. That was actually cheaper at the time even in the eighties than doing it here. We were getting huge quantities of these cards back.” The sheer volume of returned cards inspired Harris’ subscription plan. “That was kind of the insight for me,” he says. “This box is powerful. You’re reaching directly hundreds of thousands or millions of people, so you should use that box to sell things to people.” Harris decided to include a magazine subscription form in the box. “Here you’ve got your brand new computer, let’s sign you up for a subscription to a magazine that s our computer and is interesting,” he says. “That was the basic concept.” Harris was impressed that Tramiel made sure to train him enough to run the magazine as a profit center. “As the business started growing, I had no formal training in business. I had no ing experience. I was a good techie and I had an idea of how to make it work and make it sell but I couldn’t really be a business manager at that point.” Tramiel called Harris into a meeting with one of Commodore’s senior financial executives and told him, “I want you two to work together and I want Neil to

learn how to do proper P & Ls (Profit and Loss Statements) for this business so we can measure and monitor the business.” Tramiel’s dedication to the small operation impressed Harris, and he never received a Jack Attack. “You know, really I was a small part of the business but he was very pleased with the fact that I took these magazines and turned them from something that was losing half a million dollars a year to something that was making money.” * * * Although Commodore ran their computers at one megahertz, Charles Winterble claims they were able to achieve faster speeds with the 6502. “We thinned the oxides down and we intentionally did some very high-speed versions of that,” he says. “One thing MOS [Technology] had going for it was a really good quality oxide. By thinning these gates down, we were able to drive it faster and faster.” The high-speed versions had a lower yield. “Then we did selection, in of our testing, to pull out the fast ones,” says Winterble. “We actually made a couple of really hot processors for a chess tournament for somebody. He literally water-cooled it, and he ran it at something like eight megahertz. It was just ridiculous how fast he ran it.” Unfortunately, even though MOS Technology could produce faster chips, it was not possible to simply drop in a faster processor. The reason had to do with RAM speed. “That would never have worked,” says Winterble. “The speeds we had to operate at had to be a multiple of the video rate. The RAM timing wouldn’t hold up at that. With the video and the processor alternating on the bus, you’re basically running the memory at twice as fast as it would normally be run.” In addition, the costs of producing parts for a two-megahertz Commodore 64 would drastically raise the price to customers. “The other problem is that the cost goes up exponentially because not only does the processor have to run faster, but the RAM and the ROM have to run faster,” explains Winterble. “It isn’t even linear; it’s exponential. It would be too expensive.” Surprisingly, Commodore’s microprocessor speeds did not immediately hurt sales of its products. In 1982, consumers were generally not aware of processor speed since most were taking their first steps in the computer world. To them, it

was much more valuable to be cheap than fast. However, Commodore’s engineers realized they would need to keep pace with rival microprocessors from Intel and Motorola. In attempting to develop the B256 computer, they began to reach the same conclusion Chuck Peddle realized earlier—the 6502 was no longer up to the task of powering a business computer. Chuck Peddle originally planned to follow-up the 6502 chip with a more powerful version. “The 6502 was never intended to be an end,” he says. “The 6502 was to get us into the market selling cheap controllers, but we had always planned on a 16-bit version.” The success of the PET sidetracked his plans. Synertec began development on a 16-bit 6502 they called the SY6516 in 1978, but the project evaporated when the company folded. In 1982, MOS Technology began working on a 16-bit processor. “There was a 16-bit 6502 program,” reveals Charpentier. In August 1982, Byte printed rumors that Commodore had working prototypes of their new family of 16 and 32-bit microprocessors. Then in November 1982, they began calling the new chip the 65000, shadowing its closest rival, the Motorola 68000. The article even speculated Commodore would use initial production runs of the processor in a new computer of their own. “We tried to do the same thing at MOS with a few people there,” says Winterble. “We didn’t start until ’82 or so. Probably as they started working on it they let it leak out for publicity sake.” Leading the design of the chip was one of Chuck Peddle’s former engineers. “It was Will Mathys,” says Charpentier. “He was part of the original [6502] team.” Robert Russell also recalls Mathys at MOS Technology. “He was working on 16bit designs and consultation with Commodore, but nothing really came of it because we were too busy with the VIC product line,” he says. Although Winterble was the engineering manager of MOS Technology, he had little enthusiasm for the project. “It was at a difficult time, now that the C64 was done,” he says. “We hired back the architect who did the original 6502, who was local up there, and we brought in a couple of other guys and we put together a team to do that. But about then was when Al and I were less enthusiastic, and we started thinking about where our future’s going to be.”

Winterble felt the team should keep the 16-bit 6502 as simple as possible. “We allowed the creative people to go ahead and do their freewheel thinking about fancy new architectures and stuff like that,” he says. “[They discussed] different approaches to doing a processor. I was not in favor of this too much. I just wanted to do essentially the same thing [as the 6502], but just 16-bit. Then the issue became, well why bother doing that?” According to Winterble, the project team lost focus soon after forming. “It got to the point where they were doing some layouts, but the guys who were doing the architecture were sort of undisciplined people,” he explains. “After a certain point it became, ‘This is not going to fly.’” Normally, if Tramiel pushed for a product, his engineers became highly motivated and completed the product. However, Tramiel seemed indifferent to a 16-bit processor. “There was no push at all,” says Winterble. “We weren’t getting anywhere and it never really got going.” The 16-bit project eventually faded away at MOS Technology. “It just never got any kind of in the organization,” says Charpentier. “To be in the microprocessor business, you have to sell microprocessors into the OEM market. Commodore had never developed an OEM semiconductor market. They were a systems house that developed computers with it. It just wasn’t a focus in the company.” Lack of funding also hindered the 16-bit 6502 project. “I think the amount of investment they put into it wasn’t what a chip company really needed to stay competitive,” says Commodore engineer Dave Haynie. “Microprocessors were just not in the cards.” The lack of a 16-bit processor severely hurt Commodore’s chances of remaining competitive in the future of computing. * * * In August 1982, Commodore finally produced enough C64 computers to start making small shipments to dealers. “They started selling them through computer retail stores like ComputerLand and places like that,” says Bob Yannes. For now, Kmart and other mass-merchandising retailers would sell the VIC-20. The sparkle bug continued appearing in some computers, even as Commodore

shipped. Charpentier focused his search on the VIC-II video chip, but it appeared to be working fine. Finally, he went to his logic analyzer and eventually tracked the bug down to a ROM chip. The production facility developed comically primitive testing procedures for the computers. “We had to come up with our testing techniques,” explains Winterble. “How do you color back in those days? For a while there, we had operators standing there in front of test sets looking at the colors saying, ‘Oh yeah, that’s blue. Oh yeah, that’s blue.’” Al Charpentier read early reviews of the C64 and was surprised to hear it criticized for its garish colors. This seemed odd to him because he thought the C64 colors were vibrant and well chosen. He looked further into these criticisms and found out the assembly line workers were adjusting the color output on the circuit board until they saw color. As a result, many just turned up the color saturation all the way, resulting in overpowering colors. The inadequate testing of the C64 soon became apparent to others outside Commodore. As an Epyx programmer later reported, “I’ve opened up brand-new Commodores and found traces cut. They obviously use a power screwdriver to assemble the C64, sometimes miss the screw, and chop the traces. How, you might wonder, could that have ed the final inspection? Well, those traces are hooked up to the disk-drive connectors, which they obviously don’t test.”[1] The lead production engineer for the C64 was David Ziembicki.[2] “We weren’t board people,” its Winterble. “We made mistakes trying to save money, like trying to use the same tooling as the VIC-20 case.” In the end, Commodore ended up spending more money than they would have if they retooled a larger case. Adding to Commodore’s woes were house fires started by the Commodore 64. Robert Russell saw the results of the fires first hand. “I had to go do fire investigations as part of my career,” he says. “It wasn’t the VIC-20 so much as the Commodore 64 that I investigated. The Commodore 64 was much worse because the power supply was much more overloaded. It burned down buildings and houses because people had managed to put their draperies over top of the power supply.” Yash Terakura was satisfied with the result, considering the rushed

circumstances. “I think for that price, we did a pretty good job on that one. It was a pretty good entry level computer. It could have been better of course. We could have done a better design of the power supply and we could have made it a little more flexible in some areas, but for the price it was pretty good.” According to Bil Herd, when production lines encountered a shortage of VIC-II chips, technicians went to the bin labeled “defective chips” and took the sign off it. “The idea we had was trying to make it stick under the tree,” says Herd. “Meaning that we could sell all we wanted in December, but they came back in January.” Commodore hoped their customers would exchange the computers for working ones. With so many problems, it looked like the Commodore 64 would be a disaster. However, the team had managed to work through the challenges. As Ziembicki later recalled, “The key is to be able to solve your problems while you’re running.” * * * By the fall of 1982, Ally & Gargano had a collection of ads ready to air on prime-time television. The ad agency knew Tramiel was a thrifty CEO, so they offered Spencer a suggestion. “Our advertising team had a great idea. Once in a while, rs pull out of ads or cancel ads at the last minute, like airlines,” explains Tomczyk, referring to standby seating. Commodore could purchase cheap airtime using last minute advertising space. The thrifty strategy fit Commodore perfectly. “They committed a large chunk of our advertising budget to buying that time,” says Tomczyk. “We doubled our advertising exposure by using that strategy. We were really lucky to have that high powered advertising team.” Soon, ads began appearing alongside programs like Happy Days, Magnum, P.I. and Dallas. “Commodore may have been the first one to run television ads for a computer in prime time,” claims Yannes. In one VIC-20 commercial, a thirty-year old burnout sits in front of a fishbowl pretending to play a video game, staring eye to eye with his goldfish. The meaning was clear. Many parents feared video games did little to expand intelligence.

Bob Yannes approved of the educational angle. “Part of Commodore’s marketing program was, ‘Do you really want to buy your kid a game console that is going to rot his mind, or buy him a computer, which by the way can play video games too.’ That was a very successful marketing campaign,” he recalls. Given the climate of uncertainty towards computers in the early eighties, and the unproven record of computers in education, some California educators felt the commercials were disingenuous. “A lot of people objected to them,” says Dave Haynie. The first thirty-second commercial to introduce the C64 stressed memory and price. The ad presented the computer on a black pedestal while Henry Morgan informs the viewer it contains 64 kilobytes of memory for only $595, compared to the IBM PC with 16 kilobytes for $1565, the Apple II with 48 kilobytes for $1530, and the Atari 800 with 16 kilobytes for $899. It ends with the prediction, “The Commodore 64 could be the microcomputer industry’s outstanding new product since the birth of this industry.” The ad agency also created a series of short but potent commercials targeted against specific computers. These commercials were cheap but allowed Commodore to stay in the public consciousness without much expenditure. The first ten-second spot against Atari claims consumers would have to buy four Atari 800 computers for $899 each to get the same amount of memory as one Commodore 64. The commercial against Apple was simple but visually effective. It showed a green apple that gradually disintegrates as someone takes bites out of it. The soothing voice of Henry Morgan says, “There’s a new personal computer from Commodore. With a third more built-in memory than the Apple II at half the cost.” By the end of the commercial, all that is left is an apple core, which falls over. The IBM ad was much in the same vein as the Atari ad, only it used animation. It shows a lone cartoon elephant while Morgan asks, “If this is the memory of the IBM PC for $1565, then what’s this?” Three more elephants squeeze into frame, jostling for position. “The memory of the Commodore 64, for under $600.” In the final thirty-second commercial for the series, Spencer turned the other

computers against their own companies. It shows a BASIC code listing from an Apple computer, and the announcer explains, “We asked the computers which was better on the basis of price and memory.” Of course, the Apple II, IBM PC and TRS-80 all choose the Commodore 64. It concludes sardonically, “That’s what we like about our competition; they’re so honest.” The campaign gave Spencer a sense of satisfaction. “Ally & Gargano really got it right, better than I would have even envisaged,” he says. “They took the concepts and they did a superb job.” Over the holiday season, viewers came to identify Henry Morgan’s voice with Commodore. Perhaps the most memorable part of the commercials was the background music. Ally & Gargano licensed music from Switched on Bach by Walter Carlos[3], which featured Two-Part Invention #13 by Johan Sebastian Bach. The catchy tune presented the perfect impression of technology, sophistication, and playfulness. Viewers who heard the commercials could not stop thinking about the tune and ultimately could not stop thinking about Commodore. Viewers came to identify the tune as the Commodore theme song. Spencer also produced radio jingles, creating the famous ‘I Adore my 64’ theme song. The commercial jingles were as memorable as those by McDonalds and Coca-Cola. The first song sounds suspiciously like a Beatles tune:

I adore my 64, My Commodore 64.I sing with it, write with it, figure my path to flight with it, My Commodore 64. I rate with it, create with it, telecommunicate with it, My Commodore 64. The other jingle used slightly edgier music:

I adore my 64. My Commodore 64. I cook with it, steam with it, educate the team with it, My Commodore 64. I pitch with it, catch with it, count my chicks before they hatch with it, My Commodore 64.

None of the advertising touted the superior sound and graphics capabilities of the Commodore 64 since the in-house software from Commodore in late 1982 did little to exploit the abilities of these chips. Consumers would have to discover these capabilities themselves. Spencer also produced print ments for magazines. The first ment ran the headline, “When we announced the Commodore 64 for $595, our competitors said we couldn’t do it. That’s because they couldn’t do it.” In contrast to the television and radio advertising, the print advertising contained technical details. Spencer emphasized price, memory, software, peripherals, games, graphics, and sound. The ments also promised an optional /M cartridge, a promise that would eventually land Commodore in legal trouble. In total, the advertising campaign produced over 12 different television commercials, radio spots, and magazine ads. Commodore ran the ads intensely before Christmas and then played them over the next year. Surprisingly, Spencer exceeded his initial $12 million budget with Tramiel’s blessing. “The total budget in the first year of launch was over $20 million,” says Spencer. “As percentage of revenue, we didn’t have a particularly high advertising expense in consumer marketing norms.” Ally & Gargano’s advertising campaign was critically well received, taking both

major awards for advertising—the Clio and the Effie. “We managed to win both in one year, which was pretty amazing,” says Spencer. The campaign produced Clio awards for the Commodore television commercials and the radio campaign. “We won a Clio, which is the advertising industry for creative content,” says Spencer. The Effie was the more cherished award, since it focused on results. “We also won an Effie, which is the advertising industry award for effective advertising,” says Spencer. “They are the Oscars of the advertising industry.” The leading industry magazine, Advertising Age, also named Ally & Gargano the agency of the year in 1982. Kit Spencer’s campaign did exactly what it was intended to do: it got the message out. Spencer had chosen an effective strategy with his bold comparisons against the competition. His commercials, and Commodore’s pricing, would soon bring Atari, Radio Shack, and even Apple to their knees. It was a golden age for Commodore 64 advertising. “The VIC had taken massive market share away from Atari, which is why they ended up in such a mess. The strategy for the 64 was to attack the USA computer market that Apple had at the time head on,” says Spencer. “And just like the VIC, it was successful. We took away large market share again and it caused Apple problems at that time.” Unfortunately, they would never recapture the same level of media saturation attained by Spencer. Spencer credits much of the success of the advertising campaign to the engineers. “The Commodore 64 was a great product and became outstandingly successful. If Charlie and those guys hadn’t created a great product, it wouldn’t have mattered what we did moneywise to promote it.” However, Neil Harris believes none of Commodore’s products would have been quite as successful without Spencer. “I believe that Kit Spencer is the one that put Commodore on the map in the mass market,” he says. * * * After the departure of Chuck Peddle, Tramiel was uncertain if he had capable engineers to design his computers. These fears quickly dissipated when his MOS Technology engineers created the most successful Commodore product yet.

Unfortunately, the engineers felt strong dissatisfaction within Commodore. Bob Yannes still felt bitterness from the reneged bonus. “That was just something that stuck in everybody’s craw,” he says. “It was one of the points that helped me realize that being at Commodore wasn’t likely to get me very far up the ladder of financial success.” Due to the success of the Commodore 64, the engineers expected a bonus for their work. Despite delivering an exceptional product, Tramiel was not happy with the engineering delays. “We were supposed to have it done in three months and it took us six,” recalls Russell. “I don’t even think any of us got a bonus for the C64 because it was late according to Jack.” No one needed a bonus more than Yannes. “My salary wasn’t particularly high,” he says. “They did give me a bonus for the VIC-20 concept, which was pretty nice, but I never got any kind of bonus for the Commodore 64 for all the extra work I put in. That was a little bit of a slap in the face.” Tramiel compensated Charles Winterble and Al Charpentier well, but they began wondering what to do after the C64 project. “We weren’t dissatisfied,” says Winterble. “Quite the contrary, we were all excited over the fact that now [the C64] was going to be sold, but our role was no longer as important as it was before. Our role was more like fixing production problems. Who wants to do that? We were a creative design team and we wanted to continue with the creative design stuff.” Both Winterble and especially Charpentier knew what they wanted to do next, but they no longer had the freedom they once enjoyed. Tramiel wanted a product to make use of Commodore Optoelectronics. “Jack wanted us to do a handheld CMOS computer,” recalls Winterble. Although a handheld LCD computer was an innovative concept, the engineers had no enthusiasm for the proposal. “There was nothing lined up that was going to be as exciting as what we were working on.” Charpentier’s approach to engineering was fundamentally different from Tramiel’s philosophy. “The real market to develop was the business market,” he says. “Jack always felt that low-end was more appealing. That’s where he lived.” Charpentier approved of the low-end market, but he wanted to go after the highend market where technological innovation occurred. However, Tramiel wanted

an even cheaper computer like the Sinclair. “I wanted to take aim at Apple, but Jack saw it as a different market,” says Charpentier. “He felt [a cheap computer] was going to be more appropriate. It was just a disagreement of where things should go.” The disagreement between Tramiel and his engineers was eerily similar to his disagreement with Chuck Peddle. Charpentier put forward a plan for Commodore’s next machine. “After the Commodore 64, my next proposal was a C80, if you will,” reveals Charpentier. “A Commodore 80 with 80 columns to compete head on with the Apple II. Jack Tramiel and I had many, many, many discussions on that.” As a video chip designer, Charpentier knew which features he wanted to add to the next VIC chip, the VIC-III. “The next thing I was going to do was create variable sized sprite structures,” he explains. The Commodore 80 would be unable to connect to a standard television. “Since TVs can’t do 80 columns, it would have had to have been a separate CRT monitor,” says Charpentier. “It was going to use a higher speed floppy, an 80column display, and up to 256K of RAM.” Charpentier would have attempted to make the computer backward compatible with the C64. “It probably would have been a sensible thing to do, but it never got to see the light of day,” he says. Despite the recent success of his engineer’s designs, Tramiel overruled him. “It was clear that Jack was going to be calling the shots,” says Charpentier. “He did a reasonable job in his own company, but he didn’t understand what was going to happen with the computer market. I knew that the [cheap computer] was the wrong choice. I tried to convince him otherwise. He didn’t agree, and that was his shekels.” With his engineers rebelling once again, Tramiel would have to look elsewhere to find someone to build his handheld computer. Charpentier believes Commodore was lacking top-level executives who knew about computers. “The only people that were computer s were the engineers and the geeks, and they weren’t in management,” he says. Bob Yannes also felt management did not give the design team much respect for their past successes. “It was pretty clear that in the future, any products we did were going to be predefined for us by marketing,” he says. “We wouldn’t have

any freedom to do what we wanted.” The worst insult to the engineers came in the midst of financial success for Commodore. With his yearly performance review approaching, Yannes expected a raise for his efforts on the C64. “They were having such an enormous growth in their earnings and stock value,” he recalls. “When we were to have our reviews and get our salary increases, they announced that salaries would be frozen.” To the young engineer who had done so much to create Commodore’s success, it was too astonishing to believe. The explanation was even more astonishing. “If they paid us more money, it would weaken the curve of the profit growth of the company, and they didn’t want the financial analysts to see the curve leveling off,” explains Yannes. “They wanted to keep it looking like it was going up and up.” Yannes is unsure who decided to freeze their salaries. “I wouldn’t be surprised if it came from Jack,” he says, but given Irving Gould’s past obsession with the stock price, it seems more likely it came from him or his board in New York. Whether it was Tramiel or Gould, the shortsighted attempt to help the stock undoubtedly weakened the long-term prospects of the company. The financial success of Commodore was due in no small part to the efforts of its engineers. Now, because of their astonishing success, the engineers received financial punishment. It was intolerable. “We were supposed to understand that? ‘Oh yeah, that’s great! No problem,’” says Yannes. “That helped motivate me to leave.” When Charpentier called Yannes in for his review, he presented a new opportunity. Charpentier said, “Here’s the bad news: your salary is frozen. Here’s the good news: I’m starting another company. You want to come me?” Yannes was ready to accept. “I think we all realized we were instrumental in the Commodore 64,” he says. “We were the marketing team and the design team and everything. We really didn’t have any outside influence over us and it was pretty much a home run. So we felt, ‘Why are we doing this for someone else?’” Winterble initiated the plan to leave Commodore. “I had an idea for a product and I talked to Al,” he says. “Al and I joked over wine about the million-dollar

idea. We talked about what we could do in the way of a product that we could do independently [of Commodore].” With the Atari VCS at its peak in 1982, Winterble conceived of a new product to expand the system. “We had an idea to turn that into a computer, called My First Computer,” he explains. “It was essentially a membrane keyboard that fit into the VCS slot and we put BASIC in a little ROM. For $29.95, instead of buying a ROM cartridge, you buy a little computer. It was a great idea.” Winterble brought his idea to Atari. “All we had was this vague idea,” he says. He presented his idea to Atari CEO Ray Kassar, who agreed to purchase the product for a million dollars. Atari would develop the software themselves. Charpentier barely hesitated when he left Commodore. “Once Jack had said he wanted to do the [cheap computer], I said I’m out of here,” he recalls. “I felt very good about our decision. Opportunities were presenting themselves and it was a pretty exciting time back then.” Bob Yannes found it difficult to leave Commodore. “I actually had a lot of angst in leaving,” he says. “I enjoyed working there. It was a lot of fun.” Yannes eventually decided his opportunities were greater with Charpentier and Winterble. “We were gone in late August, early September of 1982.” To avoid the wrath of Jack Tramiel, the engineers made their escape while Tramiel was away on a business trip, scouting for someone to manufacture his LCD computer. “We intentionally did it when he was in Japan,” says Winterble. “One of the concerns was whether they would beat up on Yannes. We figured, ‘We need some breathing room. Let’s do it without people being able to jump all over us.’” Winterble attempted to recruit other Commodore engineers. Due to Andy Finkel’s programming skill, the engineers felt he would be a valuable addition. “They offered me the chance to go with them on that project, but I was happy on the east coast,” says Finkel. They also asked Robert Russell. “They asked me to go along with them,” he recalls. Russell knew it was a good chance to earn substantial wealth. “They basically got their million dollars if they got that thing done.” As Russell recalls, the plan called for the engineers to develop the hardware

while Atari would deliver the computer software. He realized there was little need for a software engineer like himself. “I would have gone with them if I thought they needed me,” he says. Reluctantly, Russell declined the offer to their startup, saying, “I’ll come you sometime in the future. Let me know when you’re doing software.” Despite the promise, he never ed. Russell felt devastated at losing with his best friend. “You’ve got to realize, Bob Yannes and me were really good friends,” he says. “We’d go out for lunch together practically every day.” In total, five of the six key of the C64 project ed Winterble, including production engineers David Ziembicki and Bruce Crockett. The sudden departure of Charpentier left many of the VIC-II chip features undocumented. “We knew that eventually we would want to get more of that [information] out, but we ended up leaving before I could write the definitive manual on how to do some of the tricks,” he says. [1]IEEE Spectrum journal, “Design case history: the Commodore 64”, (March 1985), p. 48. [2] David Ziembicki later invented a sensor to detect obstacles in the path of vehicles while backing up. He subsequently filed lawsuits over the patent. [3] Walter Carlos also composed the score for the movie Tron.

CHAPTER 30

Encore 1982

After the departure of several key engineers, MOS Technology had a problem. How could it continue advancing sound and video chip technology without chip designers? “There weren’t that many engineers,” says Robert Russell. “Bob Yannes and Al Charpentier were the main chip engineers.” Charpentier believes Commodore engineering possessed talent but they were missing a long-term strategy. “They had some designers, but it’s the vision that they lost,” he explains. “It was just misunderstanding the market. In 1981, the PC comes out and the market changed drastically, and Commodore didn’t recognize the opportunity.” With Jack Tramiel’s most trusted engineers gone, Robert Russell and exCalifornia engineer Shiraz Shivji moved up in the ranks. Tramiel was now so busy managing the company that he rarely visited his engineers. Instead, he relied on Shivji to carry through with his instructions. “Shiraz was his mouthpiece back then,” says Bil Herd. The two engineers met with Tramiel to decide their next product. “Me, Shiraz, and Jack were working on how to go forward with engineering at that point in time,” says Russell. “I was in a lot of the meetings, strategizing, trying to come up with it.” Robert Russell acknowledges that he did not have visionary aspirations as an engineering manager. “My whole goal as an engineer was always to accomplish a product, not to be the furthest out into the future,” he says.

Three projects came out of those meetings: a handheld computer, a portable computer, and a computer to rival the new Sinclair ZX Spectrum. Neither of the computers challenged the IBM-PC. As was typical with Tramiel, he wanted them shown at the January 1983 CES, mere months away. Tramiel still wanted a handheld computer to compete with the Radio Shack PC-1 (Pocket Computer), which was in fact a clone of the Sharp PC-1211. The Pocket Computer was very calculator-like, and Commodore had the infrastructure needed to build such a device. When he was unable to convince Charpentier to develop a handheld LCD computer, he departed for Japan to see if he could by his reluctant engineers and obtain the product he desired. Tramiel returned with a Toshiba IHC-8000. The tiny computer looked like a calculator, with a single row of 24 characters on the LCD and a tiny rubber QWERTY keyboard. He rebranded it the HHC-4 (Handheld Computer), and replaced the Toshiba decal with a Commodore logo in order to display the product at the CES. In the past, Tramiel reacted to the market, often making decisions based on what his competitors sold. In 1982, the Osborne 1 portable microcomputer was selling well, and with it, the Osborne Computer Corporation began remarkable growth. Tramiel decided he wanted his engineers to produce an Osborne-killer. With little time remaining before CES, the engineers decided to piece together a prototype using a stock C64 motherboard, a small color CRT (Cathode Ray Tube) monitor, and a disk drive. The third computer devised by their meetings was borne out of the high failure rate of VIC-II chips. “There was a problem on the early Commodore 64s,” recalls Neil Harris. “The graphics chip in that computer was fairly complex for its era and they were getting a very low yield.” The low yield made it costly to produce the C64. “What happens is, you have a die and on that die you print a whole bunch of chips, and then you punch them out put them in a housing and test them,” explains Harris. “Some percentage will work and some percentage won’t. So if you get a low yield it means the cost per chip is much higher on what you are actually getting into production.” Commodore was about to embark on a strange path. Tramiel decided to by the problem completely by creating an easy-to-manufacture computer.

According to Harris, “The engineers said, ‘Why don’t we make a simpler graphics chip where we’re likely to get a much higher yield and we’ll build a computer around that.’” Tramiel wanted the computer to compete with a new Sinclair color computer, the ZX Spectrum. The mere fact that a cheaper computer existed worried him. “I was involved with that project real early on because I helped specify it,” recalls Russell. “We were supposedly doing an anti-Sinclair device, trying to beat out the color Sinclair. It was supposed to be a less than $100 computer to compete against the Sinclair. It had to be made for $50.” According to Russell, software compatibility with the C64 was not in Tramiel’s mind. He viewed computers as disposable. “It wasn’t to be compatible or the next generation of C64,” he says. “They were thinking about it, but Jack was still like, ‘They’ll just throw away the computers.’” Tramiel’s failure to realize the importance of software compatibility was about to fail the company. “Compatibility was the catch 22 of the computer industry back then,” explains Kit Spencer. “To make the quantum leap forward, you often lose compatibility and you need to make the quantum leap forward sometimes to make the breakthroughs. On the other hand, one of the biggest things going for you with computers—which is the difference between calculators—is software.” Computer owners spent a small fortune building up a software collection, as well as creating programs and documents. They did not want to start their software library from scratch by buying an incompatible computer. “If you don’t capitalize on an existing base, you are losing out on a tremendous marketing opportunity,” says Spencer. “If you’ve got a big installed base, you have a big advantage with your product over everybody else.” Tramiel wanted a Sinclair-killer as Commodore’s next major release, but the obvious follow-up would be a backward compatible C64 with more memory, a faster U, more sprites, more voices on the sound chip, higher screen resolutions and 80-columns. At least, that was what the market wanted. However, Leonard Tramiel disagrees with following the market. “If you ask the market what they want a product to be, they will never, ever, ever come up with something new because they don’t know what’s available,” he says. Tramiel felt confident a cheap computer was right for the market. He ignored

previous objections over the VIC-20 and witnessed incredible success. The VIC20 had only 5 kilobytes of memory when Commodore had been selling 32kilobyte PET computers. He had little reason to doubt he could make another success. MOS Technology would develop a new video and sound chip for the computer. “At the time it seemed like a good decision so they started working on that graphics chip,” says Harris. MOS Technology began deg the low-cost TED chip, which stood for TExt Display. “We had a VIC chip, and this was the TED chip,” says Dave Haynie. “It did slightly better graphics, [no] sprites, and the sound was pretty basic.” Inconceivably, Commodore was now on track to build a computer with fewer features than their previous entries. * * * Following their clandestine departure from Commodore, the former C64 engineers began working on My First Computer for the Atari VCS. “After we got out of Commodore, we worked out of my basement and we worked with no money,” says Charles Winterble. Soon after forming their new company, Peripheral Visions, the ex-Commodore engineers received a lawsuit from Commodore. “Jack hears we’re doing something and he wants to know what it is, and he sues us,” recalls Winterble. “We’ve got papers coming in and it’s like, ‘Well, for what?’ ‘For stealing trade secrets.’ ‘Well, what exactly did we steal?’” The lawsuit did not even specify the supposed trade secrets. It was obvious Tramiel did not know anything about their project. Winterble made an agreement with the Commodore lawyers to disclose the details of the project. “If we told him what we were working on, if it was not anything that Commodore was working on, they would drop the lawsuit,” he explains. “So we said cool. We told them what we were working on. He said, ‘Sorry, that’s my product. You can’t have it. I’m going to sue you anyway.’” The lawsuit made little sense to Winterble. “We didn’t work on it at MOS Technology or anything like that,” he says. Although Tramiel may not have had legal justification to sue Winterble, he had

good reasons to be angry. Winterble, one of Tramiel’s trusted inner family, quite obviously looted some of his best engineers. Furthermore, the engineers went straight to a competitor, Atari. “There was a little revenge in there because he was mad, and I know he was mad about this,” its Winterble. “He was probably pissed at me because we executed this while he was not there.” The seemingly frivolous lawsuit disappointed Winterble. “We helped Commodore, and our reward is this?” he says. “It just hurt our feelings. I mean, you are allowed to leave a company! Slavery was outlawed long ago. Why should we be punished for something we had a right to do?” During meetings with their lawyers, Tramiel let his ex-Commodore engineers know exactly how he felt about them siding with Atari. “If you left, you were a traitor and Jack could be very vengeful even to the detriment of his company,” says Yannes. “I have a vivid memory of him sitting across the discovery table from us and growling how ‘this troika had betrayed him.’ If I had still been employed by him, I would have been frightened, but it was almost comical. It reminded me of Nikita Khrushchev banging his shoe at the UN.” Robert Russell found himself in an awkward position, torn between his loyalty to Commodore and his former coworkers. “I heard that [Jack] wasn’t happy and that there were some lawsuits they were trying to pursue,” he says. “I kept my head down about that whole thing because they were personal friends of mine.” The case should have been simple to decide, but in 1982, law courts were unfamiliar with technology issues. “If anyone was knowledgeable about technology it wouldn’t have lasted a second,” says Winterble. “In other words, if this had been filed out in California where you had people who were used to dealing with technology issues every day, it was a non-issue. The difficulty was that we had to educate a court system and that’s an expensive proposition for a couple of little guys.” Surprisingly, a few managers from Commodore Optoelectronics stepped forward to help Winterble prove that Commodore never had an Atari VCS computer project. “Their management, Elton Southard and a few other people, said that’s baloney,” recalls Winterble. “We had their deposition saying this was never worked on. Their people were agreeing with us.” Commodore lawyers even tried to claim that the Atari VCS project was

infringing on intellectual property merely because the keyboard contained a 6502 processor. “We designed a product that used the 6502 and Commodore owns the 6502, therefore we’re infringing on them,” laughs Winterble. “The 6502 is sold publicly to anyone who wants it. The data sheets are available to anyone in the world who wants to design with it. This is a good explanation that takes about five minutes—it takes three years in court.” Winterble eventually won, but it had the desired effect of slowing down Peripheral Visions. “By the time we educated the court system and worked our way through, it cost us about $300,000,” he reveals. Despite the harsh treatment from Tramiel, the engineers still valued their time with Commodore. “I enjoyed it a lot,” says Charpentier. “Jack was tough to work for, but in the engineering area he really gave the engineers a lot of freedom to be creative. I had a lot of fun there.” Charpentier still values the lessons he learned from Tramiel. “He was a hard charging guy who I learned a lot from,” he says. “I learned a lot of things not to do. I certainly wouldn’t want to emulate the way he did some things, but in of building a company, it was a good training ground for me.” Winterble has equal praise for his former boss. “Even today, I look back at Jack Tramiel and I thank him,” he says. “I learned a lot from him. It’s unfortunate the we separated weren’t the best in the world, but it was a great experience.” * * * The Commodore software group had been headed by Joe McEnerney at MOS Technology since early 1982. During that time, software development slowed to a trickle. “He took them over for six months and not one single game came out of the group in that six months time,” says Michael Tomczyk. “They always had weak managers,” says Russell. “They did a few games but they were not very good.” The most marketable games came from HAL Laboratory. The development of computers tended to distract the game developers. Andy Finkel, the lead developer within the group, was often in demand due to his skill at code reduction. “While games writing was fun, I was also kind of tempted by the idea of working on the operating systems of these things,” he says. “Because

I was really experienced with shaving bytes, I used to help out with shrinking down BASIC and the kernel to try and fit more stuff in.” Finkel believes the engineering department was not suited for game development. “In of getting games out, it wasn’t the right environment,” he says. “The game industry doesn’t work like the computer industry worked at that time. There was a lot less communication with end s in engineering, and a lot less communication with marketing, so it really wasn’t the best place for a game group.” Michael Tomczyk, the disgruntled former manager, needed a way to take out his frustrations. “There’s a little doll the marketing department used to keep on a shelf,” he says. “It was a little hollow plastic doll that looked just like Jack. People used to walk by and give it a whack. We called it the Whack-Jack.” When the Whack-Jack was not enough, he took his complaints directly to Tramiel. He told the CEO, “That’s one of the worst decisions you ever made. That was very insulting to me and look what happened: six months without a single frigging game. You shouldn’t have done that, and all because some stupid senior executive convinced you that he should be in charge.” “That’s one of the few times he itted he was wrong,” says Tomczyk. “He said, ‘You are absolutely right, Michael.’” After the C64 designers left, the software division bounced around from one manager to another, including Robert Russell. “It ended up me being in charge of them after their manager [Charles Winterble] left,” says Russell. “They were doing games that were really pretty good based on the VIC-20 and then translating them to the C64. For the amount of people they had, and I don’t think it was ever more than ten or so, they were pretty impressive.” Tramiel decided to make software its own division with professional management. “They hired a professional software manager and moved in a professional sales guy to start heading up a new Commodore software division,” says Finkel. “Jack decided to split it out and make it its own division, rather than being a profit center in either engineering or marketing.” The software group moved back to Commodore’s Wayne, Pennsylvania headquarters. Bill Wade, the man responsible for taking Commodore into the retail business, became the head of the new software division. “Bill was more of

a sales guy,” says Harris. “He had been with Jack Tramiel and Commodore for many, many years and had started the software business.” The new middle manager for the division, replacing Joe McEnerney, was an experienced software manager named John Mathias. The developers were happy to be in their own division, “even though it meant losing some of the fun things about hanging around with Al and Bob,” says Finkel. The ever-reliable HAL Laboratory of Japan produced many of the earliest C64 games. Commodore now understood the rules of copyright so there would be no more infringements. HAL Laboratory produced 1982 classics like Night Driver, Le Mans, Avenger, and Jupiter Lander. There was no disk drive available yet, so Commodore distributed all their games on cartridge. Of the four, Le Mans was the only one not appearing on the VIC-20 earlier. Like most HAL Laboratory games, it was a blatant rip-off of an existing arcade game, this time Sega’s 1979 arcade classic, Monaco GP. One of the most lucrative relationships for the C64 continued from the VIC-20 days. Commodore purchased the rights from Bally-Midway for many of their popular arcade games, including Sea Wolf, Clowns, Kickman (originally titled Kick in the arcades), Blueprint, Lazarian, Omega Race, and Gorf. For every game they produced, Bally-Midway sent an arcade machine, which collected in a special games room in Commodore. Robert Russell re his envy of the game developers, who played games on a daily basis. “They had a game room over on the software side,” he recalls. “They had all the latest arcade games. I don’t even know if I ever played any games in that thing, but I always wanted to.” Bill Hindorff coded Blueprint, a relatively new game released in 1982 that did not prove to be a big hit in the arcades for Bally-Midway. Hindorff’s version was a credible port of the original, however. Andy Finkel was in the midst of finishing the C64 version of Omega Race. Because the C64 graphics and sound chip was so different from the VIC-20, he had to recode the game completely, using very little of his original Omega Race code. Finkel was most proud of the sound design for the C64 version. “I designed a simple sound engine for Omega Race,” he says. “[Eric Cotton] would listen to

the original sounds from the arcade machine and then, using my little sound engine, make the sounds using the little software tool that was in Omega Race.” Another programmer named Jeff Bruette worked on the Wizard of Wor game for the Max Machine. Finkel also began development on a 1981 arcade game in the same mold of Gorf. “I did Lazarian, which was another Bally-Midway game,” he says. The US software division attempted to produce their own original game hits. The programmers based the first game on a failed premise from an early VIC-20 idea. “Garden Wars was about trying to grow a garden, and you’re being attacked by various bugs, and you have to use the right insect sprays on them,” recalls Finkel. “It was a different game. With a year of development time, it would have been a really interesting game.” However, the team decided to abandon the game and move on to simpler projects at the time. “At its first review, it just wasn’t fun enough, so we decided that we might as well shelve it and go on to work on something else,” he says. When the C64 came around with increased memory, they decided to give the concept another try. “Years later, we gave it a different design.” The new game, titled Greenhouse, had the same gardening concept. Another game, Tooth Invaders, was an attempt to appeal to younger kids by protecting a mouth from cavities. The programmers developed the games simultaneously for the VIC-20 and C64. Both Tooth Invaders and Greenhouse featured simplistic graphics and animation, and met with mediocre success. They also made Speed Math/Bingo Math, an attempt to tap into the education market. * * * In November 1982, Commodore received a new president, Robert H. Lane, to replace Jim Finke, who resigned the previous June. Lane started his career as product manager for Good Humor Ice Cream in Canada and eventually became company president of Northern Telecom (later Nortel) in North America and Europe. “Bob Lane came in as president, very much brought in by Irving, not with Jack approving of it,” says Spencer. In theory, Lane would run the day-to-day operations, while Tramiel gave Commodore the long-term corporate vision. However, with Tramiel’s hands-on

approach to running Commodore, the potential for conflict was strong. “That was the start of things going wrong and the beginning of the rift between Jack and Irving,” says Spencer. In late 1982, Tramiel decided software was a serious business and he appointed his friend Sigmund Hartmann to operate the software division. “Sig had worked with Jack previously,” says Neil Harris. “He had been out of the company for some period of time and was brought in to head up software. I know that Bill [Wade] was not really pleased by that decision.” Hartmann, a likeable yet shrewd businessman, was born in and educated in Belgium, where he earned a master’s degree in mathematics from Liege University. For 18 years, Hartmann worked at TRW Automotive, which included managing engineering projects for NASA and GM. With his thick German accent, Hartmann frequently apologized when Commodore employees could not understand him and often spelled out words to make sure people understood what he was saying. Under Hartmann, the newly structured software division would operate as an independent software company within Commodore. Tramiel wanted to turn the software division into a true powerhouse within the company. “Sig came in after we launched the C64 to build up the software division,” says Spencer. “We recognized that as a business we needed to build it up, but it’s a dangerous business because you can be stuck with inventory too.” Andy Finkel liked the new vice president. “Sig Hartman was great,” he says. “He was a California kind of guy. He’d come in with these open shirts with the gold chains around his neck. It was fun watching him work. He was a good person to work for too. He knew what he was talking about.” The software division contained three subdivisions: games, applications & productivity software, and marketing & documentation. In organizing the division, Hartmann shrewdly hired former software managers. Michael Tomczyk became the head of the marketing & documentation group. Paul Goheen also had experience with Commodore software from the PET days. “Dieter Ammon was in charge of the software division for the Commodore PET business computers. He was probably gone from the company fairly early on in the VIC-20, C64 days,” says Neil Harris. “He had a guy named Paul Goheen working for him as his number two guy. Paul stayed around for many years and

ended up working for Sig Hartmann in the software group.” John Mathias continued running the games group. The previous head of the division, Bill Wade, now ran operations of the software division. Hartmann also named Andy Finkel as the overall technical manager for the group. “Once I started working for Sig Hartmann in software, I got promoted too, so I was working directly for Sig as opposed to working for John Mathias,” says Finkel. “At that point I started going to Jack’s weekly meetings that he held with the VPs and directors, where he’d either praise people or yell at them depending on what circumstances required.” Though many would shrink at the thought of going to meetings where Tramiel might yell at anyone, after a meeting or two it started to seem normal. Hartmann’s marketing group, along with Kit Spencer, also courted third party software developers to create applications for the C64. “As a marketing organization, we set up technical presentations for software developers,” says Spencer. “In those presentations I would try and give them marketing input to convince them this product was going to be successful. Typically we started off by showing them the commercials, predicting what we are going to sell, so they get enthused, and then we gave them the technical information to let them do their part and create software.” To lead the educational software division, Hartmann hired Dr. Dan Kunz, who was formerly responsible for selling PETs to schools. Some of the first C64 software titles came from the Educational Software Program. Hartmann recognized there was not much of a market for selling educational software, so he released the software as public domain to boost the educational image of the C64. The Educational Software Program mostly converted BASIC educational titles from the PET and VIC-20. In total, there were 96 educational titles created between 1982 and 1983. The games division had greater success with their software sales. With the exception of the smash hit Choplifter by Broderbund, Commodore dominated the C64 software sales charts, not merely because of superior distribution, but because of the surprising high quality of many of their games. Tramiel gave Hartmann a mandate to create a full Commodore 64 software

library for the June 1983 CES show. Many of the titles would come from outside developers. Robert Russell feels Hartmann’s greatest strength was negotiating to acquire software titles. “He could deal with the outside companies,” says Russell. “That’s what made him effective.” “He loved to make deals and he made really good deals for Commodore,” says Finkel. “I learned a lot about negotiation from Sig.” Tramiel wanted a good spreadsheet application to defend against VisiCalc on the Apple II. Hartmann decided to go to Microsoft, who had a new spreadsheet application called Multiplan. He travelled to Seattle, Washington with Andy Finkel, John Campbell, and Gail Wellington. “We got to travel to the Microsoft campus and see the luxury where they lived,” recalls Finkel. “There were drink machines on every floor, every office had a door, and every office had two development machines. It was very nice.” When the Commodore employees met with Bill Gates’ team, Hartmann knew it would be a difficult meeting. “They weren’t really interested in really ing Commodore,” laughs Finkel. “He was always kind of cordial but they wouldn’t really put any effort into working with Commodore.” Gates’ reluctance went back to his earlier dealings with Jack Tramiel. “My theory was Bill Gates always resented the deal he’d made for BASIC with Commodore,” says Finkel. However, Hartmann persevered. “While I was there when they went to Microsoft, I really wasn’t saying much during the meetings except to say whether or not we could do something, or this is how we’d have to do it on our platform or that sort of thing,” says Finkel. Surprisingly, Hartmann convinced Microsoft to make the deal. Microsoft wrote Multiplan using a pseudo code C language, which made it easy to port to a variety of computers. “Other than Multiplan, which was really easy to port because of the way it was written, we never got anything on the Commodores,” says Finkel. * * *

Hartmann chose not to continue the deal with Scott Adams’ adventure games, perhaps because Tomczyk had been too generous with his royalties. “The main series was never licensed by Commodore for the C64,” says Adams, who went on to convert them for the C64 himself through his company Adventure International. Hartmann instead went to Infocom for many of their most popular text adventure games. “I think they felt that the Infocom games showed off the platform better as they required a disk drive,” says Adams. Hartmann travelled with Andy Finkel to their Cambridge, Massachusetts headquarters. “It was kind of like a college building,” recalls Finkel. “I just fairly large whitish interiors with glass and a good amount of light.” As the technical manager of software development, it was up to Finkel to advise their programmers on porting the titles to the C64. “The Infocom games were a pretty big deal,” he recalls. “I was mostly there for technical during the meetings, rather than playing an active role in the negotiations.” Finkel marveled at Hartmann’s ability to make a deal, which was almost the opposite of Tramiel. “I learned that while you could negotiate deals that were really good for you and bad for the other guy, you shouldn’t do that,” he says. “You should always make it so that when you walk away from the table, both sides are at least reasonably happy. That was the big thing I learned from Sig.” After the deal was complete, Finkel ended up staying in Cambridge to help the programmers port their adventure game interpreter, which they called the Zmachine. “I made quite a few trips to Infocom helping them port the Infocom adventures to the Commodore 64,” he says. During these trips, Finkel became fast friends with Infocom founder Joel Berez, one of the programmers behind the Z-machine. He also became friends with Steve Meretzky, a humorous game tester and programmer who went on to develop Planetfall and other games. Commodore distributed Deadline, Starcross, Suspended, and the complete Zork trilogy. Of these, Zork was the most popular and spawned a dedicated following. In 1984, Infocom ended their relationship with Commodore once they established their own software distribution channels. * * *

The 1600 VICModem for the VIC-20 became a hit for Commodore, selling over 100,000 units, making it the largest base of telecommunications in the world at the time. The Compve online service also grew due to the trial subscription with every modem. As a result, Commodore created a modem for the C64, called the 1650 modem. Hartmann spotted an opportunity for Commodore. “Commodore had a modem that was coming out that was the latest generation,” recalls Neil Harris. “Sig Hartmann originally negotiated the deal with Compve.” For the deal, Commodore agreed to produce an online magazine called the Commodore Information Network, accessible through Compve. It would include helpful computer advice, troubleshooting tips, and the latest news. Kit Spencer wrote the agreement. “We had a royalty on connect time coming to us in the agreement that I drew up with them,” he says. “The basic deal was that we were getting a royalty on the revenue that was generated in those forums specifically, which was a pretty common deal for Compve to do at the time,” explains Harris. “The percentage was probably 15 or 20%. They were billing people by the minute for usage.” Spencer kept current on the latest technological developments from Commodore’s engineers, and he used that knowledge when creating the contract. “I always made a point of being friendly with the techie guys with the beards and the weird attitudes,” says Spencer. “I grabbed a beer with them.” With software s in their pioneering days, it became another potential source of revenue. According to Spencer, “They were telling me, ‘Everything is going to be ed. You won’t have disks or hard drives—it will all be down the modem in a few years.’” Even though software s were not yet part of Compve’s service, Spencer included it in the contract. “We couldn’t do it at the time, but we built in a royalty for ing software,” he says. “It would have been a good one a few years later.” After g the agreement, Hartmann realized he did not have anyone to create the Commodore portal. According to Harris, “He did the deal and then as things started moving forward I think Sig realized, ‘Wait a minute, we’ve just done this deal to run this Commodore Information Network on Compve and we don’t

actually have anybody to run it.’” Spencer and Hartmann hired some junior developers to set up the Commodore Information Network. “Then they said, ‘We need somebody to actually manage these people. It’s kind of like a publication, so let’s give it to Neil.’ So they gave those people and the responsibility to me. I didn’t set it up but I got to manage it.” Within days, Harris found himself an addict. “I became absolutely hooked on online services,” he says. Compve became a place for s and especially developers to gather for all things Commodore. “Every Commodore modem that went out had a Compve brochure in it and anybody that ed from that would come into the Commodore Information Network,” says Harris. “We basically made that the central hub of Commodore s online. We reached out to people in the development community and people who were publishing software and making hardware to provide and information.” Within months, the Commodore Information Network became the most popular portal on Compve. “It really became a tremendously large area,” says Harris. “In fact, it became the single biggest area on Compve during that timeframe.” Spencer likens the Commodore Information Network as a precursor to the Internet. “On Compve, we effectively had the first ever webpage,” he says. “We employed a technical guy to answer all the questions online and help do bits in the magazine. It was all part of what I help put in as a network. When you look at where things are today, that was a very important breakthrough by Commodore.” * * * After the successful debut of the C64, Commodore planned to release the Max Machine. Tramiel hoped it might ride some of the success of the C64. “When we first designed the Commodore 64, it was going to be a $600 computer and the [Max Machine] was going to be a hundred-something dollar computer that was going to use the game cartridges that were developed for the Commodore 64,” says Yannes.

Once Yash Terakura completed his design, Tramiel ed the project to Japan. “We had put together the architecture for it,” says Winterble. “Japan grabbed that and worked on the [production] design.” Tramiel was attempting to duplicate the success of the VIC-20 by releasing the low-cost computer in Japan first. He set a target price of $179. Once again, he was trying to meet his Japanese competition on their home ground. Once the machine had a foothold in Japan, he would launch it in the United States.

Releasing the ill-fated Commodore Max Machine.

The Max Machine outclassed the other leading game consoles. In technical specifications, the Max Machine would have had approximately the same graphics and sound capabilities of the hugely successful Nintendo Entertainment System, released years later. Unfortunately, out of the many variations of the Max Machine that Terakura designed, Tramiel chose the small computer design rather than the dedicated game console. As a computer, the machine was lacking. With the Mini BASIC cartridge, only 510 bytes of memory were available for programming. This was obviously not useful to anyone. Another BASIC cartridge included an extra two kilobytes of RAM—only nominally better. The decision to push the product despite objections within his company showed Tramiel had not yet made the transition from calculators to computers. He believed the low price would make up for a weak product. Back in North America, Tramiel placed Kit Spencer in charge of preparing an advertising campaign for the Max Machine. Spencer began preparing his advertising campaign with the New York ad agency Ally & Gargano. Michael Tomczyk would help design the packaging and manual for the machine. Both men saw the obvious, but they dutifully went to work anyway. According to product literature of the time, Commodore intended the Max Machine as a third generation game console. To become a true console, the engineers needed to take a stand and eliminate the keyboard. Without the manufacturing cost of a keyboard, Commodore could release a cheaper console with superior graphics and sound. Instead, it remained a poorly defined computer. Commodore Japan manufactured and released the Max Machine in their home market, alongside the Japanese Commodore 64. As many had predicted, the Max Machine was a monumental flop. By November 1982, Commodore was still planning to go ahead with the North American release. However, the situation had changed during the development

of the machine. When the project began, the VIC-20 was selling for $299. The $179 Max made sense, but by the time the Max Machine was ready for release, the VIC-20 was selling for under $200. By early 1983, even the Commodore 64 was closing in on the Max Machine price. “It wasn’t long after the introduction of the Commodore 64 that it was down in the $300 range,” says Yannes. “There just wasn’t any reason to produce the Max Machine. Why buy the Max with its membrane keyboard and tiny amount of memory when you can buy the C64 for [nearly] the same price and get a whole computer?” Tramiel handed off the North American product release to consumer product manager Bill Wade, who clearly saw the product had no future. The VIC-20’s price was now lower than the Max Machine’s price, making the Max irrelevant. He wanted to abandon the machine and this time Tramiel agreed.

CHAPTER 31

Commodore Mania 1982-1983

The Christmas season of 1982 was a turning point for Commodore. The C64 had a massive advertising campaign with regular television commercials, distribution through thousands of dealers, and demonstrations in stores. It was now up to consumers to decide if Commodore would become a success. Through 1982, Commodore sold the C64 through computer dealers. In the first two weeks of the August release, Commodore sold 12,000 units. Between September 1 and December 1, they sold 65,000 systems. The system exceeded buyers’ expectations because Kit Spencer’s television advertising stressed memory and price but said little about its superior sound and graphics. Commodore began to earn fierce brand loyalty. Magazine reviews were unanimous in their praise. Byte contributor Stan Wszola applauded the C64’s sound and graphics, though he mentioned the lackluster BASIC 2.0. He also pointed out Commodore’s poor quality control, since he had to return two computers before he finally obtained a working system. Creative Computing’s Ron Jeffries appreciated the amount of computing power available for under $600, though he found problems with the video display, which appeared blurry. Commodore received another piece of free advertising from William Shatner when Paramount Pictures released a comedy sequel on December 10, 1982. “In his movies there is usually a Commodore something-or-other on the set,” says Andy Finkel. “In Airplane II, he uses a VIC-20.”

Finkel and Neil Harris made it a game to see every Shatner movie and try to spot Commodore products. “It was consistently showing up for years afterwards, so I always wondered if he just gave them to the prop department whenever he needed a computer,” says Finkel. Christmas sales sured expectations, but 1983 was really the year of the C64. On January 4, 1983, Time magazine selected the computer as their man of the year for 1982. A generic computer graced the cover, making it the first year in which a human failed to win. To its credit, Compute! magazine was one of the first publications to recognize the potential of the C64. In January 1983, publisher Robert Lock announced a Commodore spin-off magazine he titled Commodore Gazette (later renamed Compute!’s Gazette). Commodore sold over 800,000 VIC-20s for the year along with PETs, C64s, printers, disk drives, software, and semiconductor components. The company reported sales of $304 million, with an impressive profit of $40.6 million due to Jack Tramiel’s streamlined operations. With profits so high, the board of directors declared a 50% dividend, meaning they took half of the profits and divided them among shareholders. Tramiel received a check for $1.6 million, while Irving Gould received $3.6 million. A golden age for Commodore had begun. Throughout his success, Tramiel remained low-key, rarely granting interviews. “Jack had some very bad experiences with PR (Public Relations) by people misinterpreting what he said or misquoting him or not giving him credit for things,” explains Michael Tomczyk. “He basically was a little bit PR averse.” Throughout most of 1982, after Tramiel rudely dismissed him from his marketing post, Tomczyk represented Commodore to the outside world. “He actually designated people like myself to be PR spokesperson. I was one of the few people allowed to speak to the press whenever I wanted, and that helped change things a bit.” Amidst Commodore’s success, Forbes magazine released a scathing attack on Tramiel. The teaser read, “Jack Tramiel’s cloudy past and autocratic personality mar the otherwise bright future of Commodore International.” It was a curious article to appear during Commodore’s best year ever.

According to Tomczyk, the negative article originated due to Tramiel snubbing a social invitation. “Malcolm Forbes sent him an invitation to come and spend time on his Yacht, which Malcolm Forbes was fond of doing with Chief Executives,” he recalls. “Jack totally ignored it and never even responded. He was simply too busy to be bothered.” About a year later, a reporter named Subrata Chakravarty ed Tomczyk in order to write an article on Commodore’s recent success. “He told me what he was going to write and how he would frame it. I said, ‘Well that’s fine, we’re finally going to get some credit,’” recalls Tomczyk. “I talked Jack into doing the interview because I knew the reporter.” Chakravarty wrote the article and sent the first draft to Tomczyk. “The reporter actually sent me a copy of the article which looked fine,” he recalls. “Between that time and the time it was published, apparently Malcolm Forbes came to the reporter and forced him to change the story. The reporter indicated that Malcolm was very upset that Jack never responded to his request to go to his yacht, and as a result, he decided to trash him.” On January 17, 1983, an article entitled “Albatross” appeared in Forbes magazine. “The title implied that Jack was an albatross around Commodore’s neck, which was just the opposite,” says Tomczyk. “Jack was the inspiration behind Commodore.” The article surprised Tomczyk. “Malcolm Forbes was so petty and so vindictive that he would turn what he deemed to be a social affront into a journalistic trashing. That was hugely unethical,” he says. “It was one of the most bitter experiences that I had during my time at Commodore. It reflected negatively on me, since I had arranged the interview in good faith. I guess I was naïve. At that point, I totally understood why Jack shunned the media and public relations.” In the article, Gould bolstered his own contribution to Commodore. The article read, “[Irving] kept Tramiel on a very tight financial leash. ‘I wouldn’t give them enough money to take any real risks,’ Gould says.” Along with Gould, ex-Commodore president Jim Finke expressed his frustrations with Tramiel. “ Machiavelli? The strongest kingdom had the strongest barons and a strong king. There are no strong barons in Jack’s kingdom. There is only a strong king.”

The negative article did not surprise Bob Yannes. “The financial community did not like Jack at all, supposedly because he was really unpredictable and he was totally in charge of the company,” he explains. “If he made a bad decision, there wasn’t anything to temper his decision.” * * * Every year at CES, Tramiel attempted to make a big announcement. “He planned to reduce the cost of the Commodore 64 but he wanted to wait until after Christmas to get the last bit of money,” says Dave Haynie. Just weeks earlier, in December 1982, Atari dropped the price of the Atari 800 to under $500, edging out the $595 price of the C64. Tramiel held a meeting with Kit Spencer and managers from the sales division to figure out how low they could go. “We would always want to be the most aggressive on pricing,” says Spencer. Tramiel was very clear in illustrating his point to his marketing people. According to Spencer, he told them, “Both the calculator business and the computer business are like the fruit market on a Saturday night. If you don’t sell it at five o’clock, the price is down tomorrow because the fruit’s no good the next day. You’d better sell it now because the price will never go up in this business.” However, they could not arbitrarily lower prices without examining all the costs. “After the original development costs, you’re looking at just the straight production cost of materials, and that’s going down all the time,” says Spencer. “The engineers would see the cost-price of components when they first develop it, but they wouldn’t see the whole situation with shipping, marketing, overheads, customs duties and 101 things that go into different products before they reach a .” One of the biggest issues affecting manufacturing costs was the initially low yield of VIC-II chips. By January 1983, the yields were much improved, which lowered the costs of the chips dramatically. The other factor was Kmart, which was just about to release the C64 to their store shelves. “You have to make decisions about what volume you’re going for and what period of time you’re at in a product cycle,” says Spencer. “The market dictates a lot of that.” Throughout their discussions, Tramiel did not consider the existing dealers who

had purchased C64 inventory at full price. According to Yannes, “He went to ComputerLand with the Commodore 64 at $595 and they said, ‘You’re not going to do what you did with the VIC-20, are you?’ and Jack said, ‘Oh, no.’ And then of course, he did.” At the January 1983 CES, Tramiel announced a new price for the C64 of $399. The sudden move left computer dealers with overpriced stock. Curiously, Yannes blames ComputerLand for doing business with Commodore in the first place. “You know, how stupid could ComputerLand be? Jack didn’t care if his dealers made any money on them. He would cut his profits as much as he could to try to take over the market.” The move alienated both large and small dealers. “Commodore made some blunders there,” says Dave Haynie. “The classic story was that the local mom and pop dealer could buy the C64 for less at the local Kmart than they could buy it from their local distributor, which gave Commodore a really bad name.” Tramiel could have warned his dealers of the price drop and given them a chance to sell off their inventory. Instead, he undercut his own distribution network. “Charlie [Winterble] told me this joke, in relation to Jack,” says Yannes. “There was this turtle that was going to cross the river and there’s a scorpion there. The scorpion asks the turtle if he’ll take him across the river. The turtle says, ‘No, you’ll sting me and I’ll die.’ The scorpion says, ‘I’m not going to sting you. We’ll both drown.’ So the turtle says, ‘Okay.’ The scorpion climbs up on the turtle’s back and the turtle goes across the river. Suddenly he feels himself being stung. He turns to the scorpion and says, ‘Why did you do that? We’re both going to die!’ The scorpion says, ‘It’s my nature.’” After the announcement, the sales force had to deal with angry dealers coming into Commodore’s booth to complain. In response, Commodore encouraged their dealers to concentrate on the new P and B series, which Tramiel still hoped would be able to compete with the Apple II line of computers. The suggestion did little to placate anyone. In one move, Tramiel had virtually destroyed Commodore’s reputation with dealers. CES was the major show for Commodore to display the actual production models of the C64. Spencer was justifiably proud of how it turned out, despite

the crippling problems during production. “You’ve got to say with the C64, we could have done a lot of things better, given time and money. But then you’ve got probably the most successful computer ever. You’d have to say the technical guys got it right.” Commodore displayed their newest C64 games, including the Bally-Midway conversions and HAL Laboratory games. Even though Bob Yannes was no longer at Commodore, he was thrilled to see a software library building up around the machine he designed. “I used [a C64] all the time,” he says. “I enjoyed playing Omega Race.” Later at the show, Commodore demonstrated prototype products, including the C64-based Osborne-killers, and the HHC-4 handheld computer. The HHC-4 received little enthusiasm. With only one line of text display available, it seemed stunted compared to other computers. It would be unable to display games or applications, making it little more than a digital notebook. Tramiel decided to terminate the product before release. The Osborne-killers met with a better response. Commodore demonstrated the SX-100, a portable computer with one disk drive and a black & white monitor for $995. Their slightly more advanced DX-64 came with two disk drives and a color monitor, for $1295. The machines looked similar to the Osborne 1 and Compaq Portable, both of which resembled the Xerox NoteTaker prototype from Xerox PARC. Tramiel decided to go to production with the Osborne-killers. At the show, Commodore signed up Sears, who would soon become a major computer distributor. Later in the year, Sears’ stores opened up a separate area for computer sales, each with about five demonstration C64 computers complete with floppy drives, tape units, printers, and monitors. * * * Apple released the IIe in January 1983 to strike back against the C64. The main selling point was 64 kilobytes of RAM, up from 48 kilobytes in the old II+. The IIe also had the new ability to display upper and lower case characters, a first for Apple, and display 80-columns of text on a monitor. Apple charged $1,395, which did not include a disk drive or monitor. Tony Tokai released a Japanese version of the C64 in early 1983. The machine

had Japanese Katakana characters in place of many of the PETSCII characters and it booted up to black text on a pink background as opposed to the familiar blue colors. It sold for a price of 99,800 yen (approximately $400 US). Commodore s in Japan received from a magazine called Vic! The Magazine for Computer Age.

Computing with the portable C64.

Although the C64 was unable to dominate the Japanese marketplace, it was responsible for keeping Japan from entering the North American market. In a broadcast of Computer Chronicles, Tramiel told his hosts, “As far as the Japanese are concerned, I was able to keep those people out of the US market and almost the world market for the past seven years. … What I’m trying to do is come out with the best product, the best quality, and the best price and by doing so, I keep those people out. Thanks to God, I’ve been successful so far.” Co-host Gary Kildall agreed, saying, “The C64 was definitely one of those devices that kept the Japanese machines out.” * * * At any other company, Kit Spencer would have been highly valued after his successful launch of the VIC-20 and C64. “Kit Spencer was key,” says Russell. “He was really an important person.” “It was quite an interesting period,” recalls Spencer. “When I arrived in the States we were doing $50 million dollars in sales, which was actually less than we were doing in the UK at that time. Eighteen months later, we were doing $500 million and were well back in profit.” At Commodore, however, it seemed like his record meant little. Curiously, most of Spencer’s problems came from Gould’s recent hire for Commodore president, Bob Lane. Many people thought Lane began making changes within the company too soon, without first studying the company and finding out its strengths. “He didn’t understand the industry and didn’t seem to want to learn,” says Spencer. “I found I couldn’t get on with Bob Lane.” The sudden growth of the company gave Spencer a choppy ride. “Once the C64 became successful, the whole company started blowing up as far as personnel,” says Russell. “They started doing magazines and stuff like that. Kit was getting shuffled around.”

Spencer began to detest his time at Commodore. “I enjoy starting up things. I’m generally pretty good building new markets and operations, but when they get really big, it’s a different operation,” he says. “Politics come in much more, companies often lose focus on what is really important and managers can become too involved with their own empires rather than what is best for the overall company.” Russell felt sympathy for Spencer. “The company got bigger and there got to be more politics. People like Kit and me were doers, we weren’t politicians. We didn’t want to go up the corporate ladder and fight politics; we wanted to do what we liked to do.” Despite his key contribution to two of Commodore’s greatest successes, events pushed Spencer to the sidelines. One of the first decisions Lane made was to replace Spencer with another Commodore employee, Myrddin Jones. With Jones now vice president of sales and marketing, Spencer lost most of his responsibilities. “That happens at lots of companies where the original people do things,” explains Russell. “Sometimes they reach burnout, sometimes they need a cycle and somebody gets put in their place, and sometimes they just don’t want to play the politics anymore. They get tired of, ‘Why do I have to argue with these idiots who came in yesterday about how things should be done. Let them go fail and I’ll go do something else.’ It happens all the time and sometimes it’s not worth the battle.” Tramiel tried to save Spencer. “When Bob Lane came along, obviously we were not getting along very well, because quite frankly the guy was an idiot,” says Spencer. “Jack asked me if I would go back to Europe and run operations there. I said, ‘No, as far as I’m concerned I’m going to change my lifestyle, cash my shares in and go to the Bahamas.’” Tramiel knew Spencer had made up his mind, so he persuaded him to stay part of Commodore in a minor capacity, telling him, “Okay then, I’d like to help you be down there. Let’s keep you involved in some way.” As Spencer recalls, “Jack was as good as his word on that one and I took up a position as VP of Commodore in the Bahamas where our holding company was ed. I think Jack was hoping I would get bored in time and would want to

get back fully in the company again, which is a typical Jack move.” Although Bob Lane was the trigger that caused Spencer to quit day-to-day operations, he felt he would have left eventually anyway. “I would have stayed on a bit longer,” he says. “I had always intended at some time to change my lifestyle.” Spencer was financially secure after building up Commodore stock since the seventies. “He made a fortune in Commodore stock and he said that’s enough,” says Leonard. “He said, ‘I’m going to play tennis now.’” “I just decided I had a chance to give myself security for life if I cashed my shares in the Bahamas, which is very good if you’re British,” he says. According to Charles Winterble, the official company headquarters in the Bahamas was a front for tax savings. “I used to go there once or twice a year, but there was nothing there but an office,” he says. In his new position, Spencer held little power. “Kit could get out of there and go hide in the Bahamas and enjoy the life there,” says Russell. “He occasionally showed up for meetings and then he wrote the Commodore company newsletters that he published out of the Bahamas.”[1] Kit Spencer’s marketing reign lasted 15 months. Before him, product advertising was deplorable. The loss of Spencer was one of the biggest blows to Commodore, who could have used his talents launching more computers. Though Lane failed to realize it at the time, Spencer’s superior marketing had gained Commodore an unbreakable lead. He made a success out of the PET, even though it sold for double the price in the United Kingdom. The VIC-20, a modest piece of technology, was the first computer to break a million units. Now with the C64, Commodore would have years of high sales ahead. He had the ability to turn almost any product into a marketing success. Unfortunately, with Spencer moving to the Bahamas, Commodore’s marketing and advertising reign was at an end. * * * Tramiel planned to release the P & B computers to replace the antiquated PET line. The P series were personal computers (with no monitors), while the B

computers were for business with 80 column displays and monochrome monitors. Commodore handed out demonstration units to select dealers. In November 1982, Byte magazine announced the P128 for $995, the B128 for $1695, and the BX256 for $2995. Al Charpentier feels the P128 was no longer a sufficient Apple-killer, due to the release of the Apple IIe in January. “The whole problem with them was that they didn’t have sufficient graphic information when you looked at the Apple doing 80-columns,” he says. In Europe, Commodore created an elaborate naming scheme for the P & B series. They dubbed the new line the CBM-II. The CBM 505 and 510 were the personal computers, with 64 kilobytes and 128 kilobytes respectively. The CBM 610 and 620 had 128 kilobytes and 256 kilobytes memory. They also planned to sell high-end 710 and 720 computers, which included detachable keyboards and swivel monitors. The design group came up with a revolutionary look for personal computers. The 700 series were the best-looking computers in the early eighties, far ahead of their time. They used rounded moldings reminiscent of the rounded look Macintosh monitors would use decades later. The CBM 710 and 720 models even housed two floppy disk drives. Byte magazine viewed the computers at the National Computer Conference in Texas. They remarked, “Commodore gets the prize for the wildest styling of any computers we saw at the show.”[2] Although Kit Spencer had little impact on the design of the C64, he made marketing suggestions for the P & B series. He had targeted the computers against the IBM PC and pushed engineers to develop an optional /M expansion board, which included a Z80 microprocessor. Commodore produced limited numbers of the computers and released them to dealers. s had little reason to purchase the computers, which received almost no software development. The success of the C64 was overshadowing the P & B series. * * *

To casual observers, it looked like Jack Tramiel was completely in charge of Commodore. However, despite founding Commodore and holding the CEO position, Tramiel did not hold the most power within the company. The real power belonged to Irving Gould, who owned 17.9% of Commodore stock— more shares than any individual did. Gould held residences in three locations: New York (where his wife lived), Toronto, and the Bahamas, ostensibly for tax purposes. “He had a whole schedule set up of how many days he could spend in the United States and how much time he had to be out of country,” explains Russell. “Irving would show up unexpectedly because it was time to put some time in the United States.” According to Chuck Peddle, Gould relished the fame brought to him by Commodore. “Irving was rich,” he says. “He started a container business and he sold his container business. But being rich in New York wasn’t enough. Being the top dog in a hot company put him in a totally different social structure. Irving liked being the owner of this hot company whose stock was running.” Rumors floated that Gould thought Tramiel was getting too much credit for Commodore’s success. The financier is an important part of a company and perhaps Gould felt he should share more of the spotlight. Through the years, Gould earned a reputation among Commodore employees for living an ebullient lifestyle. “Irving was the guy who flew in the jet with this gorgeous babe for as long as I worked at Commodore,” says Russell. “That’s what Irving was. I never saw Irving with what was identified as a wife, but he would show up with this gorgeous babe every now and then.”

Irving Gould overseeing Commodore as the chairman of the board.

Gould’s vast wealth allowed him to do almost anything. One of his ions was collecting expensive oriental carvings and vases. The Toronto Star even featured an article on his famed collection. Gould’s other ion was his seventy-foot yacht, which was manned 24 hours a day with a captain and crew. Leonard Tramiel visited Gould’s lavish residence in the Bahamas. “People with a lot of money like to travel a lot, and he certainly liked to travel and had a lot of nice places to go,” says Leonard. “I had been on his boat in the Bahamas and seen his house down there. He certainly had nice stuff.” Despite his regular visits to the United States, Gould remained a cipher to those who worked with him. “Even though we were important to Irving, Irving was never close to any of us,” says Peddle. “He was a very cold, distant guy. I liked Irving, but Irving was the kind of guy who could reach across the table, cut your heart out with a penknife, and never crack a smile or feel bad about it.” “I’m not saying he didn’t have the ability to be socially gracious, but underneath it there was nothing,” explains Peddle. “He literally does not have emotions. There’s just nothing underneath that front layer. He was one of the most absolutely ruthless, focused people I’ve ever met.” Engineer Robert Russell spoke with Gould on several occasions, but was unable to pierce through his veil. “Irving was always a really tough guy for me to read,” he recalls. “He was more the type of guy who nodded his head.” Leonard Tramiel found it difficult to acquaint himself with Gould. “I certainly met him a few times but I never really spent any time talking to him and didn’t really know the man,” he explains. Charles Winterble had similar experiences with Gould. “I used to do the financial presentations,” he recalls. “I’d give presentations to stock analysts [in New York and San Francisco]. He was always nice to me and complimentary on the stock presentations. He would have his Rolls-Royce pick us up in New York

and drop us off, which was rather nice. But I would not even pretend to know him very well, if you know what I mean.” Kit Spencer had frequent meetings with Gould. “I knew him better than 99% of the people at Commodore, but I still didn’t know him that well,” he says. Gould met Tramiel in 1966, when a financial scandal in Canada threatened to destroy Commodore. “That was big time shady stuff, but I think that was more Jack’s partner,” says Russell. “Irving didn’t really let that type of stuff happen, as far as I knew.” For eighteen years, Gould and Tramiel operated the company in relative peace, though never shoulder to shoulder. “After I had known Irving for several years I realized that I’d met him quite a few times but never actually on the Commodore premises,” says Spencer. “It was somewhere he never really came to while Jack was there.” There was no doubt they were responsible for each other’s success. Tramiel transformed Gould’s $500,000 investment in 1966 into over $200 million dollars by 1983. For his part, Gould was chiefly responsible for saving Commodore when no one would loan money to Tramiel. In fact, after the scandal, Gould had to guarantee every loan Commodore received. After Tramiel turned Commodore around in the early 1970s, Gould rewarded him. As a gesture of gratitude and trust, Gould sold Tramiel 8% of Commodore at cost. This was an enormous gesture, worth $90 million by 1983. As a publicly traded company, Gould wielded his power as chairman of the board. If push came to shove, the board would Gould over Tramiel. Tramiel might run Commodore, but Gould was the final word on any decision. “In the USA, it was much more political for everyone, including for Jack as Irving and the board were also mostly in the North American area and so interference from everyone was much greater,” says Spencer. In 1983, as Commodore approached $1 billion in annual sales, Gould felt the company should operate more like a Fortune 500 company. He wondered if Tramiel had the nuanced ability to run a large company. “You’re talking about a company that was run as if it was a small garage shop,” says Yannes. “He was in charge and he made all the decisions. The fact that it became a billion dollar company was almost a fluke. Left to normal circumstances, a company run that

way would never have gotten that large or that successful.” Al Charpentier believes Tramiel had some big faults. “He didn’t listen real well,” he says. “He made up his mind and it was very difficult to change his mind. That’s not unusual for people who are running companies because they believe in what they are doing, but that led to most of his problems.” The recent Forbes article also had Gould and others thinking. “The story in Forbes influenced Irving Gould,” says Michael Tomczyk. “This was not fair, but business is not fair, and neither is war.” The quirkiest aspect of Tramiel’s business style was not allowing budgets on projects. “He didn’t believe in budgets because he believed budgets were a license to steal,” explains Yannes. “If you said you needed fifty million dollars to do something then you were going to spend fifty million regardless of whether you needed it or not. So he didn’t let his people have budgets. How can you run a business without having budgets?” Tramiel personally signed all company expenditures over one thousand dollars. “You would tell him what you were going to spend and he would either approve it or disapprove it,” explains Yannes. “How can you get your work done under those conditions?” Spending came to a halt when Tramiel was away on business trips. “He was all over the place. We had stuff in Japan, the Bahamas, California, Pennsylvania, and Europe, plus he took a lot of time off for himself. You just can’t run a big business that way.” Peddle believes Tramiel felt insecure delegating major aspects of his business to others. “In order to make it work as well as he could have made it work, he had to bring in all these people that were better at things than he was,” he explains. “If you read entrepreneurial 101, you hire people better than you at everything. That’s the way to be successful, because they’ll take the company places you couldn’t take it. A lot of entrepreneurs miss that step, and that’s why they fail.” Tramiel made sure he was the only one in Commodore with real power. “Jack never brought in anybody that would have even conceivably been any threat to Jack,” says Peddle. “Sometimes he would bring in some guy and say, ‘I’m going to start taking time off and give this guy part of the action.’ We would say, ‘Hohum,’ because we knew it wasn’t true. Then Jack would give the guy some rope, he would hang himself, and then Jack could kill him.”

Commodore executives rarely lasted for more than six months. In the Forbes article, ex-Commodore president Jim Finke claimed, “five key executives have departed in the past few months.” It might have been impossible for anyone to last near Tramiel. The situation produced poor leadership. Commodore needed seasoned managers and executives who could take on the microcomputer industry. Instead, executives barely had a chance to learn about Commodore before they were out the door. * * * One of the biggest criticisms the Commodore 64 had to overcome was the lack of software on release. “The people who bought Commodore 64s didn’t buy them because there was lots of applications software,” says Yannes. This seems incredible, given the unmatched software library that eventually developed, but software development was slow for the first year, mainly because of the lack of volume disk drive production until late 1983. Commodore advertising initially promised /M compatibility and the sizable /M software base. Unfortunately, due to engineering difficulties, Commodore was unable to launch the /M cartridge alongside the C64. An unnamed engineer dubbed “Shooting Star” was in charge of the /M cartridge. Robert Russell hoped to speed up development of the cartridge by analyzing the competition. “I gave him the [Apple II /M card] schematic and the documentation and said, ‘This is what we want.’ I didn’t care whether he copied it or not, but I’m sure he did because he wasn’t that great an engineer.” The /M cartridge for the C64 remained under development by “Shooting Star” but technical problems plagued the project. “It took him a long time to even make it somewhat functional,” says Russell. “There were some serious timing problems with that card that kept it from really coming out and succeeding.”[3] /M was never the focus of C64 software, however. The openness of the Programmer’s Reference Guide gave Commodore more third party over competing companies. “[The reference guide] not only helped sell the product and create more software but also made a lot of money in their own right,” says Spencer.

Surprisingly, the publishing division of Commodore produced significant revenues. For every five computers Commodore sold, one customer purchased a Commodore manual. As a result, Commodore sold 600,000 books in 1983. With the release of the Commodore 64, game publishers quickly shifted away from Apple in favor of Commodore. The C64 received early software development by the top game publishers of the day, including Broderbund, Epyx, Electronic Arts, and Sierra Online. Other publishers came along and disappeared just as quickly, such as Spinnaker, Sirius, Hesware, Creative Software, Synapse, and Tronix. The Apple II and TRS-80 dominated a 1982 catalog from Eypx, with some for Atari and IBM games, and a handful for the VIC-20 and PET. By 1983, the situation changed dramatically, with games like Pitfall and Jumpman Junior exclusively for Commodore and Atari computers. In 1983, TRS-80 games all but disappeared. Commodore engineers found time to play Epyx’s first hit. “Jumpman was probably the most popular game with some staying power throughout engineering and I assume some of the other departments,” says Bil Herd. Commodore even published a game submitted by two young Canadian programmers called Jack Attack. It was a joke most Commodore employees could appreciate. “You would have to be an insider to know what that meant,” says Charles Winterble. Originally, the game had no title, so Tomczyk and some of the software developers decided on the new name. “Jack Attacks happened so frequently, it was common jargon around the company, so we decided to name a game that,” says Tomczyk. [1] According to Russell, Kit Spencer published the company newsletter until the early 1990s. [2]Byte magazine, (Vol. 7, No. 9, September 1982), p. 61. [3] The delays of the /M cartridge caused s to complain to the Federal Trade Commission. In 1984, the FTC charged Commodore with false advertising. “Commodore Business Machines signed a consent agreement in Aug-84, under which Commodore agreed not to capabilities that don’t

yet exist.” Byte magazine (October 1984), p. 9.

CHAPTER 32

New Projects 1983

As the failure of the P & B computers became more apparent, Commodore began an aggressive search for new technology. Rather than relying on Jack Tramiel to spot trends and drive new products, Commodore turned to their engineers. Tramiel wanted a group of his engineers to make a concerted effort to study the technology scene and then come up with new products for Commodore. It was reminiscent of the deliberate planning that led to the creation of the VIC-II and SID chips. Andy Finkel, the technical manager of the software group, was one of four employees who helped with the search, along with Eric Cotton, Benny Prudin and Judy Braddock. “Our task on the east coast was to look for and evaluate new technologies and see how Commodore could productize them,” says Finkel. “We were doing things like looking at LCD screens, new operating systems, thinking about whether Commodore could do a Unix machine cheaply enough.” Two new computers would come out of the search, giving Commodore a total of three major systems in development, including the Sinclair-killer. Russell had little enthusiasm for the Sinclair-killer, but he was more inclined to go along with his boss rather than resist because he also had plans for an IBM PC-killer. “It was a less than a hundred dollar computer. What did I care? At that time I was trying to get something more advanced together,” he says. To create an IBM PC-killer, Commodore needed a 16-bit microprocessor. Unfortunately, MOS Technology had failed to produce one. While Motorola and Intel delivered 16-bit sequels of their chips, MOS Technology lagged behind. “I don’t know precisely why they didn’t [produce a 16-bit chip], other than

Motorola is about a thousand times larger,” says Commodore engineer Dave Haynie. “Over time, fewer and fewer companies were successful with microprocessors. The only ones who did really well in the merchant market were Motorola and Intel.” With a limited workforce through the early eighties, Robert Russell believes Commodore had higher priorities. “We wanted a 16-bit 6502, but it wasn’t as important as fulfilling the computer demands,” he explains. “We had a successful VIC-20 and we wanted something else to put on the shelf. We had a successful 80-column PET, the 8032, and it was doing huge gangbusters as a business machine. Nobody was saying, ‘I need a faster computer.’ They basically wanted more features.” The closest thing to a 16-bit 6502 was the Motorola 68000, but Commodore was still in lawsuits with the company. “At that point in time we were still butting heads with Motorola, so Motorola wasn’t an option, which is what we really wanted to design with,” explains Russell. Instead of hiring engineers to develop a 16-bit processor, Tramiel looked to acquire the technology elsewhere. He found a potential match with the Z8000, a 16-bit microprocessor by ZiLOG. “The only reason I ever got involved in deg for the Z8000 project was because we were going to buy ZiLOG,” reveals Russell. Federico Faggin, one of the Intel 8080 engineers, founded Zi 1976. ZiLOG’s first microprocessor was the 8-bit Z80, followed by the 16-bit Z8000. “We went and did a lot of meetings with ZiLOG,” recalls Russell. “The idea was that Commodore was going to purchase them, so we started deg a business computer around the Z8000.” The Z8000 based computer would use Unix, an acclaimed operating system for minicomputers. “With the Z8000, we were trying to run a Unix like operating system and it was clearly a business machine,” says Russell. At the time, AT&T owned the most popular commercial version of Unix. Computer pioneers Dennis Ritchie and Ken Thompson developed the Unix system (originally called UNICS) in 1969, along with the C programming language. The source code for the operating system was freely available. Companies quickly began customizing and selling different flavors optimized for

different machines.[1] Tramiel made a deal with a smaller company for a Unix operating system, which they called Coherent. “There was some work done with a company in Chicago called Mark Williams,” says Leonard Tramiel. “The Mark Williams Company had a knockoff of Unix, and there was some work done on a computer with them. There was also some other stuff done looking at regular licensing from AT&T.” Coherent was compatible with AT&T’s Unix System V. Coherent even boasted a graphical interface similar to the Unix standard, X-Windows. Commodore announced the Unix machine in Byte magazine, calling it Next Generation. Systems engineers Frank Hughes and Robert Russell started the new business computer in early 1983. “Frank [Hughes] was in charge of the Z8000 project,” says Bil Herd. “He was from the west coast, like Shiraz and Bob Russell.” Hughes had previously headed the Commodore cash project, but after Tramiel cancelled it, he moved to the east coast. The engineer was not popular among management. “He was a rebel in his own cause. No boss liked him,” says Herd. “I liked him. I was one of the few people who got along with him later. The secretaries realized he was more of a pussycat when he was trying to act like a big belligerent bear.” The new Z8000 machine, dubbed Next Generation, would require a video chip. The VIC-II chip was not adequate for business computers, since it could not natively produce an 80-column display. The same applied to the TED chip, meant for the low-cost Sinclair-killer. It was up to MOS Technology engineers to create a next generation video chip. Russell would attempt to reawaken technology from the Chuck Peddle days using a prototype given to him by Bill Seiler. “We had really neat stuff that [Bill] Seiler did, like the high resolution board that he just kind of walked away from,” says Russell. “That was one of the things he was working on in the TOI era. I was trying to get it resurrected as a project.” After the mass exodus of talent from MOS Technology, it was uncertain if the remaining engineers could continue advancing Commodore’s semiconductor designs. At the time of his departure, Bob Yannes recalls, “There weren’t very many chip designers there.”

Al Charpentier believes MOS Technology had design talent. “There was a hole in the organization after we left, but I think there were still some good people there,” he says. While the rest of Commodore moved to Wayne, Pennsylvania, the MOS Technology engineers remained at MOS headquarters in Norristown, Pennsylvania. MOS Technology engineers began work on an 80-column video chip designed by a husband and wife team. “Kim [Eckert] was very bright, and so was his wife Anne,” says Herd. “They both happened to be redheads from Texas.” Like Peddle and his team, Eckert was from the more formal atmosphere of Motorola. “Kim had come from Motorola and had designed and patented some of the logic blocks in use on the Motorola 68000,” explains Herd. “He had great stats on paper.” Motorola ran on a corporate nine-to-five schedule, while Commodore ran on ion. “It was a clash of corporate culture,” says Haynie. “We didn’t like Kim Eckert too much. He wasn’t a typical Commodore chip guy.” The new chip used the CMOS process, so rather than making the chip part of the 6500 series, the name jumped ahead to the 8500 series. They called the new chip the 8563 Video Display Controller (VDC).[2] Given the long design times for semiconductor chips, it would be approximately 18 months before engineers could expect production quantities of the new video chip. * * * In 1983, with the success of the C64 becoming apparent, it was clear to Robert Russell that Commodore should do a sequel to the C64. Russell hired an engineer named Cong Su to lead the project. The proposed sequel would have 128 kilobytes of memory, and it was the successor to the C64 so they dubbed it the D128. The D128’s lead designer possessed limited practical abilities. “One of the engineers was Dr. Cong Su, a Chinese guy working in the United States,” explains Dave Haynie. “He was a very clever theoretical guy who should never have been near an engineering laboratory.” Haynie believes they might have attempted some backward compatibility with the C64 sequel. “They had previous failures to make a new, compatible

Commodore 64,” he says. The C64 sequel had more in common with the P128 than the C64, however. “They were calling it the D128, but it was really a spin-off of the P128,” says Herd. “It didn’t have any C64 compatibility.” Surprisingly, MOS Technology did not attempt an improved VIC-III chip with VIC-II backward compatibility. “From a design point of view, once Charpentier left, that was a big deal for MOS,” says Winterble. “The VIC chip wasn’t the most stable thing in the world. It was so complex that after he left, they were afraid to touch it. That was a problem.” The D128 had almost the same specifications as the failed P & B series, except it would use the same video chip as the Unix machine, the 8563. “The D128 was a version without the [VIC-II] chip in it,” says Russell. Like the P & B series, it contained a 6509 microprocessor and a SID chip. Robert Russell managed the D128 team. “We were basically taking a C64, putting a [video] chip from the Z8000 [project] in it, and doing an add-on processor card slot to make it more standardized,” explains Russell. “We were kind of using stuff from the P and B, and blending that together with more C64 compatibility than those machines had.” Without a physical 8563 video chip to work with, the engineers would attempt to emulate the features of the 8563 using other circuitry. Tramiel expected to unveil the project at the January 1984 CES show. * * * Tramiel’s low-cost Sinclair killer had languished since its inception in November 1982. “It had a rocky start in that there was no real ownership of the design,” says Bil Herd. “They had spent five months trying to emulate it using a VIC chip, which was stupid because the TED chip had nothing to do with the VIC chip.” The first engineer on the TED project did not last long when Tramiel realized he was making little progress. “He was too conservative for Commodore,” says Herd. “All he ever tried to do was prove that it wouldn’t work. You can’t exist at Commodore trying to figure out why it won’t work. You’ve got to figure out a way to make it work.”

After the lead engineer left, TED ed to various engineers who were apathetic towards the project. “They weren’t very aggressive engineers,” says Herd. Eventually, the project found a new engineer who showed genuine enthusiasm for the TED. Bil Herd was yet another Midwest Commodore engineer. “I was born in Ames, Iowa in 1959,” he recalls. He was 22 years old when Commodore hired him. Herd first became interested in electronics after his father brought home a gift. “My dad had bought me one of those 150-in-1 electronic kits. In the third or fourth grade, I was making electronic circuits. I taught myself as much as I could with that.” The inquisitive boy began taking apart other electronics devices in the house, such as his father’s light organ (a popular device in the 1970s that pulsed lights to music). His father encouraged him in his learning. “He handed me four dollars after I replaced the silicon-controlled rectifier and I said, ‘No, no, I don’t need your money.’ He said, ‘Son, you should be paid for your talent.’ I thought, ‘Maybe this is something that I could do.’” The family later moved to Indiana. In his teen years, the rock group Emerson, Lake, and Palmer influenced Herd. “Listening to [Keith Emerson’s] music gave me the drive to design a synthesizer to replicate what I was hearing him play,” he says. With encouragement and financial from his father, the longhaired teenager began building his own synthesizer. “He subsidized the build a few bucks at a time,” recalls Herd. Herd’s education came to a premature end when his high school denied him entry into an electronics class. “I was from a small town in Indiana and I was having problems staying focused in school,” he recalls. “When I wanted the TV repair class, they wouldn’t give it to me. They felt I wasn’t academically inclined enough. So I dropped out. I’m a high school dropout.” Herd soon left home, determined to himself with his electronics knowledge. “I was one of those people who went off to see if I could make life harder on myself. I don’t know of any better way to describe it.” The teenager improved his knowledge of electronics on the streets. “I raised

myself,” he says. “At the age of 16 or 17, I had my TV repair license and was fixing electronics for my dinner, literally. I would hang out in a park and fix tape players for enough money to get myself something to eat. You learn to be good when your dinner depends on it.” Herd followed the path of Chuck Peddle and Jack Tramiel before him by enlisting in the armed forces. “One day I actually walked in and signed up for the National Guard,” he recalls. Herd subsequently had his trademark mane shaved off. “I went through basic training and got trained as a teletype repairman at the age of seventeen.” During this time, Herd’s father remained ive. “He didn’t hassle me when I took off from school to work on my synthesizer, nor when I dropped out, nor when I ed the National Guard in lieu of high school.” After six months of basic training, he returned home and received a new offer from his high school. “When I got back they said, ‘Well, we can get you in that electronics class now if you want.’ So I went back and ended up dropping out again,” he says. “After you’ve been in the army it’s hard to put up with the rules of high school. With all that said, they did send my diploma home three years later with my sister.” Herd soon found other ways to combine his love of music with electronics by becoming a roadie. “After I had gotten back from basic training, I did lights for a local rock group in Indiana called Primo,” he recalls. “They were well known throughout central Indiana. Back when you could not buy light controllers, I had built my own. I made my own PCB boards and I had an organ keyboard where each key was a different string of lights. The ones on the end were flash pods, so you had to be careful with those. That was a cool time.” Herd received his first technical position helping to manufacture digital weight scales. “I had taken a job at Pennsylvania Scale Company out in Leola,” he recalls. “We made digital scales, and the guy who had first designed that had been a pioneer in his own right. We used a chip called the 6530 … made by MOS Technologies. We were one of the first customers for it.” Herd made a rapid ascent within the company based on his engineering prowess. “I had been working on the factory floor as a technician and then I worked my way up repairing them,” he recalls. “We had some problems with the circuits out

in production. Rather than just complain about it, I bought data books from Radio Shack and learned enough that I wrote a paper on what was wrong.” The paper earned Herd recognition in the company and he soon moved to design engineering. “I actually got sucked into engineering as an associate design engineer, so that’s how I got into the biz of put-up or shut up,” he explains. “Then I worked my way into full engineer. We had a scale with a seven-segment LED display that was just incredible in the late seventies to early eighties. I was the guy in charge of trying to bring green LED technology and backlight display to Penn. Scale.” Herd also had an opportunity to program. “At Pennsylvania Scale, we did both hardware and software.” The digital scales relied on a small amount of code, which the engineers developed on one of Commodore’s early computers. “They had done the first one on a KIM-1,” he says. On its next project, Pennsylvania Scale acquired an AIM-65 development system by Rockwell. “It’s like a big KIM-1 with almost the same kind of circuitry except it had a dumb terminal, so you could actually type your commands in,” explains Herd. “You would hit compile and it would compile your 2K of code in a couple of hours, and it was great!” The AIM-65 captivated Herd and he began spending more time with it. “I started dragging it home at night and bringing it back every morning. They weren’t small.” The late-night sessions made Herd an expert in 6502 machine code. Bil Herd now held much of the necessary knowledge to develop his own computer systems. In many ways, he was an engineering version of Tramiel. He was unschooled in his field, but in some ways more impressive than his collegeeducated contemporaries. “I think there is probably a little Jack Tramiel in me these days,” he reflects. He also continued his National Guard duties during his early career, attached to the 103rd Medical Battalion. With his unruly hair and even more unruly behavior, he received a nickname. “My handle when I was in a tank unit in the National Guard was Animal,” he recalls. “I had it painted on my CVC helmet. It had started as a derogatory reference that I took as a compliment.” Herd grew back his long hair in the later days of his service. While in uniform, he pinned up his hair and wore a shorthaired wig. In 1982, the Pennsylvania

National Guard awarded him the Army Medal of Commendation. “I was told that I was the only soldier to receive an ArCom while wearing a shorthaired wig and long hair,” he recalls. In March 1983, Herd’s position at Pennsylvania Scale grew tenuous. “I had a falling out with my boss,” he says. “[Hedley Davis] was reading the paper and he said, ‘Hey, do you want to go get a job interview?’ I said, ‘Where?’ He said, ‘Commodore.’” * * * As Commodore began developing several projects simultaneously, Robert Russell became desperate to find new engineers. Compared to the Commodore of old, the new Commodore was about to go large. “Teams were real small up until the C64 started expanding things,” says Russell. Commodore hired a professional service to find more engineers. “They would go to these headhunters and the headhunters would put eight or ten people in front of them every night,” says Herd. “They would go up and down Highway 202 combing these different places looking for talent. We used to call it Silicon Valley East.” Herd was one of many candidates interviewing for a job. “That night, Bob Russell and Frank Hughes were there,” he recalls. Herd first interviewed with Hughes, the prickly lead engineer of the Unix machine. With a lack of formal education, it was easy for Hughes to dismiss Herd as a wannabe engineer. Herd feels a personality clash made his first interview fail. “Frank didn’t like to feel like he wasn’t the most important person in the room,” he says. “I ended up talking about how this thing called a magneto-optical drive was on the horizon and how exciting it was. He was like, ‘Yeah, uh-huh. Okay, I’m done.’ I was like, ‘Crap, I blew it!’ I was so excited and I wondered, ‘Why couldn’t I talk to this guy?’ I thought he would be a good electronics guy.” Herd sensed his interview went poorly, but he had one more chance. “All you needed was one of the two guys to invite you in,” he explains. “I was done with Frank Hughes and it was a no go. I interviewed with Bob Russell next and he was just beat-up from interviewing people.” Russell discussed a project at Commodore and mentioned several op-codes from

the 6502 chip. Each time he mentioned one, Herd parroted the hexadecimal instruction. “I was repeating the machine code to the instructions he was saying, under my breath,” says Herd. “Bob was a pretty observant guy and he caught on to it, and the interview was over.” It was obvious Herd knew 6502 assembly cold. This was enough to persuade Russell to set up another interview. Herd was now on his way to visit MOS Technology. “They made an appointment for me to go to King of Prussia and actually interview at Commodore,” he explains. “They brought me into Commodore and had me interview with Shiraz Shivji.”

Shiraz Shivji discussing computers at Commodore.

Herd made a poor first impression. “I jumped into Shiraz’s chair because he had his desk facing the opposite way that I was used to,” he says. “He walks in and says, ‘Excuse me, you are in my chair.’ He was holding it against me! I thought, ‘Oh no! I’ve blown it again.’” Herd also appeared unprepared. “Shiraz says, ‘Do you have a resume?’ and I said, ‘No, I didn’t bring one in with me.’ They hadn’t told me I was interviewing with other people when I was coming in, so I didn’t bring resumes. The way Bob [Russell] was so nonchalant about it, he said come in to talk to me or something, and he already had my resume. Frank came to my rescue and brings in a resume from a couple of nights before.” Herd began talking with the engineers about a personal electronics project. “I showed my ingenuity and love of electronics and that’s what finally got across to them. That’s what got me hired,” he says. He was now a Commodore employee. * * * Shivji hired Herd in March 1983 as a mere technician to aid the design engineers. He did not have an engineering degree, but he did not feel any less capable. “I had never been to college, but by the age of 22 I actually had about four years of experience as an engineer, which felt like a lot back then.” Most of Herd’s coworkers were also relatively new. “I started at the MOS building in King of Prussia where the walls were this bright blue color and there were three of us to a room,” he recalls. “When I got there, there were literally only the holdovers like Yash Terakura and Bob Russell and some of those guys from the California Commodore days.” Herd initially approved of the casual atmosphere of Commodore, where employees were addicted to C64 games. “My first day I noticed people playing games in their offices as you walked by and no-one saying anything,” he recalls. “I liked the thought that you would be judged on your work output and not the

appearance of work. Later, I found out that a good number of the people I saw playing games didn’t really do anything useful. So much for the real version of utopia.” On his first day working at Commodore, Herd sat in a room with engineer Benny Pruden, the ROM programmer for the Pet 4040 drive. “When I got there, they didn’t know what to have me do, so they were going to have me do code for one of the disk drives,” he recalls. Shivji had decided on a software role for Herd because of his knowledge of assembly code. It was a disappointing appointment for someone who loved hardware his whole life. Herd had a momentary reprieve when Pruden departed on a research trip. “The guy who was supposed to show me the boring software wasn’t there that week,” says Herd. Instead, Pruden left him with a stack of books to read. With his lead engineer gone, Herd decided to wander around MOS Technology. He stumbled on the lab where the 6502 originated. As he looked around the empty room, he felt saddened. “I had just missed them. The cigars were still burning in the ash trays, the seats were still warm, but there was no one around anymore.” On the wall, he spotted a curious photograph. “There had been a picture of the PET design team hanging up and they would black out the image of everyone that was no longer with Commodore,” he recalls. “In the end, there was only Yash Terakura and Michael [Angelina] left amongst spooky black outlines.” Eventually, Herd wandered into the systems lab, where engineers were busy with Tramiel’s latest low-cost project. “I asked, ‘What’s this?’ They said, ‘This is the TED.’ I’m like, ‘Cool! Are you doing bus sharing?’ I just did something at home like that. I wire wrapped my own 6502 board with bus sharing. I think eyebrows went up as I’m peering into it and poring over the schematic. That’s how I got picked to be the project leader for it.” “Bil was originally hired as a super-tech and he just fell into this job of being the chief systems guy at Commodore,” marvels Dave Haynie. Herd was no longer in software engineering. “The first week I came off of doing code for the disk drive to being in charge of the TED line,” he says. When Benny Pruden returned from his research trip, he must have been shocked to find

his new hire in charge of systems engineering. “I took the hardware path, thank God, because that’s my favorite part.” Herd now had the responsibility of delivering Commodore’s next consumer product. With no formal education and lots of bravado, he now held one of the prime jobs at Commodore. It was a rapid ascent for a new engineer. * * * The first thing Herd noticed on the TED was the lack of a circuit to reset the computer. To make the reset work properly, he needed one more chip. “The TED had one stipulation: they said you need to do it in nine chips,” he recalls. “Nobody wanted to go tell Jack that they needed to add a chip. Well, I didn’t care. I said, ‘You almost did it in nine, we need ten.’” Shivji reluctantly ed the news to his boss. “Somebody went to Jack saying, ‘We added one last chip to make the reset work right.’ I’m told he just said, ‘Fine.’ In other words, he didn’t care,” explains Herd. “He got them to do what he wanted them to do, which was make it in ten chips, not 28 or 30 or some number like that.” Robert Russell was curious if his new hire had the ability to design a complete system. “Honestly, I wasn’t too sure about Herd to begin with, but he knew his stuff,” he recalls. “Once he came in and started working, geeze! He’s a guy that gives 200% and then has a good time in between.” Although Russell and Shivji enjoyed relative autonomy in the months following Charles Winterble’s departure, Tramiel eventually replaced Winterble with a new systems manager. Russell now reported to engineering manager Joe Krasucki, who gained a callous reputation. “He came in … and he was stuck in my office, smoking a frigging cigar the first day he was there, which I never liked,” says Russell. “He was an asshole manager who may be able to do some hardware design.” Tramiel attempted to change the structure of Commodore engineering so MOS Technology no longer designed the systems. “Before the new Commodore, everybody was working for MOS Technology in the chip group,” says engineer Dave Haynie. “This was as things were changing into their modern form, where you had separate groups for software, systems, and chips. At that point, every director reported to the vice president of technology, Joe Krasucki.”

Krasucki now held a powerful position. Prior to Commodore, the engineer worked for Sperry where he helped engineer the Sperry System 80. Before that, he worked for Burroughs where he engineered military mainframe systems. The engineer earned an advanced degree in electrical engineering from Villanova University near Philadelphia. Herd formed an agreeable working relationship with Krasucki. “I was the one guy who got along with him,” he says. “When he came in, everybody hated him. I heard him coming down the hall, asking people what they were working on and they would be like, ‘Why are you asking that? What do you need that for?’ Well, I had all morning to hear what he was after, so by the time he walked into my office, I had it all ready to go. That helped me advance because I didn’t fight him.” Most managers at Commodore had a difficult time retaining their jobs with Tramiel around, but Krasucki held on. “If he didn’t have people like Bil Herd and Greg Berlin to slug it out and make it work, he never would have got anywhere,” says Russell. Much of the success or failure of engineering VPs had to do with whether their engineers could deliver results, and Bil Herd would prove that he could deliver. “I was his cash cow,” he says. Predictably, Herd’s rapid ascent to lead engineer perturbed other engineers who had been with Commodore longer. “There was some resentment from some other people,” says Russell. “Some people said, ‘Oh, he’s just a favorite of Joe Krasucki.’” However, Herd was quick to win over these detractors with his disarming sense of humor and fierce work ethic. He did not back away from a conflict, even with the upper echelons of Commodore management. Sometimes the conflicts turned physical. “[Krasucki] worked for Lloyd Taylor, who was the vice president under Jack,” says Herd. “Lloyd Taylor was this big strapping guy, kind of obnoxious and rude. His name had been Red, but he had grey hair by then. Everybody was afraid of him.” Red Taylor was one of Tramiel’s inner family. “He had this old car horn. When he wanted his secretary, he would reach down and honk the rubber ball to the horn to call her,” recalls Herd. “She’d roll her eyes and walk in there.” The drywall at MOS Technology barely stood a chance when Herd was around.

One day, the engineers heard a series of loud thumps. “Three heads stuck around the corner into the hallway to see what was going on, and me and [Lloyd Taylor] were literally wrestling in the hall,” says Herd. “I’ve got hold of his leg by the knee and I’m slamming him into the wall.” To the Z8000 engineers, it looked like an Alpha-male Klingon ritual. “He was kind of one of those brutish guys, so at some point our personal zones crossed each other and we got into a wrestling match,” explains Herd. “He pushed and I pushed back. So much for being afraid of the vice president.” The altercation had no ill effects on Herd’s career. “We were laughing while we were doing it,” he recalls. “I still him walking back to his office with his shirttail hanging out. After that, we got along great because I think he respected somebody who stood up to him a little bit.” * * * Surrounded by engineers grounded in theory, Bil Herd was a powerful force, respected because he knew practical electronics like no one else. According to Herd, the universities that educated his coworkers often overlooked practical aspects of hardware design. “What they weren’t teaching in school was how to tell how hot your part would run and the real life situation of it.” Day by day, Herd proved his value to the team. Herd’s unending source of energy amazed his engineers. “He was a wild-man back then,” says Haynie. “He would be working for days straight and hoped everybody else would be along for the ride too. It was one of those things where you’re young, you’re single, you’ve got no kids, and you would party as hard as you worked.” Unlike many engineers, Herd was a genuine extrovert who spent as much time interacting with coworkers as with electronics. “Herd got along with people,” says Russell. “He fit the loose Commodore culture but he worked his butt off if he wasn’t drunk or high on something.” At the heart of the TED computer was a processor with a new name, the 7501 8bit microprocessor. The 7501 contained the 6502 core, but it used the new HMOS design (High-speed NMOS). The processor would now run at 1.76 megahertz.

With Al Charpentier and Bob Yannes no longer at MOS Technology, the title of resident semiconductor genius fell to Dave DiOrio. “DiOrio was great. He was my counterpart in the chip designers as far as nailing this stuff,” says Herd. “I knew I was working in a cool place.” DiOrio’s goal was to reduce the number of chips inside the TED. He set out to design the TED chip as a low-cost graphics and sound chip similar to the original VIC-I chip, but with higher resolutions and more colors. This was one of the last chips to follow in the grand tradition of friendly three-letter names. Originally, MOS Technology attempted to create a prototype of the TED chip using a VIC-II chip. “They dicked with a wire-wrap board based upon the VIC chip which didn’t relate to the TED design at all,” recalls Herd. “I threw the wire-wrap away without ever turning it on.” The TED chip was a competent chip for the time. It displayed 128 colors—16 colors in eight different shades. In effect, TED eliminated the need for five separate chips. The main deficiencies were the lack of sprites and inferior sound.

Dave DiOrio (left) and Bil Herd drinking and deg (photo courtesy of Dave Haynie).

As with any new semiconductor chip, Herd had to deal with problems in the early revisions. “The TED chip had to go through many revs before it was usable, mostly due to the lack of checking tools,” he recalls. “I the day we got TED up with a processor. The system would come up and then puke on the screen. I sat and stared at this thing for a couple of hours. There really wasn’t a lot of tools at my disposal so I had to think about what, if anything, was fixable in the mess.” Later that night, the engineers gathered at engineer Terry Ryan’s house for a casual get-together. “That night was a bad beer and bad movie party at Ryan’s featuring Schmitz beer, Plan Nine from Outer Space, and Attack of the Killer Tomatoes,” says Herd. “I took out my frustration by challenging a couple of people to keep up drinking with me.” Herd, a self-described alcoholic, had an unfair advantage over the other engineers. “At one point, a junior programmer named Mike Isgar went running for the backyard to throw up, only he didn’t get the screen door open before going through it,” he says. “I heard a story about my first ‘programmer kill’ that night and went with that as a theme of the differences between men and programmers.” After his first programmer kill, the engineers began to look at Herd as a carnivore thinning out the pack of programmers. Terry Ryan told Herd, “Try not to kill more than you can eat.” Herd functioned much like Chuck Peddle, who found parties to be a good atmosphere for coming up with creative ideas. In the back of his mind, he continued thinking about the TED chip’s problem. “That night I figured out what the problem had to be,” he explains. “I walked in the next morning, moved the wire, and had it working in five minutes. Alcohol and problem solving seemed to go hand in hand for me. That’s the problem with being an alcoholic.” Terry Ryan developed most of the ROM software for the TED computer. A friendly rivalry developed between the software and hardware engineers. “Terry

Ryan was the programmer who wrote BASIC version six,” recalls Herd, who found the programmer standoffish at first. “He was like, ‘Leave me alone, I’m working.’” Herd decided to use 16-kilobyte DRAM chips. MOS Technology bought the design from a company called Micron, but manufactured the chips themselves. “I they couldn’t get this one spec down and when they called Micron about it and they were like, ‘Yeah, we couldn’t either.’ So they ended up buying a bad DRAM process,” says Herd. Predictably, Tramiel sued. “We had sued Micron and a lot of other companies,” says Herd. “Part of it was because Micron made them wrong and part of it was because we didn’t use them right. That’s what happens when you have chip designers doing hardware design. But Jack Tramiel got the lawsuit to stick. It was pretty cool.” Ira Velinsky, the engineer responsible for the P & B series cases, created the TED case. Tramiel called him his Jewish-Japanese engineer. “He worked out of Tokyo for Sam [Tramiel],” says Herd. “He looks every bit Jewish, he speaks enough Japanese to get in lots of trouble, and he was a real character. I’d go over there and he’d get me in trouble. You had to learn not to trust what he told you the Japanese meant at first.” The TED computer received a proprietary joystick port, despite the popularity of the standard D-Shell connector by Atari. “The reason given to us by Ira Velinsky was that it was the only place it would fit,” says Herd. “Sure enough, if you look at the case, you would not have room for two of the D-shell connectors.” The small TED case was reminiscent of the computer it sought to overthrow. “The original case was this tiny sloping little Sinclair case,” explains Herd. TED would also receive a new joystick by Commodore. “Ira Velinsky had designed another joystick and brought it in,” says Herd. “This was from a softtool, so it was a $40,000 [prototype] joystick. It had a handgrip which was pretty cool but it came down to a real skinny 3/16 of an inch diameter before it went into the base of the joystick.” Herd found fault with the design. “Everybody is saying, ‘This looks like it will break.’ And Ira is saying, ‘No no, that will never break. It’s been designed.’ That was the word he was using.”

“Everybody is saying, ‘Ira, this is going to break. Kids are rough. It’s got to last a year.’ He said, ‘No, no, it’ll never break. I guarantee it.’” Velinsky held his ground, so Herd took the argument to a new level. “Then we got onto something else and I’ve got the joystick in my hands,” recalls Herd. “And you heard, ‘Snap!’ You can see me pushing this broken joystick back into the middle of the table. It’s true, I had to bend it a little harder than I thought I would, but the point was if I could break it sitting there, it was going to get broken a lot. So I broke a $40,000 soft-tool prototype to prove a point.” As summer approached, progress on the TED slowed down. “TED slumbered for many months until the summer before the CES show,” says Herd. Commodore was about to undergo a drastic reorganization. [1]* Decades later, Unix popularity surged in the form of Linux. [2]* The 8563 designation may indicate it was a continuation of the TOI chip started years earlier by Bil Seiler.

CHAPTER 33

Ted 1983

After Chuck Peddle and his group of engineers departed in 1980, there was no one left to further the PET line of computers. Commodore released one further model, the SuperPET CBM9000, developed in conjunction with the University of Waterloo in Canada. It was the last product in the evolution of the PET. In the spring of 1983, Commodore employees travelled to Hanover, on the PET Jet. “On a trip to Hannover Messe, we flew from Pennsylvania to Teterboro [airport in New Jersey] and then a stop in Boston to pick up [Commodore president Jim] Finke,” says John Feagans. “We were carrying the prototypes for the SuperPET.” Already the short flights across the east coast had eaten into the PET Jet fuel reserves. Transatlantic flights were often nerve-wracking owing to the limited range of 2,770 miles for the PET Jet. “I looking at the pilots fuel gauge saying it had only 20 minutes left as we were leaving Boston,” recalls Feagans. Luckily, the PET Jet had the ability to tap into additional fuel reserves to complete the 2,440-mile flight to Reykjavik. “My jump seat was full of extra fuel tanks for the Atlantic crossing. We ended up in Iceland for the evening and completed the flight to Braunschweig the next day.” The engineers planned to land in Braunschweig, a mere 67 kilometers from Hanover, in order to visit a local Commodore factory. “We flew over with my engineers in the corporate jet carrying all the prototypes for the Hanover show,” recalls Robert Russell. “They didn’t have enough fuel to fly there so they diverted to Hanover.”

Unknown to Russell, the townspeople of Braunschweig made plans for receiving the Commodore jet. “The jet had never been in before,” explains Russell. “They thought Jack was coming.” The diversion to Hanover saved Russell and his engineers some potential embarrassment. “The thing that killed us was the pilot came on and said, ‘Oh, you just lucked out because they had the mayor, some dignitaries, and a band thinking Jack was going to be on the plane.’” According to Russell, no one adored Commodore more than the Germans did, especially near the Braunschweig factory. “Those people still it with great pride,” he says. Jack Tramiel, Sig Hartmann, Andy Finkel and other Commodore employees came out on a later flight. It was Hartmann’s first show as a representative of Commodore, and Tramiel wanted to make sure he had a good time. Andy Finkel recalls the introduction. “I got to act as chauffeur for him and Sig Hartmann in at the Hanover Fair one year,” he says. “We had a really nice fast German car on the Autobahn taking them from Braunschweig where Commodore had its offices to the Hanover Fair.” Russell first met Hartmann during the show. “I was at the Hanover show, sitting up in the booth in case anyone needed specifics,” he says. “Jack came to me and said, ‘Hey, take this friend of mine Sig around and show him what you find interesting at the show, Bob.’” Harald Speyer, general manager of Commodore , continued his marketing success. “He was the guy who drove the C64 in Europe into being such an important product,” says Chuck Peddle. “Harald drove Commodore to the top of the pyramid with the work he did with sponsorships. He sponsored an around the world race and two or three things that put Commodore on everybody’s lips all over Europe.” Russell agrees. “He made the whole business thing,” he says. “We were bigger than IBM for a long time in as far as product sales, volume, and reputation. We were fighting with HP for quality reputation.” Speyer was in high demand in Europe, but he remained loyal to Commodore. “IBM was killing themselves trying to get him in the early PC days to come over to their side because he was a smoking marketing guy,” recalls Russell. “He

wouldn’t work for IBM in because he considered it a step down from Commodore.” After the show, Tramiel and Hartmann celebrated on the return trip to Commodore’s German offices. “On the way back, they sat there in the backseat of the car drinking and telling old stories,” recalls Finkel, their driver. “That was kind of fun. They’d be in the back seat singing songs in German.” The songs surprised Russell. “I would listen to them singing their German marching songs. I thought, ‘That’s pretty bizarre,’” he recalls. “That’s what he ired about the Germans when he was a young kid in Poland. He ed the songs.” “Jack told me when he was a young child and the Germans invaded Poland, where he was at, that he was impressed with the German army because they were all in shiny uniforms marching in sync, singing the German songs,” says Russell. “As a child, he thought that was great when they came into Poland. Apparently it was quite a grand parade marching through.” * * * Although Paul Goheen was responsible for producing a line of business applications for the C64 and B series of computers, John Feagans, who still resided on the west coast, initiated Commodore’s most successful productivity application. In the early eighties, computer enthusiasts near Silicon Valley were taking notice of the graphical interface (GUI), first developed at Xerox Palo Alto Research Center (PARC). The idea of a GUI intrigued Feagans while he was still working on PET computers. Xerox PARC veteran Robert Metcalfe had visited Commodore in 1980, and relayed the amazing developments occurring at the research center. “Perhaps this was the inspiration [for deg a GUI],” says Feagans. The Apple Lisa GUI is an example of engineers copying the Xerox PARC work after viewing it firsthand. John Feagans never saw the GUI—he merely heard it described. Nevertheless, Feagans did his best to implement the GUI in a project for Commodore. “I was playing with GUI concepts on a PET,” he says. “My first GUI was an animated filing cabinet done entirely in PET graphics characters.” Feagans’ GUI shared similarities with a common menu bar. It borrowed from the

index card paradigm, such as library card catalogues found at libraries at the time. His GUI for opening files literally looked like something in an office setting. “Open a drawer and you saw tabbed folders,” he explains. “Select a tabbed folder and pages with file names appeared.” After the closure of the Moorpark R&D office in 1982, Feagans had little structured work to do and he began developing his GUI project for the C64. The C64 did not have a mouse peripheral at the time, but it had a joystick port, which allowed Feagans to control a mouse pointer on the screen. He recalls, “Using the joystick to navigate, you clicked on a text document and it displayed the contents on the screen.” While Andy Finkel was out scouting for projects for Commodore, he saw a demonstration of Feagans’ GUI and suggested to Sig Hartmann that Commodore should develop it into a full product. “When Andy came out to work on the C64 with me, we created all the other desktop objects as sprites,” says Feagans. “I wrote the original concept in BASIC, and then used a compiler I wrote to turn it into assembler.” Feagans already had a system to manage files using a GUI and a joystickcontrolled pointer, but he, along with Andy Finkel, designed a simple wordprocessing application. He called his application Magic Desk. “John was the type of guy who liked to design things,” says Russell. “He did some really interesting graphics stuff later on for the C64—the Magic Desk.” The programmers were in a rush to have his product ready for the June 1983 CES show. They called on Michael Tomczyk to fly down to the west coast to help with the design. Tomczyk hastily drew up the first screen on graph paper during the flight, which showed a desk with a clock, a door, a wastebasket, and other office accoutrements. It was -friendly, but it took the GUI metaphor too literally. Though it was primitive, it was a preview of next generation interfaces. Feagans worked out of at 3330 Scott Boulevard, the same building as Jack Tramiel. Minor earthquakes were a regular occurrence in California, but engineers designed the bunker-like building to handle earthquakes—in theory. “There were huge 20 foot glass panes,” says Russell. “During an earthquake, when those front s flexed, everybody was running towards the [walls] because they didn’t want to get toasted.”

On May 2, 1983, the Coalinga earthquake hit California. “When the earthquake started, it was flexing up and down and the windows were actually bowing. Of course, the only one that ever broke was in Jack’s office,” says Russell. The glass shattered over Tramiel’s desk. Luckily, the earthquake struck close to midnight when hardly anyone occupied the building. * * * Just prior to the June 1983 CES show in Chicago, MGM released WarGames starring a young Matthew Broderick. The cold war film glorified computer hacking, a subject briefly touched upon in Tron a year earlier. It went on to gross almost eighty-million dollars in the United States alone, a reflection of mainstream interest in computers. With so many competitors in the microcomputer industry, a price war was inevitable. In February 1983, Tramiel’s old rival Texas Instruments had reduced the TI-99/4A to $150. They were going head to head with the VIC-20. In May 1983, Atari offered a $100 rebate on the Atari 800, bringing it to the same price as the C64. It was time to retaliate. Tramiel wanted a major announcement for CES, and he knew exactly how to generate it. However, he was mystified at the extremely low price of the TI99/4A and ordered his engineers to dissect one of their computers. “Our engineers had back-engineered the TI-99 series of home computers and we confirmed that there was no way they could manufacture that product at a profit,” recalls Tomczyk. “We determined that they were making all of their profits from software and peripherals, and losing money on the U.” Tramiel made it his mission to match Texas Instruments in price, even though the C64 was technologically superior to the TI-99/4A. Since the January price cut to $399, many factors allowed Commodore to produce the C64 cheaper. MOS Technology began to achieve truly high yields of the VIC-II chip. The only other major components in the C64 were the SID chip and memory, which was rapidly falling all the time. On top of that, he paid nothing for including Microsoft BASIC in the machine. Tramiel now had a dramatic announcement for the CES that would leave the competition gasping for air. Commodore sold a million C64s sometime in the first half of 1983. The

engineers decided to flaunt their success at the CES show in June. According to engineer Bil Herd, “We were going to the show and somebody said, ‘You know what? We need to see the millionth one.’” Realistically, no one was waiting on the assembly line to pull out the millionth C64. “I just happened to walk back into the lab and Gayle Moyer was back there rooting through this big pile of trash that accumulates in these labs,” recalls Herd. “I asked, ‘What are you looking for?’ He said, ‘I’ll know when I find it.’ Then he goes, ‘Aha! Here it is!’ He pulls out of this pile of junk the millionth Commodore 64!” “In reality, it was a random one,” says Herd. “He took it, sprayed it with gold, and put it in the display at the CES show.” At CES, Microsoft watched as the computer with their BASIC achieved new sales records. “The Commodore 64 must have really rankled Microsoft because I don’t think they made a dime on it, even though there were millions and millions of Commodore 64s sold,” explains Bob Yannes. Tramiel still used the same contract written in 1977 for the PET. “I’m sure that the people who were on the other side of that contract weren’t real happy,” says Charles Winterble. “The relationship between Commodore and Microsoft wasn’t that great.” Tramiel used CES to study new products from his competition. This year, the maker of the popular ColecoVision game console announced the Coleco Adam home computer for $525, including a printer. The new computer introduced the concept of a suite of built-in software applications, which Coleco called the Electronic Typewriter. The idea intrigued Michael Tomczyk and Sig Hartmann of the Commodore software group. Tramiel was intensely interested in Texas Instruments. Of his competitors, only they had vertical integration down to the semiconductor level, allowing them to compete with Commodore on price. Industry analysts had a favorable image of Texas Instruments, largely unearned according to Michael Tomczyk. “There was a huge amount of falsehood and really bad fact reporting among the research houses,” he says. “In 1983, Business Week showed a pie chart that showed Texas Instruments with a bigger market share than Commodore and we knew that to be patently false. We told

them that and they ignored us.” The president of TI’s home computer division, Jerry Junkins, came to the show in full force. “At the summer 1983 CES, we were in a hundred-thousand dollar booth next to a million dollar booth that Texas Instruments had,” recalls Tomczyk. “They had this huge gigantic two-story black and silver booth with all their products including their computers. And we had a little tiny grey carpeted booth. It looked like we were serfs next to the castle.” Commodore employees felt humiliated next to such an ostentatious display by one of their chief rivals. “We were beating TI in the home computer market and we knew it,” says Tomczyk. “So everybody just felt really chagrined that TI was acting like they owned everything. We were really pissed about that.” Tramiel sensed his employees were becoming demoralized due to the huge booth, and decided to do something impulsive in order to boost their spirits. From his engineers’ dissection of the TI-99/4A, he knew Texas Instruments was not making profit on their computer. At CES, they further reduced the price to $99, when it actually cost around $125 to manufacture the product. “We knew that if we took away their profit on the software and peripherals, we would kill the product,” explains Tomczyk. “So he translated that into a very dramatic and risky business strategy.” Considering Tramiel already had a price announcement for the C64 that would stun the industry, it seemed like supreme overkill to do any more. “He walked into the booth and said, ‘I’m going to cut the cost of all of our software and some of our peripherals in half,’” says Tomczyk. The sudden decision left Commodore employees stunned. Although the CES booth was relatively modest, Commodore rented the S.S. Milwaukee Clipper, a lavish 360-foot enger ship, harbored off Navy Pier in the Chicago Harbor. “The Milwaukee Clipper was a ferry that ran from Milwaukee, Wisconsin to Muskegon, Michigan until it was retired and berthed in Chicago,” says Feagans. “Commodore rented the ship and painted the side facing the pier to use for private exhibit space at the 1983 Summer CES.” Commodore rechristened the ship “Commodore Clipper” and painted over the old name on the bow. Commodore hosted parties on the boat, conducted media briefings, and

displayed Sig Hartmann’s new software lineup. “We featured ‘Boatloads’ of software,” says Feagans.

Marketing software on the Commodore Clipper.

The director of business applications, Paul Goheen, brought along an uninspired line of business applications written in BASIC. The applications included Easy Calc 64, Easy Finance I through V, Easy Mail, Easy Spell, and so on. They also produced a series of small business applications with names like General Ledger, s Payable, s Receivable, and Inventory Management. Unfortunately, BASIC math was limited to performing calculations on numbers less than 10,000,000. The titles met with limited success, and Commodore ended up with excess inventory. During Commodore’s press briefing, Tramiel announced he would reduce the price of the C64 from $360 to just $200, one-third the original price a year ago. Intense free market competition had produced a consumer paradise. Tramiel also made his announcement to cut the price of software and peripherals in half. Sig Hartmann could not have been happy to see Tramiel undermine his profit model, from which he had been basing his forecasts. Tramiel’s announcement brought an immediate response from TI’s Jerry Junkins. According to Tomczyk, “As soon as the announcement went out, within hours, the president of TI’s home computer division came storming into our booth and shouted at Jack, ‘You can’t do this! There is no reason to do this. Why would you do something so reckless?’ And Jack just grinned.” That afternoon, Tomczyk learned that TI was pulling out of the home computer business. “The president of the consumer products division put everybody on a plane and took them back to Texas,” he recalls. “On that Friday, a few days later, they announced they were exiting the home computer market. That’s how fast it happened.” Tramiel’s impulsive decision would later have unintended consequences that cut into Commodore’s profit margins. As a result, his CES announcement would ultimately undermine his relationship with Irving Gould. “That was where the

initial seeds were sewn,” says Russell. Soon, Texas Instruments predicted they would lose $100,000,000 in the second quarter of 1983 alone. It was the first loss for TI since its inception in 1930. Meanwhile, Commodore showed a profit of $25,000,000 in their third quarter. It was clear Commodore could outlast its strongest competitors. While the C64 sold well, its offspring did not. The Executive 64, or simply SX64, which was the production name of the portable Osborne-killer computer, had been selling since May 1983 with very few orders. In total, the machine would sell approximately 10,000 units before Tramiel discontinued production. By that time, Tramiel no longer needed an Osborne-killer. The Osborne Computer Corporation was already struggling with bankruptcy. Commodore also displayed a plug-in keyboard synthesizer called Music Synthesizer and a drum set called Digi-Drum, both for the C64 and available for $59.95 each. After the show, Commodore had an executive shakeup. Commodore often strategically fired executives after major trade shows when business politics would get lost amid the other news. In June 1983, the board of directors dismissed company president Robert H. Lane, the man who had persecuted Kit Spencer Typically, the president of a company deals with the day-to-day operations, however that was impossible with Tramiel around. “Jack was very much the man in charge of day-to-day operations and all technical decisions,” says Spencer. Lane was the latest in a string of presidents that lasted only a year or two. Gould saw Lane as a way to inject professionalism into Commodore. It was his hope that he could keep Tramiel and use Lane to transform Commodore into a Fortune 500 company. An interim president, Don Richard, took over while the board searched for a replacement. Tramiel also abruptly terminated the PET sales organization, a clear indication he wanted to focus exclusively on the consumer market. Commodore also weeded out their smaller computer dealers. Now, instead of thousands of small computer stores, Commodore only dealt with approximately 65 major distributors.

* * * Prior to the departure of Charles Winterble and his team, Tramiel had discussed reorganizing the company headquarters at one centralized location on the east coast. “We were just getting ready to move down to the new facility,” says Winterble. Tramiel had his reasons for the move. “He just recognized that it was a lot less expensive to get people in Pennsylvania than it was in California,” says Al Charpentier. “I think he also was thinking that he was going to use the big facility for manufacturing.” Tramiel began making plans to move to a larger facility, but he did not want to interrupt operations during initial push for the C64. The new premises would relieve the cramped conditions in the MOS Technology building. “We were at the limits of our building,” says Russell. “We had two years where we were on top of each other after the VIC-20. There was no more space for anybody. When the C64 became real popular, all they did was put more people in the same office space. West Chester was really the thing that relieved the pressure.” Up until 1983, Commodore had been a diffuse collection of operations dispersed across California, Pennsylvania, Texas and even Arizona. Now Tramiel moved his company into a massive facility in West Chester, Pennsylvania at 1200 Wilson Drive. “Jack’s idea was to consolidate everything in the West Chester facility,” says Winterble. The facility had a history of manufacturing. “It was an old Zenith TV tube manufacturing facility,” says Charpentier. “That was an expensive building. Commodore went in and gutted it and renovated it.” According to Russell, Hallmark Cards owned the company after Zenith. “I went there when it was full of frigging greeting cards because it was a greeting card manufacturing plant,” he says. In the summer of 1983, Commodore employees moved into the massive building. “It was a two story building and 585,000 square feet,” says Russell. “It was mostly production. In the front of the building they put engineering, corporate, software, and those things.”

Commodore used most of the building for manufacturing and warehousing. “It was an assembly line and it had huge amounts of storage space,” says Dave Haynie. “They did all of the final assembly for the Americas there.” The size of the building amazed Russell. “That building was so huge compared to what we were used to,” he says. “We opened up some warehouse doors and there was a whole atmosphere environment and rain clouds in some of those huge rooms.” The move to West Chester was hectic, especially with C64 production at a peak. “We had just a lot of irons in the fire at that point in time,” says Russell. The move also disrupted system designers like Bil Herd, who was hard at work on the TED computer. “In the summer to late fall I trying to get prototypes built while we were moving,” he says. Russell and Shivji continued their hiring spree to restock the Commodore talent pool. “Shiraz and I just interviewed everybody in creation,” says Russell. “From 1982 until 1985, I hired about 80 people. We had a full boat at West Chester. We were always trying to get personnel, but it was hard to find anybody who knew anything.” It was up to Russell to move Commodore’s development systems. “I was in charge of moving the VAX computers out of the MOS facility to the [West Chester facility],” he says. “I guess I was the highest low-guy on the totem pole.” The new West Chester facility created self-sufficient community. It even contained a small electronics store to sell Commodore products to employees at discount rates. “They would sometimes sell liquidation things as well as just the normal everyday Commodore stuff,” says Haynie. “You could buy Commodore calculators there, which was pretty much the only place in North America that still sold them. Anything that was being discontinued or liquidated would show up at the Commodore Employee Store.” Almost everyone on the west coast who refused to move to the east coast became unemployed. “Even though Commodore itself was doing fine, Jack and Irving had decided that division was no longer necessary,” explains Andy Finkel. “So all those people had a mass layoff.”

Since very few Californians wanted to live on the east coast, he terminated most employees in the group. “He was constantly firing people and paring the organization down,” says Tomczyk. “There was a tug of war between Jack and the rest of organization. He would say things like, ‘If you turn your back on an organization, in a couple of weeks it will double in size.’” The old Santa Clara offices were now no longer the company headquarters, but they remained open. “The facility in 3330 Scott Boulevard, Santa Clara was manufacturing,” says Feagans. “It built PETs, VIC-20s, C64s, and warehoused product coming from Asia. The MOS Technology sales rep that serviced Atari and other Silicon Valley s was there as well as the regional sales office.” Although Tramiel moved the bulk of his employees to the east coast, he remained close to home with his family. “Jack Tramiel never moved to West Chester and only visited there,” says Feagans. “He always maintained his office in Santa Clara.” The offices at Scott Boulevard were a mere 20 minute drive from his family home in Monte Sereno. John Feagans refused to come to the east coast, but Tramiel could not stand to lose an inner family member. “Because of his relationship to Jack, he was the one person left who stayed to help an orderly transition of the west coast intellectual property and stuff to the east coast,” says Finkel. With the larger workforce, the PET Jet began to reach its limits too. In 1983, Gould and Tramiel decided to upgrade the 7-seat Westwind jet to a British made Hawker-Siddeley HS-125 (known as the British Aerospace BAe 125). The new jet seated two crew and up to 14 engers. * * * In the summer of 1983, Bil Herd began hiring engineers for the TED project. “I went in with Joe Krasucki, my boss at the time,” recalls Herd. “We went back to the same office where I had been hired, to do the same thing where they run a dozen people by you every night.” Herd recruited two of his former coworkers, Hedley Davis and Terry Fischer. “They came after I did, from Pennsylvania Scale Company,” recalls Herd. “Hedley Davis was a well known programmer and designer. He came out of school with an electronics engineering degree.” Terry Fischer, a technician, created printed circuit board designs.

Herd also hired a young engineer named Dave Haynie. “In 1983, I was brought in to work on the TED project,” recalls Haynie. “Bil Herd was the chief engineer on that project and he was looking for somebody who could work under him who knew things he didn’t know.” Haynie met Herd while waiting in the lobby. “I met him for the interview before I knew he was one of the guys I was going to be talking to,” says Haynie. “I liked him right away.” Herd was relaxing in the lobby when an unusual young engineer arrived early. “I’m sitting in a chair reading and this kid comes in and plops down in a homemade shirt,” recalls Herd. Most candidates dressed up for their interview, but the young engineer seemed unconcerned about his appearance. “It was black and he had cut it out and sewn it himself, so it was kind of like a sweatshirt and it wasn’t real symmetrical. Picture a sweatshirt without a real good neck hem or without arms on it.” Herd looked too young to be a recruiter. “[Haynie is] excited and wild-eyed. He said, ‘So, are you here for the interview?’” recalls Herd. “I said, ‘Yeah, something like that. Where you working now?’ He said, ‘GE aerospace missile division.’ I asked, ‘Well, why do you want to leave?’ He said, ‘Oh, I don’t want to make missiles that hurt people and stuff.’” Herd hid the fact that he was from Commodore. “I was asking him all these questions and we’re having this enthusiastic heart to heart talk in the lobby,” he recalls. “He really didn’t know who I was at this time. Then they call him in to talk to Joe Krasucki.” Herd took his seat in his office and waited for the candidates to arrive. “Then I met Bil again,” laughs Haynie. According to Herd, “When they brought him in, they said, ‘You’ll be interviewing next with Bil Herd.’ I looked up at him and I said, ‘We’ve met.’ The look on his face was great.” Herd went straight into the technical interview. “He asked me a few analog questions and things I’ve never used yet, like Laplace transforms, but he just wanted to make sure that somebody on the team knew this stuff if it ever showed up.”

According to Herd, there was no doubt Haynie was the right match for Commodore. “What he didn’t know was he was already hired at that point,” he says. “He had gotten his whole interview sitting in the lobby.” Herd was impressed with Haynie’s skills and knew they would help augment his own. “Haynie was definitely the kid of the group,” he says. “The closest I found [to a systems engineer] was Dave Haynie, who had two math degrees from Carnegie-Mellon.” * * * Dave Haynie, fresh out of University, desperately wanted to use Unix. “We had one Unix machine, which was pretty much there because a lot of the programmers and other people demanded a Unix machine,” he says. “So we were running DEC Unix on a small VAX 11/750. That had all the email and everything.” Prior to the arrival of the VAX, engineers were stuck using PET machines to compile code for the system ROMs. “I had had it with that stupid PET they had given me to work on,” says Herd. “I wasn’t in the mood for the problems it was throwing. It wanted to see two different floppies in the same drive at the same time if I recall.” It seemed like the end of the line for Herd’s PET. “I grabbed it off my desk and stepped into the hall with it over my head shouting profanities. I said, ‘Take it or I am throwing it!’ I didn’t really want to throw it, but I would have.” One of the low-level technicians overheard the strange offer, stuck his head into the hall, quickly realized the opportunity and scurried over before Herd could disintegrate the machine. “Someone who hadn’t rated a PET of their own rather quickly liberated me of the load,” says Herd. Haynie convinced Herd to attain VAX access. “I had used Unix systems in college and at Bell Labs when I did two summers there,” says Haynie. “So I knew what it was and said, ‘Yes, we must have this.’” Herd set his sights on the VAX system. “I went and asked for a hookup to the VAX,” he recalls. “They said, ‘Oh gee, no terminals.’ It was a lie, since they bought eight terminals. I said, ‘No problem, I have several at home.’”

When a barrier existed, Herd either smashed through it or went around. “It was all attitude,” he says. “Everybody that worked there, it was just, ‘Get the fuck out of my way, we’re coming through.’” “They were like, ‘Well, I guess you can get on it, but you have to hook it up yourself.’ They were acting like I would never know how to do that. An hour later, I’m in there and they were asking, ‘What are you doing?’ I said, ‘That’s the last of the wiring.’ We had already run the wires and everything. I brought a terminal in from home and we were up!” “I was the eighth person on the first VAX and the only person not in the chip design group,” says Herd. “Dave Haynie and I put it to good use, though he had to sit in my office originally to use the terminal.” Herd’s aggressive attitude did not make him popular with Commodore’s technical service team. “We had people who were in charge of our VAX system who weren’t there to serve us,” he says. “They were just there to keep their jobs.” Herd often clashed with the VAX . “He called his cubicle the Operations Center when he answered the phone,” he laughs. “The manager in charge of s working on the VAX was into telling me what I couldn’t do and I replied I could be on the highest level of the system in a few minutes if I needed to. He made that bet.” It was unwise to bet against the cocky young engineer. Herd used a simple method to spy on the manager’s keyboard commands. “I took out my screwdriver and moved the wire from my terminal to his, back at the wiring block, and simply watched him type his to logon,” recalls Herd. “It actually took less than five minutes as he got [curious] and wanted to see what I might be trying on the system level.” * * * Back when Kit Spencer was involved with Commodore, he predicted 70% of s would purchase a disk drive for the Commodore 64. It was a bold prediction given the low number of VIC-20 owners who purchased one, and it could potentially leave Commodore with a costly surplus of product. Consequently, 90% of North American Commodore 64 owners purchased a drive. “I suspect they sold a printer for every five or six computers and they

probably sold a floppy disk [drive] for every one,” says Chuck Peddle. “Lowcost peripherals were very important for the success of Commodore during that time.” Commodore experienced supply problems very early with the 1541. In August and September 1983, 1541 drives all but disappeared from store shelves because Commodore was unable to meet demand. According to Robert Russell, “At the peak of Commodore 64 popularity, we weren’t able to ship disk drives.” Russell sheds light on why Commodore had problems delivering. “They weren’t working,” he says. “I had written a real simple production test early on to validate that the drives were good and aligned. It basically beat them around and validated them.” Either Russell’s validation test was too strict or the drives were unreliable. “At that time, the drive company was Japanese and the drives were assembled in Japan,” he says. “One time they quit running my software test to validate 1541 disk drives because it failed too many drives, according to the Japanese.” Soon, s began reporting high failure rates with the disk drives. The drive heads were becoming misaligned. Tom Halfhill, editor of Compute!’s Gazette, complained that four of the seven drives owned by the magazine failed in this way. Russell re the masses of disk drives that began piling up behind the West Chester headquarters. “We had an employee parking lot in the back of the company, and half of it was semi-trailers hauling stuff in from overseas,” he recalls. “We had 30 semi-trailers full of 1541 disk drives in the back parking lot sinking into the asphalt. I don’t know how many hundred-thousand drives were out there.” It was not all bad, however. Third-party companies soon eliminated the C64’s slow loading times. “Later on they came out with software revisions that allowed it to run much faster,” says Yannes. Several ingenious companies released fast loader cartridges. A game company named Epyx released the most popular cartridge, which increased speeds by five times. * * * In late 1983, Commodore began preparing for the release of the high-end P & B

computers. Commodore had already built and released preproduction models of the P128 to Commodore dealers in the US (excluding retail stores) for demonstration. However, it became redundant when the C64 established itself as a personal computer.

Commodore’s West Chester facility (2005).

Jack Tramiel decided to cancel the computer due to low demand. Dealers ended up selling their preproduction models to customers, landing Commodore in trouble with the FCC because the computers had not received final approval yet. In response, Commodore recalled most of the units and destroyed them. Later, Commodore released the business oriented 80-column green-screen B128 and B256 computers against the IBM PC. “In hindsight, no computer in that day was really powerful enough to run a business,” says Neil Harris. “It just wasn’t the right era. It wasn’t until years later that computers with hard drives came around that you could really effectively operate a business on a PC.” Predictably, they fared poorly. Commodore liquidated its inventory to a Chicagobased company called Protecto Enterprises, which sold the computers through mail order. The ads appeared in computer magazines for years, with Protecto unable to sell their remaining computers.[1] In Europe, Commodore released the business computers under the name CBMII. At the time, the IBM PC started to dominate the business world and the CBM-II line shared a similar fate as it had in the US. The C64’s success gave customers no reason to spend more money for a product that had very little software development. It made more sense for Commodore to concentrate on selling the C64. Following Tramiel’s price cut in January, the C64 enjoyed massive popularity. “The worst thing that happened was when guys were stealing stuff from the production floor,” says Russell. “They would basically re-consign whole truckloads out of there.” Even worse, it soon became a challenge for production to keep up with demand because the electronics industry had not yet adjusted to the sudden demand for computers. “The industry was building a few thousand computers and all of a sudden you’re talking a million,” says Robert Russell. “People never heard of the volumes we wanted. We ended up trying to buy parts from Russia because you just couldn’t get them anywhere else.”

Tramiel attempted to exploit the Eastern Europe market for parts. “Jack flew around in the jet and went to Eastern Europe, and they flew an escort fighter alongside him when they were over Hungary or one of the eastern European countries,” recalls Russell. “It was still the Iron Curtain.” Russell often felt nervous crossing borders into Eastern Europe during the Cold War. “We got held up in customs because we had all these strange boxes with computers that the customs guys didn’t have a clue about,” he recalls. “They wanted to know what the hell all the early computer prototypes were. I couldn’t speak their language and they had machine guns, so I didn’t want to pick a fight.” During one border crossing into East , Russell, engineer Frank Hughes, and a coworker almost met with disaster. “It was extreme paranoia,” says Russell. “We were a weird group because Frank [Hughes] was a big old Southern guy, I’m a Midwesterner, and we had an oriental girl Madeline Shaw who was in the back of the car. They were kind of wondering what the hell we were in [East ] for.” Unfamiliar with border crossing procedures, Hughes inadvertently alarmed the border guards. “The guards were motioning to him to come over and show them his papers and Frank thought he was supposed to go in the booth,” recalls Russell. “He’s there trying to open the door and the guard freaked out. All of a sudden all these guns swing around at us. The guy almost machine gunned the bunch of us.” From the car, Russell recalls screaming, “Frank, don’t open the door!” Although the tense situation resolved itself when Hughes took his hand off the door handle, Russell was not eager to return to Eastern Europe. [1] Actor Michael Douglas’ production company purchased several of the discounted computers as props in the film Jewel of the Nile (1985).

CHAPTER 34

Marketing Interference 1983

In 1983, Bally-Midway offered Commodore three of their newest games: Tapper, Domino Man, and Timber. “We had a right of first refusal for a number of years and some of them we took and some of them we didn’t based on whether or not we thought the game was fun,” explains Andy Finkel. “I we got some machines in. We could have handled Tapper, but we didn’t think it was much fun.” Commodore ed on the others too. “We also got Domino Man at the same time. We decided not to do it because it was so frustrating to play. You would be setting up the dominos and the little old lady with her shopping cart would knock over what you were setting up.” Of Bally-Midway’s latest offerings, the games group decided to port Satan’s Hollow. With no other licensed games to develop, it was up to the programmers to develop their own original titles. Commodore had their first legitimate hit with International Soccer, developed by Andrew Spencer of Commodore UK in 1983. The game was a landmark game for sports titles. It featured details like players running onto the field at the beginning of the game, a trophy presentation to the winning team, and an animated crowd. “I thought it was amazing,” says Finkel. “The players were well articulated. They seemed to move kind of like real people. Given the pixels he had to work with, I thought he did a wonderful job.” Fans around the world, and especially Europe, embraced the title. However, naming the game proved controversial within Commodore. While Americans

call the sport soccer, Europeans know the game as football. “There was a fight to change its name,” says Finkel. “Obviously when they came out with it, it was called football, but we wanted to call it soccer.” Finally, an American working for Commodore UK came up with the solution. “Gail Wellington came up with the compromise name, International Soccer.” According to Robert Russell, Commodore was concerned with software piracy even before the VIC-20 days. “We were trying to figure out how to do things protection wise,” he says. “The problem was, we’d come up with a scheme and then we’d sit down and figure out in an hour how to bust that scheme.” In an ironic twist, Commodore turned out to be its own worst enemy when it came to piracy. With the release of the 1541 disk drive in late 1983, software became much easier to distribute, for both publishers and crackers. “Here’s a little known fact: Commodore was probably the biggest source of cracked software,” reveals Bil Herd. “The worst thing you could do was submit a copy of something to the games and applications group.” Even game cartridges were not safe from illegal copying. “Someone at Commodore would hack the game cartridges and put them on a floppy disk,” says Bob Yannes. According to Herd, a cracker who used the name The Giant Clam actually worked in the software group. “The rest were good people, don’t get me wrong, but there were a few nefarious types that would generally make sure a cracked version of the game was available within a week,” he says. “The Giant Clam worked at Commodore.” Russell was also aware of The Giant Clam. “That would have been one of the guys over there in [Andy] Finkel’s team,” he says. “There were a lot of weirdos in that group.” The Giant Clam remained undetected for a time until Commodore received a preview copy of International Soccer. “It was the soccer game produced by Commodore England,” says Herd. When pirated copies of the game began appearing in North America, it was clear the pirate was an insider. “Later the FBI got involved because the only people that had even received the game were from Commodore West Chester.” It is unknown whether the FBI apprehended The Giant Clam.

Although most game conversions were of high quality, Commodore still had problems maintaining consistency. The first version of Wizard of Wor, released in 1982 for the Max Machine and C64, did not match the quality of the arcade game. Commodore was determined to get it right, however, and they released an improved version in 1983, which became an ired classic. Like the arcade game, it even included speech when used with Commodore’s Magic Voice speech module. When Andy Finkel finished with his Bally-Midway conversions, he began working on an original game of his own design. “I did a game called Dragon’s Den,” he recalls. “I started with what I thought would be an interesting game—a knight flying on a Pegasus. Then I decided on having multiple levels. Basically I tried to put in some of the things that I knew the C64 could do pretty well.” The game, known for its unique game play and outstanding graphics for the time, was a variation on Joust. “Jeff Bruette was our graphics artist programmer,” says Finkel. “He did a really nice job on a lot of the graphics in that.” The game even included half-time animations similar to those found in Pac-Man. Commodore was rapidly developing into a legitimate game company. Finkel started another original game called Guardians of Time. “The game’s gimmick was you could press your time-jump button and reset the whole battle back to an earlier time,” he explains. “I had the basic play mapped in there with a spaceship having a space battle around a fortress.” During one of Commodore’s game review sessions, Finkel’s group decided to put the project on hold. “During each game’s development process, we had a couple of points where we would get together and talk about the game and whether or not we should continue with it,” he says. Unfortunately, he never completed the game. “It just turned out not to be a very fun game, flying around trying to defend your fortress against these marauding ships. The head of that group has a lot of influence on the decision. I wasn’t really fighting for that one because it just wasn’t that much fun.” Finkel thought he might return to the game eventually, but for now he was about to the TED project. “While I might have liked to spend another three months on it, thinking about trying to make it fun, maybe add more elements, it had to be shelved. I had to move onto other projects.”

Due to the success of the Commodore 64, 1983 was a surprisingly successful year for the Commodore software division. According to the Wall Street Journal, Commodore was the third largest software producer for 1983.[1] “That was when our software group was peaking,” says Russell. “Sig Hartman was responsible.” * * * After the interruptions and distractions of moving to a new location, the TED project fell behind schedule. With the January 1984 CES approaching, the engineers had all the motivation they needed to complete the project. “In late summer we kicked it into gear,” recalls Herd. “I don’t think we knew how to make good use of our time when we weren’t in imminent peril of CES.” Unfortunately, the project was about to get a lot more complicated. The engineers housed the near-complete TED prototype in a VIC-20 case painted black. Dave Haynie feels Herd and his team accomplished the goal Jack Tramiel set for the TED computer. “There were a lot of good ideas in everything that was done there, both the hardware and the software,” he says. Unfortunately, the original reasons for the computer were no longer valid. “By the time that computer was ready for the marketplace, at least on the hardware side, the yield problem on the original C64 graphics chip had been resolved,” says Neil Harris. With its radically lowered price since its introduction, the C64 became an Apple killer, an Atari killer, a Texas Instruments killer and a Sinclairkiller all rolled into one. The move to West Chester brought the engineering group into closer with marketing and software people, such as Sig Hartmann and Michael Tomczyk. According to Harris, “The company has this product that they’re building, and they say, ‘What are we going to do with it?’ The software division says, ‘We have the answer. We will come to your rescue. We will take this application software and build it into the hardware so that when you buy the computer it will come with what turned out to be the software called Magic Desk. It will have a simple word processor, database and spreadsheet built into it. That way we can sell it for a price over the Commodore 64.’” As international product marketing manager for the software group, Michael Tomczyk now held a role similar to his early role with the VIC-20. “I was

helping to start the software division at Commodore at that time, working for Sig Hartmann,” he says. “I helped select the applications that were going to be involved and basically I was helping to drive that product.” Hartmann and Tomczyk had seen built-in applications in the Coleco Adam at the recent CES show. They found the idea intriguing. Normally, when they released a software product, only a fraction of computer owners purchased it. By including the applications in ROM memory, they could effectively force 100% of computer owners to purchase the application. The built-in software would allow them to increase the price of the computer, with a portion of the revenues going to the software division. Hartmann and Tomczyk proposed adding permanent applications, transforming the Sinclair killer into a Coleco killer. They renamed the computer the 264, and later Plus/4. “The Plus/4 got horribly mutilated by the people in marketing of all places,” recalls Herd. Leonard Tramiel felt marketing should have stayed away from computer design. “They don’t know what can be done,” he says. “If they knew, they would be product developers. As soon as you have product design being done by people who don’t know how to design products, you can’t progress.” Tomczyk explains his vision. “The next generation was going to be a very cool, sleek, small computer almost the size of a netbook computer. All the software was built into it.” Tomczyk wanted a built-in word processor, spreadsheet, database, and graphing utility, much like the popular Lotus 1-2-3 suite. “That was a forerunner of some of the software suites that we have today,” he explains. “Many software suites include those four applications, most significantly Lotus. I thought that was a genius idea. It even had speech capabilities.” To carry out the idea, Hartmann pulled in Andy Finkel from his Guardians of Time project and assigned him to the TED computer. “I was software project manager of the Plus/4, so I was the guy who was responsible for getting all the apps up and running on the Plus/4 with all the third party developers,” explains Finkel. Finkel embraced the concept of applications in ROM. “I thought it was a valid concept,” he says. “The idea is that you turn on your computer and it’s there with

your preferred application at all times. I thought that was pretty powerful.” Finkel likens the idea to computers with applications preinstalled on a hard drive. “You turn on your computer and there’s Internet Explorer or Firefox these days. People keep coming back to that idea with the thin client computers and the little boxes that just do email and so on.” In the past, Tramiel saw each new computer as replacing the previous model. Now he wanted to try multiple products at different price levels. “Commodore really didn’t have the idea of keeping multiple computers in the consumer marketplace until the Plus/4,” says Finkel. “Jack’s thing was always if you have to compete with someone, you should compete with yourself.” The idea of built-in software was controversial to the hardware engineers, while the software engineers ed the idea. Practically, s would be unable to update the built in ROM software, which would quickly become outdated as other software programs continued evolving. Surprisingly, Tramiel backed his marketing people. Herd was at odds with his employer, but he had little recourse. “I respected his decision-making capacity, if not his decisions every time,” he recalls. With from Tramiel, marketing took control of the system. “Sometime during the summer of 1983, it started growing features,” explains Haynie. “By the time I came along, all these features were set and it was just a matter of getting them to work.” Haynie thought it was a mistake to allow others to step in and change the course of the TED computer. “You see that from time to time in management,” he says. “Whether it was due to ego or something else, these things happen.” The relatively new engineers felt they were obliged to listen to marketing managers, rather than chart their own course with the computer. “As it was happening, I was still kind of new at this,” says Haynie. “It wasn’t necessarily Commodore engineering’s job to decide what kind of model should be made.” Russell felt it was pointless to attempt to improve on the minimalist system by adding features. “None of the engineers liked it because we all knew that fundamentally it was meant to be $50 computer that we would sell for a hundred bucks,” he says. “For them to add software and call it a Plus/4 was wrong.”

Tramiel understood the rules of business like no one else in the computer industry. His philosophy was simple: low prices sell more merchandise. Nevertheless, Tramiel still thought of computers as calculators. “He felt that the marketplace would go for lower cost, mixed-function computers,” says Al Charpentier. “Jack ‘grew up’ in the calculator market and he looked at the computer as just another big calculator. That’s why the follow-on versions were somewhat crippled.” Unfortunately, computer software and calculator functions were not the same thing. “That’s what he didn’t understand.” * * * Within the Commodore software group, there were three divisions: business applications, home software, and education. As a result, Hartmann decided to create three versions of the TED computer, which Andy Finkel refers to as flavors. They would create a business flavor with sophisticated applications, a home flavor with an ultra-friendly joystick controlled interface, and an educational flavor. In order to expedite the business flavor of the TED computer, Commodore located a company in Orange County, California named Tri Micro that had an existing software suite called Trilogy. “We got an office suite from International Tri Micro, which had a word processor, a spreadsheet, and a database,” says Finkel. The vice president of software engineering at Tri Micro, David Johnson, developed the product. He would work with Finkel in porting the suite to the business machine, which Commodore later dubbed the Plus/4. It was up to Finkel to use his talents shrinking code in order to make the suite fit into the system ROMs. Perhaps the greatest potential for built-in ROM software could have been the inclusion of a graphical operating system. The closest thing Commodore had at the time was Magic Desk, which they previewed at the June CES for the C64. John Feagans was still ionate about graphical interfaces, and he ed the TED project to port Magic Desk. Press releases claimed the “menu system uses icons, or picture-symbols, rather than words to convey the different functions (similar to Apple’s more complex business-oriented Lisa).”[2]

Creating Commodore’s first Graphical Interface.

Andy Finkel flew to California to work on Magic Desk at Scott Boulevard. After a favorable reception at the recent CES, Feagans expanded his team to complete the project. “I managed a team of four engineers doing Magic Desk,” he says. Finkel enjoyed working with Feagans. “John and I worked on a couple of things together. He was great, always a lot of fun. I was always a little envious of his creativity,” he says. “He came up with the Magic Desk thing very early. Other people have come up with it since then, including Microsoft, but he was the first one I’d heard about that metaphor from, long before I’d seen anything like that on the Mac.” Feagans was one of Peddle’s original “weirdos,” as Peddle called his engineers. Finkel had much the same impression of Feagans. “He is kind of eccentric and he has his own ways of doing things,” he says. Feagans had a dry sense of humor, however. “He’s tall nerdy guy with glasses who would seem all serious and then he would come up with the most outrageous statement, but he’d say it with a straight face. He’d start smirking and you could tell he was really kidding. He had a very laidback sense of humor.” It was up to Finkel to shrink down Feagans’ code to make it fit in 32 kilobytes of ROM memory space. “The Plus/4 had some customizable ROM sockets that we could put in different application software,” explains Finkel. “The ROMs were given a chance to start up first and they could basically start into them. So the Magic Desk Plus/4 would have started up in Magic Desk mode.” Kit Spencer saw potential for Feagans to develop his ideas further, and infuse his GUI with a machine more like a Commodore 64. “John Feagans with his Magic Desk had potentially the same impact as the Macintosh for Commodore around then,” he says. “If we had taken that a step further with the Commodore 64 and what John was working on, we might have had another great machine. And Commodore could have kept an awful lot of the backward compatibility on software and had a Windows/Mac type interface. But it didn’t really happen.”

Finkel also travelled to Cambridge, Massachusetts in order to adapt a version of LOGO to run on the educational flavor of the TED machine. “We had one with LOGO built in, from MIT,” he recalls. “That one would have booted up in LOGO. If you start it up you see the turtle, ready to type in on the split screen. I thought that would have been a wonderful one.” Finkel saw the potential of a dedicated LOGO machine. “LOGO is a great teaching tool for schools and for kids,” he says. “It would be the cheapest LOGO station around.” Commodore even had a fourth and fifth flavor of the TED computer, the C16 and C116, which contained no built-in software. The C16 was the TI-99/4A Killer, while the C116 (a cost reduced version of the C16) was more like a Sinclair killer. “The 116 was the original design meant to compete with the Sinclair,” says Herd. The C116 would receive a calculator style keyboard, similar to the original PET 2001. “The reason it had the crappy keyboard was because that was what a Sinclair had,” explains Russell. As CES approached, the marketing team requested even more features. “TED had been mutilated, decimated, and defecated upon,” recalls Herd. “Management decided to kick the body one last time.” In Dallas, Commodore had acquired two of the best minds in speech technology at the time, the makers of the popular Speak & Spell educational tool. “We hired the guys who made the TI Speak & Spell, Tom Brightman and Richard Wiggins,” recalls Herd. “Rich has a PhD and Tom is just extremely bright. Back when the TI Speak & Spell came out, there was nothing like that! My God, they made a movie about it,” he says, referring to the 1982 film E.T. the ExtraTerrestrial. The two pioneers worked out of a Commodore research office in Dallas, Texas. Commodore gave them a mandate to create a speech cartridge for the C64 called Magic Voice. Commodore wanted Magic Voice integration with their products, and Feagans added voice to Magic Desk. “The Commodore voice team in Dallas was an interesting group,” says Feagans. “When I added speech to the Magic Desk, I worked with them.”

After Feagans easily integrated voice capability with Magic Desk, the marketing team decided it would be a good addition to the TED computer, which they would call the 364. “That was basically just a 264 with a numeric keypad and a Magic Voice speech synthesis built in,” says Haynie. Adding Magic Voice to the TED was not difficult. “The software was already done for it, because it had been the Magic Voice add-on, or as we always called it, Tragic Voice,” says Haynie, in reference to the monotone voice, which sounded like a depressed robot. Magic Voice was somewhat limited. “It was a fixed vocabulary of about 260 words and that was pretty much it,” says Haynie. “It didn’t have a phonetic language, so it was real hard to get it to say the stuff you wanted to.” Herd loved the technology. “It was literally the first talking computer I knew of, and I didn’t care how it talked,” he says. “It said ‘gerple’ instead of purple.” Magic Voice amused the engineers during their daily routine, but it paled next to the abilities of a third-party application by Don’t Ask Software. “I was working on that one day and somebody brought in a thing called S.A.M. which was a thing for the Commodore 64 that had speech,” recalls Haynie. S.A.M. could synthesize an unlimited number of words generated from text, rather than a fixed vocabulary. The engineers had a duel between S.A.M. and Magic Voice. “You could construct an amazing array of words,” says Russell. “We were trying to say dirty things about each other and I was extremely limited with the 364 vocabulary. I was just totally destroyed by S.A.M. when we were having a laugh one day.” The TED computers also included the BASIC programming language. Programmer Terry Ryan developed BASIC 3.5 for the TED series. Ryan included powerful new commands, making it easier for s to program graphics and sounds. It was an improvement over the POKE commands of the VIC-20 and C64. “When he wrote structured BASIC, which went into the Plus/4, he got in trouble for doing it,” recalls Herd. “He was told, ‘Do not put LOOP-WHILE and these instructions in there,’ and he got a bad review for making great code. The good news was his boss got fired weeks later.” * * *

Throughout the price wars, IBM remained implacable because they stayed out of the home market. Businesses continued buying IBM PC computers and the company quickly established itself as the office standard for microcomputers. In November 1983, IBM wanted a piece of the home market and announced a $700 home computer called the IBM PCjr to compete with the Commodore 64. Industry observers expected IBM to triumph in the home market space, as they had done with the business computer. The innovative design allowed the computer to connect to a television set and featured a cordless infrared keyboard, allowing s to compute from their living room chair. IBM launched the system with a series of clever Charlie Chaplin commercials featuring a baby carriage. Tramiel responded to the IBM announcement with disdain. The Wall Street Journal quoted him saying, “There are far more people with $200 than $700 for a computer.” Tramiel reaffirmed his strategy of conquering the market with low prices. In the end, Tramiel was right and the IBM PCjr eventually failed. * * * With CES closing in, the magnitude of the workload became clear. “We went like nuts for the fall through CES,” recalls Herd. “We hit time compression in the fall when the 264, 364, and 116 were all parallel developed to some degree.” Russell’s project engineers impressed him with their dedication. A tall engineer named Greg Berlin created disk drives for the TED computers. “There were certain people like Greg Berlin and Bil Herd who were core engineers,” he says. “They worked their butts off to try to fix the problems and get [the system] running. … They are friends for life after you go through that type of war.” Like Chuck Peddle before them, the engineers used cots in the workplace. “These are the guys who gave 199% working 20-hour days,” says Russell. “They would be in there sleeping because they wouldn’t go home. We had these real long cubicles and we had enough space to go under the desk and sleep. It was an intense environment.” Herd was disappointed some of the engineers were less than dedicated. “The ‘second shift’ guys—meaning the guys that showed up only after management had left—basically gave up any pretense of working that we could discern,” he says. “The true workers were there both during the day and night.”

A few weeks before CES, Herd began running into unforeseen obstacles. “The security guards started locking the door,” he recalls. “They got a rule that said all lockable doors should be locked on Friday nights. So Saturday I walk in, and I can’t get in my office. There is no key for this door because it’s new construction, and the contractors hadn’t dropped the keys off yet.” With the deadline near, Herd decided to go around the problem. “I had to get back to work,” he says. “You can’t keep me from working. So I climbed over the ceiling and got all gucky doing it, and cut myself, but I opened the door and got back to work.” To prevent the incident from reoccurring, Herd posted a friendly notice on the door. “I put a sign up that said, ‘Please do not lock this door. There is no key for it,’” he recalls. “Well they locked it again. We went through three layers of notes. The first one was real polite, the second one said, ‘Look, you locked it again. We can’t get in to do our jobs. Don’t lock this door, there is no key.’” Herd thought his notes explained the situation clearly, but the instructions from management overruled Herd’s pleas. “Well, it got locked again,” says an exasperated Herd. “Somebody came and got me and said, ‘I can’t get in the room. It’s locked again.’ This is the room they had given me to do the project, so it’s my room as far as I’m concerned.” Herd stormed towards the locked door. The previous times, he had gone around the problem. Now he decided to go through the problem. “I punched a hole through the wall to where you could reach in and unlock the door,” he recalls. “I just barely missed a light switch on the other side, which would have split my knuckles wide open.” Herd thought the faceless battle of wills had come to a climax, but it continued. “They locked the door again!” he says. “You had to reach through the hole to unlock the door and you would get your arm all chalky and everything. Finally, I had to write a note that said, ‘Look, assholes. There’s a fucking hole in the wall next to the door. You can stop locking it now.’” When Herd came in on Monday morning, management was not amused. “It happened over Friday, Saturday, and Sunday,” he says. “On Monday morning, Bob Shamus, the QA engineer, said, ‘They’re looking for whoever put the hole in the wall. They’re in big trouble!’ He’s walking around saying that, and

somebody said, ‘Bil Herd put that hole in the wall. There’s a sign there you should read,’ and he shut up.” * * * By the time Alphaville released their hit song Big in Japan in early 1984, Tony Tokai had already lost the Japanese market. Even the potent C64 was not having an impact. A new type of computer dominated Japan and threatened to spread to other parts of the world. The winning strategy for Japan came from Kay Nishi, one of the groupies who hung around Yash Terakura at Commodore Japan. Chuck Peddle first met Nishi when he launched the PET. “There was this 19 year old kid who wrote a magazine called [I/O],” says Peddle. “The first time I met him was at the Hanover Fair. He was just gaga. He took pictures, goes back home and sells [magazines].” In 1981, Nishi was working for IBM. The 26 year old witnessed the increasing popularity of home computers and decided to design one. “He and [Bill] Gates formed this team,” recalls Peddle. “They were both about the same age and he moved to Seattle. He’s pushing Gates to design a computer with the Japanese.” Nishi was familiar with the story of how Sony’s proprietary Betamax format lost dominance. He saw personal computers as home electronic devices, like VCRs. Likewise, he quickly identified software as the media of computers (like VHS tapes in VCRs). Nishi envisioned a computer where competing companies would build computers that used the same software. “It was called MSX and all the Japanese companies standardized on it,” says Peddle. Nishi claims MSX stood for Machines with Software Exchangeability. All computers using the MSX standard would have the same specifications: an 8-bit Zilog Z-80 microprocessor at 3.58 megahertz, a Texas Instruments video processor, 16 kilobytes of video memory, a General Instruments AY-3-8910 audio chip, Atari style joystick ports, and at least 8 kilobytes of memory (though most included 64 kilobytes). Sony and Matsushita (Panasonic) were the first companies to take notice of Nishi’s concept and designs. The two giants ed each other to discuss

plans to implement Nishi’s vision. Soon, 11 other companies ed, including Sanyo, Hitachi, Canon, Mitsubishi, Toshiba, Fujitsu, Kyocera, General, Yamaha, Pioneer, and JVC. The MSX standard included Microsoft BASIC on a 32-kilobyte ROM chip. “They gave Gates a paid deposit against per-machine royalties,” says Peddle. “Kay Nishi thought they were going to kill the market. His theory was to come in under us with a packaged machine that could be sold using Japanese style and Japanese customs.” And it worked—in Japan at least. The Japanese Commodore 64 met with initial success, until the MSX came along. The first MSX computers appeared in autumn 1983. As promised, they were the VHS of computers. Although the computers were technologically inferior to the C64, and cost more, the promise of a universal standard won out. Japanese buyers knew software makers would rush to MSX. The C64 withered against the might of the MSX standard. There was no conceivable way for Commodore Japan to compete against the combined advertising might of Sony, JVC, Panasonic, Canon, and Yamaha. Through the course of the MSX lifetime, developers released approximately 70,000 different MSX software titles worldwide. Commodore Japan gave up and sold off their remaining C64 inventory at below cost to cut their losses. The MSX computer also gained popularity in many parts of Europe. Philips, based in the Netherlands, was working with electronics giant Thomson on a European standard from 1982 to 1983. The project did not make sufficient progress and Philips instead decided to the MSX standard. Philips became the largest supplier of MSX machines in Europe, but made little headway in the UK. “The thing that screwed [Nishi] up was Clive Sinclair did his junk machine which moved the price point down,” says Peddle. However, the MSX computers succeeded in many smaller markets, including , Spain, the Netherlands, Korea, and even the former Soviet Union. [1]* Wall Street Journal (February 23, 1984). [2]* Compute!’s Gazette magazine, “Commodore’s New Computer Family” (April 1984), p. 20.

CHAPTER 35

End of an Era 1983-1984

By 1983, there were too many computers in the market and it was time for a shakeout. As Byte magazine noted, “System suppliers have tried to become much more aggressive in their marketing, but only the better-financed and established companies are succeeding. Add to this the current recession, and it’s no surprise that industry pundits are predicting a shakeout among personal computer manufacturers in the very near future as marketing becomes more important than the products themselves.”[1] Byte later dubbed the event “the Great Shakeout.” Commodore entered into the war with the single greatest weapon they had ever devised, the C64. For the next few years, it would be the central piece of artillery on the battlefield, with the VIC-20 in a ing position. Between 1982 and 1984, the C64 left a smoldering path of destruction in its wake. Jack Tramiel knew that to eliminate the competition, he would have to take Commodore places others could not survive. With his lean company and vertical integration, Commodore could decrease the price of their computers to a point where others would not be able to follow. On Friday, October 28, 1983, Texas Instruments publicly announced it was pulling out of the computer market and selling off its remaining inventory at bargain prices. Tramiel had defeated his former nemesis from the calculator wars. Within two days of announcing its withdrawal, Texas Instruments stock surged 30 percent. Atari had no vertical integration to compete with Commodore and even bought

its ROM components and microprocessors from Commodore. After selling twobillion dollars of product in 1982, sales dropped abruptly to $1.1 billion and Atari lost $538.6 million in 1983. Atari owner Warner Communications saw the trend and wanted out before the Atari division bled it to death. Tramiel’s “business is war” philosophy was effective. “Jack always wanted to drive the price down so that no one could compete and he succeeded,” says Bob Yannes. “The C64 was a steamroller. No one could really go up against it.” Radio Shack, once a potential reseller of the PET, also failed against the C64. The TRS-80 initially benefited by distribution in almost 7,000 Radio Shack stores, but this seemed more like a limitation as time went on. Radio Shack was unable to penetrate into the wider market. Owners also dubbed the TRS-80 the Trash-80, owing to its poor quality. Tandy eventually dropped the TRS-80 line and started making IBM PC clones. By 1982, the Apple II was starting to look dated. There had been updates, but they were not fundamental. Potential Apple customers became dismayed when they compared it with the C64. The Apple II was too expensive and game quality was not as good as the C64. Apple’s fear of Commodore went back to 1977, when they realized Commodore’s ownership of MOS Technology could give it an unbeatable advantage. With the C64, Apple’s worst fears came true. The Apple II now looked sad with its poor sound and muddy graphics. Commodore, using its own graphics chip, sound chip, and microprocessor, was able to sell an incredible computer for a fraction of Apple’s price. Its shares fell from $62 in June 1983 to $24.25 by September. Apple could have charged its customers less for the Apple II. Between 1977 and 1982, there were many opportunities to decrease the price, since the cost of components fell yearly. Instead, Apple was intent on overcharging its customers. In contrast, Tramiel routinely lowered his prices as soon as components became cheaper. The great shakeout reaffirmed Commodore’s strengths compared to the competition. Tramiel had all but destroyed Atari, Texas Instruments, Timex, Radio Shack, Osborne, Coleco, and Mattel, while weakening Apple. Only IBM remained untouched by the mighty C64. Tramiel had patiently built up a

company with unbeatable vertical integration and now Commodore was reaping the rewards. * * * The Commodore 64 entered the computer market at an opportune moment. The recent video game crash resulted in the disappearance of the Atari 2600, Intellivision, ColecoVision, and other consoles, but video games remained popular. With no heir apparent in game consoles, the Commodore 64 would fill the void for the next several years. The great shakeout proved Tramiel’s skill as a company leader. While other companies folded around him, Commodore thrived. It went from a $50 million dollar a year company during the PET era to a billion dollar company in the space of four years. In 1978, Commodore sold 4,000 PET computers over the whole year. Now they sold 5,000 Commodore 64 computers in a single day. Though sales of the Commodore 64 were on an upward trajectory through 1983, it took time to build the market. In fact, sales of the VIC-20 continued to dominate for most of 1983, largely due to faster production and an everdecreasing price. The VIC-20 received impressive but short-lived commercial software . In total, companies released 530 commercial games, 35 applications, 30 educational titles, and 15 programmer utilities.[2] Of these, Commodore produced 106 software titles. Most commercial development ceased by the end of 1983. No one enjoyed Commodore’s success more than Tramiel. Despite his uneasiness with the press, he gave an interview to the MacNeil-Lehrer Report. Tramiel offered one of his favorite sayings, “Business is like sex. You have to be involved.” He subsequently watched the report in his office with over two dozen employees. When they showed his quote, everyone burst out laughing except Tramiel, who was visibly embarrassed. Irving Gould, the top shareholder in Commodore, should have been delighted with Tramiel’s performance. His 17.9% holding in Commodore stock, which cost him $500,000 in 1966, was worth almost $200 million by 1983. However, Gould had been livid with Tramiel since the June CES show, when he

made an inspired yet impulsive decision to cut the prices of software and peripherals in half. “The ripple effect created an unanticipated situation where all of our retailers demanded stock balancing because they had now purchased their inventory at a cost that was higher than the new retail price of the slashed products,” explains Michael Tomczyk. “Retailers had paid more money for the inventory that was on their shelf at wholesale than the new retail price.” In the old Commodore, Tramiel would have found it amusing that his smaller retailers had lost out due to one of his decisions, such as when he alienated ComputerLand. The new Commodore was more responsible to mass distributors like Kmart and Sears. According to Tomczyk, the retailers demanded compensation, telling Commodore, “If you are going to cut the prices by 50%, you have to give me 50% for all the inventory that I have. You have to make it up with product. Or you have to give me a discount on all the new products that I buy.” Tramiel’s snap decision cost Commodore millions in lost revenues. “That was a serious problem,” says Tomczyk. “We had to give a lot of free product to a lot of retailers, which cut our profit margin.” The failure of Texas Instruments also had unintended consequences. “Texas Instruments started dumping their product on the market at $50 or thereabouts, which cut into our Christmas sales,” says Tomczyk. “So the combination of having to give free product to our existing retailers, combined with the dumping of Texas Instruments computers on the market just to get rid of them, really crushed our Christmas quarter. This is the biggest quarter of the year for Commodore or anybody who has consumer products.” To combat the glut of cheap computers, Tramiel came up with a creative solution that literally removed the competition from homes. “Right about the time I started working there, Jack had come up with this great marketing scheme,” says Dave Haynie. “He had an offer, strategically delivered as Tramiel always did, that said, ‘Send in any sort of computing device and you can get $100 off your Commodore 64.’” Commodore soon had a warehouse full of competitor’s computers. “We had a lot of junk around there,” says Haynie. “I looking through that, and there was a Sol-20. That was the computer I wanted when I was a kid.”

With the storage space filling up, Commodore employees began topping up their salaries by bartering with the computers. According to Bil Herd, employees could go shopping for stereo equipment and other electronics using the computers, if you were willing to sneak past manager row with the stolen loot under your arm. Commodore employees speculated that the offer actually increased sales of Timex-Sinclair computers. Customers sometimes purchased $50 Timex-Sinclair computers just to take advantage of the $100 rebate offer. The excess Sinclair computers became a running joke within the halls of Commodore. “We had all these Sinclairs,” says Herd. “I started using them as doorstops.” Tramiel’s solution helped negate the dire situation, but not entirely. “The bottom line was, we knew this was going to be a bad quarter,” says Tomczyk. Gould obsessed about year over year growth of Commodore in order to keep his stock price growing. He took seriously anything that slowed down the growth of his stock. * * * Jack Tramiel had come a long way since his days in Poland, where he dreamed of running his own factories. “There was this presence and everything, but right on his arm was that tattoo,” recalls Bill Herd, referring to his concentration camp identification number. “You would look at that, and you would know what that was and just know a little bit more about that man when you saw him.” Now, his worldwide operation turned out thousands of computers each day. His efficient, frugal way of life was present at every level of Commodore, from the boardrooms to the factory floor. Tramiel enjoyed the simple pleasure of touring his manufacturing facilities and talking with his workers. He felt like an equal among his employees. He told an interviewer, “If you ask the people who worked with me, they will say that I have not changed in the last 25 years. I was always one of them.”[3] He had lived in small apartments and driven a taxicab to make his dream come true. Now, analysts mentioned his company in same breath as IBM, the company he sought to emulate since the 1960s. Nothing gave him more satisfaction than watching computers roll off the assembly line, except perhaps watching his own boys grow into men.

Tramiel’s sons were at the center of his world. “My family became very, very important to me,” he said. “I believe that in life you have to help one another, you have to trust one another, and you have to develop continuity in all things, which I strive for.”[4] Tramiel harbored a dream of handing Commodore over to his three children, fighting the corporate war together as a family business. Over the years, he had subtly guided his sons on a path that would allow them to run a business together. He knew there were three important roles in any successful technology company: a business leader, a technology guru, and a financier. In the past, it had been Chuck Peddle, Irving Gould and himself. More than anything, Tramiel hoped to continue his legacy at Commodore using his sons in all three key roles. At first glance, Tramiel’s sons did not resemble their father. He was cleanshaven, but all three boys resembled grizzly bears with their full, bushy beards and moustaches. Tramiel was rotund, while his sons were taller and thinner. Individually, his sons could not match Tramiel’s ambition and determination, but perhaps together they could run Commodore. Sam was the most similar to his father physically. He had the same bulging eyes and intense stare. He also shared the same tough management style, though he was less abrasive than Tramiel. Trained at York University in Canada, Sam was destined to handle business affairs at Commodore. Although exiled to the Far East after his wife ed away, Tramiel welcomed Sam back after the release of the Commodore 64. According to Chuck Peddle, “He meets this girl that he’s totally in love with … but the parents don’t necessarily agree. He finally just marries her and she does the right thing: she makes a couple of babies. At that point, Jack and Helen circle back in.” Leonard, the middle child, would be leaving Columbia University in 1984 after receiving his PhD in astrophysics. According to Peddle, Tramiel was upset his son did not go into something practical, such as electrical engineering. Despite his chosen vocation, Leonard received work experience in Commodore’s engineering labs over the years. He helped with the original PET and occasionally helped troubleshoot other computers. According to Jack Tramiel, “I always strove to show them what I do, and to involve them with what I do and discuss our successes and failures.”[5]

Gary, the youngest in the family, was the moneyman. He was also the most spirited. “He was kind of the wild nutty one,” says Bill Seiler. “They were always having to get him out of some trouble. He smashed up the car. He was always doing the dumb teenaged things. Leonard was more sensible.” His wild ways reminded Jack Tramiel of his own early life. “Gary was sort of his favorite,” says Seiler. “He seemed to be able to get away with murder with Jack, but Sam and Leonard would get beat-up more.” Gary received financial training at Manlow Park College. After college, he briefly worked for Merrill Lynch as a stockbroker. He also managed the Tramiel’s multimillion-dollar investment portfolio. There is little doubt Tramiel wanted his sons in key positions at Commodore, for simple reasons. “I don’t think Jack really trusted anybody except his family,” says Yannes. Tramiel let his feelings about his sons known in an interview. “My dream was that my sons continue to try to be the best, exactly as I have tried to be the best in my area, however not to force them to work in the same industry.” He also doubtfully added, “By the way, all three specialized in different aspects of the company without any planning.”[6] With his sons’ educations ending, Tramiel planned to assimilate them into Commodore. Some employees doubted his sons and wondered if they could produce the same magic as their father. Tramiel was truly a unique individual and it was unfair to expect his sons to match up to his larger than life accomplishments. As Commodore engineer Dave Haynie says, “They were no Jack juniors.” Unfortunately, Irving Gould also had little faith in Tramiel’s sons. As Haynie recalls, “Gould, being the principle stockholder and usually chairman of the board, wasn’t happy with Tramiel’s plans to bring his boys up as his successors.” A disastrous clash of wills was now on an unalterable path. * * * Commodore always beat their competition with lower prices, but its most successful products happened when the technology outpaced that of the competition. The PET succeeded with ease-of-use by offering a complete

product. Even the VIC-20 exceeded the competition to some extent with a fullsized keyboard and superior games. The C64 dominated the competition with superior memory, sound and graphics. However, since the C64, Commodore’s attempted products did little to advance the state of the art. Most of the products were derivative of a competing product, resulting in costly failures. Without visionary engineers, Tramiel instigated products that copied the competition rather than pushing Commodore’s products ahead of the competition. Engineers lasted longer than managers did, but rarely stayed for more than one or two product cycles. In the short history of Commodore computers, there had been frequent mass exoduses of the best engineers, including Bill Mensch and the 6502 engineers, Chuck Peddle, Bill Seiler, Al Charpentier, and Bob Yannes. Commodore undoubtedly could have achieved greater success if they retained their key engineers. Part of the exodus had to do with the constant pressure Tramiel exerted on employees. “He would come up and chit-chat from time to time and offer friendly advice,” says Neil Harris. “I mean, obviously he was a pretty intimidating guy. I don’t intimidate that easily, but you know, I had seen Jack pound on the desk and yell at people enough to not look for as much interaction as I might have otherwise.” Tramiel’s business philosophy included trapping and occasionally suing other companies rather than establishing long-term working relationships. As a result, very few companies wanted to do repeat business with Commodore if they could avoid it. Tramiel’s devotion to cutting costs often came dangerously close to alienating important partners. As Steve Jobs recalled, “I couldn’t find one person who had made a deal with them and was happy. Everyone felt they had been cheated.”[7] The only people happy with Commodore were their end customers. While others pondered his leadership, Tramiel continued his obsessive drive towards total vertical integration. Due to the incredible number of computers produced and sold during 1983, a worldwide RAM shortage occurred, causing prices to climb. Tramiel once again tried producing RAM internally. He bought the

manufacturing rights for a 64-kilobyte RAM chip from an Idaho company. In late 1983, he set up a manufacturing line in a Costa Mesa, California production facility. Commodore was now on its way to being the most lethal computer company in the industry. Chuck Peddle believes Tramiel developed his tough business strategy when Commodore went through turbulent periods. “There was a definite period of time when Commodore was cash poor and Jack was using that as an excuse to do all kinds of jerking around—‘Sue me’ and all that—to stay alive,” he says. Curiously, Tramiel’s ruthless tactics continued, even in profitable years. “Clearly there was a period of time when Commodore had lots of money, and they still didn’t take care of those guys,” says Peddle. “During the C64 and VIC-20 times, they had tons of money.” Yannes felt Tramiel’s short-term tactics could not go on forever. “The whole philosophy, the whole corporate culture at Commodore was something that couldn’t be sustained,” he says. “The company had to be run differently.” Tramiel was tough, but he was also charismatic. Employees felt a keen sense of purpose when he was around. Their lives and actions seemed more real, as though what they did really mattered. “We all loved fighting for Jack Tramiel because we knew he was very hard driving,” says Tomczyk. “You could get killed fighting for him, but you would willingly die in that sort of combat if your goal is to bring computing to the world.” There was no arguing with success, and Tramiel delivered it to his employees. With Commodore becoming a billion dollar company, Irving Gould was faced with a dilemma. Did Tramiel’s good points outweigh his bad? Would Commodore do better without the founder at the helm? Yannes recalls the general feeling about Tramiel. “People were wondering, ‘What do we do about Commodore? Here’s this company that’s really successful and they’re really on the cutting edge, and they are really growing, but what do we do about this guy Jack?’” * * * The clash between Gould and Tramiel culminated when problems began to surface the previous year, with the hiring of Bob Lane. “If you go back to the

time when Bob Lane came into the company from Northern Telecom, Jack and Irving were already beginning to have a little bit of a problem,” says Kit Spencer. Soon after the departure of Lane, Tramiel wanted Sam Tramiel in the position of Commodore president. “Sam was already in the business,” says Andy Finkel. “For a while, he was running Hong Kong for Jack. But he wanted to bring Leonard and Gary in too and apparently that didn’t go over well.” Tramiel began placing his sons in key positions at Commodore. “Jack looked at this as his family business and wanted his sons to take over,” explains Dave Haynie. “From everything we ever heard, Gould would have nothing to do with that.” He gave his younger son Gary a job in finance. “I think he wanted Gary to be head of sales and Leonard to be head of engineering,” says Andy Finkel. “Between June [1983] and January [1984] is when his two sons were appointed positions,” recalls Russell. “I was told we have to find a place to accommodate Leonard, and we were trying to figure out how to do that.” Most Commodore engineers were welcoming towards Leonard. “I had been talking to Leonard off and on for a couple of years before that, since he was Jack’s technical expert in the family,” says Finkel. “Whenever Jack wanted to know something about computers he’d always ask Leonard first. He was Jack’s unofficial technical expert, one of the people Jack trusted and asked his opinion on.” Leonard was preparing to graduate from Columbia University. According to Leonard, he held no firm plans for his career at the time. “I wasn’t sure, actually,” he recalls. “I was a PhD student in Physics at Columbia and I would have been perfectly happy going to work for Bell Labs or doing some research thing, or going to work for Commodore.” “Leonard was the only one who wasn’t really locked into anything,” says Russell, who was more than happy to find a place for him within Commodore. “We got along because we were both software types. We talked about stuff and I sent him information when I was asked or when I was feeling like it.” By late 1983, Russell was ready to welcome Leonard to the Commodore

engineering group. “He was coming over into my side of things, which was the engineering software,” he says. On December 13, 1983, Tramiel celebrated his 55th birthday. The family gathered at the Tramiel ranch in California and presented him with a new sports car. Tramiel put away the tensions of work and spent a day of happiness and contentment with his family. On January 3, 1984, Leonard Tramiel’s succession to head of engineering began in earnest. “I had gotten a call actually about a week before CES,” says Finkel. “Leonard was kind of telling me that he was going to be taking over engineering. We talked about what he wanted to do there and how he wanted things organized and so on.” Finkel ed the move. “I thought it could work. I knew Leonard and his wife from shows—CES and so on—and they seemed like reasonable people. Leonard was certainly technical. I guess the only question was how he’d take to being the guy in charge. I was willing to wait and see.” Leonard claims he would not immediately have received a management position like his older brother Sam. “If I had gone to work for Commodore, I would not have initially taken on a leadership role in engineering because there was already an engineering group there,” he says. “I would have to work my way up. I would have maybe worked as an engineering manager of some sort, but more likely as a programmer.” For the past six months, Commodore had been without a permanent president. Tramiel wanted Sam in the role. Gould wanted Marshall Smith, the former president of Thyssen-Bornemisza, a multinational conglomerate in the steel industry. The conglomerate operated from the Caribbean islands of the Netherlands Antilles for tax purposes, much like Commodore International in the Bahamas. Neither Tramiel nor Gould would budge. “Jack spoke a lot with Irving and they were quite a good partnership together for very many years,” says Spencer. “I think the problem came when Commodore was so successful that there wasn’t a common enemy outside. We were no longer in survival mode. We were making money. The share price was doing well. Commodore was extremely successful. So it was maybe who runs the company, how should it go, what direction. I think it was perhaps egos coming into it.

Nobody knows the full story. It’s a whole combination of things.” In December 1983, Commodore celebrated its 25th anniversary as a company. For the occasion, Tramiel, Gould, and top executives returned to the birthplace of Commodore in Toronto Canada. Commodore celebrated with style, hosting The World of Commodore, with over 38,000 attendees. Tramiel had left Canada in disgrace years earlier after a financial scandal and now he returned a hero. Government representatives welcomed him. It was an amazing experience for the company founder, who rarely looked back. The display included calculators, typewriters, and even the original wooden PET. It was Tramiel’s opportunity to appreciate how far he had taken Commodore. It should have been a momentous time, but it was not. January would be Tramiel’s final month with Commodore. * * * January 10th was the start of CES in Las Vegas and Commodore’s annual debut of new products. It was the biggest show up to that time, with over 90,000 attendees. Commodore presented their most ambitious booth yet. In the center of the booth, writer Jim Butterfield demonstrated the 264. “The CES show centered on the 264 version with Magic Desk,” says Bill Herd. In the days before CES, the marketing department decided to raise the price of the 264 computer. “By that time the marketing department—and I won’t name names—decided that this product was better than the C64 because you could use all of the memory and it had a real RS-232,” explains Andy Finkel, referring to the C64 port.[8] “They decided it was going to be more expensive than the C64. So here we are with this computer that was designed to be less than the C64, now stuck with a price tag of more.” Finkel’s close friend, Neil Harris, had little faith in the product from the start. “It was being marketed as this next great leap forward and it was really kind of a step backward. What you had was a computer with worse graphics than the Commodore 64, and with frankly a fairly rudimentary piece of software. No knock on John [Feagans], but Magic Desk wasn’t written to be the world’s greatest anything, it was written to be a bunch of basic applications.” When Harris learned of the price hike at CES, he took his concerns directly to the interim president, Don Richard. “I talked to the president of the company and

said, ‘How is it that we’re charging a price for this?’ And he said, ‘Well we’re number one in the market so we can get a price.’ I said, ‘We got to be number one by being the low-cost high-performance guys and here we are coming out with a high-cost low-performance computer. You know, this makes no sense. It really doesn’t fit our marketing positioning.’ And of course they didn’t listen to me.” Finkel, the software product manager on the 264, recalls the response from CES attendees. “People liked some parts of it, but the big question was, ‘Well, is it compatible with the C64?’ And we would say, ‘No, but it’s more expensive!’” According to Herd, when he demonstrated the 264, people asked, “Why isn’t it C64 compatible?” Herd gazed at months of hard work and replied, “What can I tell you? I made it work! Doesn’t that count for something?” It was a letdown for Herd, who hoped his computer would ignite more interest. Afterwards, he says he felt beat up. The request for backward compatibility surprised him, since backward compatibility was not at all on his mind as he designed the computer. Al Charpentier, who still attended CES with his own company, recalls his reaction to the new computers. “Once we left, the company marched down the Plus/4 path, which was wrong. It was just wrong.”

Bil Herd attending CES in 1984 (photo courtesy of Dave Haynie).

Aside from the 264 demonstrations, Commodore showed off its new software packages, mostly for the C64, including the Magic Voice cartridge. The massive Commodore booth even housed an arcade, which displayed 30 different C64 games. It was located in a darkened room where attendees could play the latest games like Satan’s Hollow, Zork, and Wizard of Wor, which included Magic Voice capability. Jack Attack, a game whose title poked fun at Tramiel’s outbursts, was also on display. Backstage, engineers demonstrated a touch screen, which used a series of infrared LED lights and light sensors around the edge of the screen to detect finger location. They also demonstrated a light pen for drawing pictures directly to the screen. Herd had a chance to meet the Commodore Texas engineers who created the speech capability for Magic Voice. As usual, he pulled them into his hard-living lifestyle. “It was really cool running into people like that back then and having a beer with these guys,” says Herd. “Me and the one guy nobody talked with, we went out terrorizing the bar one night together in Vegas.” Tramiel and Gould delivered a t press conference, where they announced Commodore sold over three million computers in 1983. They also claimed Commodore sold over a billion dollars worth of product for the year, including software and peripherals. Tramiel also announced a new VIC-20 price of only $79.00. Neil Harris re the press conference. “Jack gave a speech at the Consumer Electronics Show and it was a packed house. He announced that Commodore has just crossed $1 billion in revenue, which was a huge milestone.” Curiously, Tramiel’s demeanor seemed off to Harris. “He was not looking positive and upbeat. It was just a whole different atmosphere, even though he was making an announcement that should have been the grand triumph of his career.” Tramiel’s behavior made Harris suspicious. “I came out of the room saying

something is wrong because this was not the Jack Tramiel that I was used to seeing at these events. That was, to me, a sign that clearly something was going on.” At the conclusion of the press conference, Commodore sponsored a dinner party for employees and industry insiders. Afterwards, everyone returned to the Commodore booth to set up for the next day. The booth contained a large meeting space above the arcade. “We had a booth at the January CES that had a second level office space—a demonstration room to show the retailers,” explains Russell. The meeting space contained sofas and loveseats for the tired employees. There was also a small room for closing deals with distributors and a small office for Tramiel. Everything appeared normal except for a rising level of noise coming from inside the Commodore booth. “They were in the conference room at the top of the booth and if you went into the back area you could hear the yelling from the room,” says Finkel. “The general CES attendees couldn’t have heard anything unless they were in the back area of the booth where only the Commodore employees could get. It was clear they were having a big disagreement.” Deeper inside the booth, Russell had a front seat to the argument as he prepared a backstage demonstration of his Unix machine. “We were doing setup early in the show,” he says. “That was where the blowup happened that I overheard. I was setting stuff up next to the door where they were going at it.” Russell clearly heard the reason for the argument. “The blowup I heard was about the sons. Jack had kind of thought his sons were going to be in positions in the company, and Irving said, ‘No, I’ve rethought having your sons in these positions in the company.’ Jack said, ‘You know that’s what I want.’ Irving was like, ‘No.’ He started ranting and raving.” “They had a blowout,” says Finkel. “I think the key issue was Jack had decided to bring his sons into the business.” For the first time ever, Russell witnessed Gould display emotion, blasting Tramiel with all the vengeance of a Jack Attack. “He got heated during that. That was the only time I actually saw him upset,” he recalls. “I’ve been to board meetings with Irving where there was much worse news and he didn’t even flinch.” Gould felt Tramiel would gain too much power with his sons in key positions.

“It was definitely the sons and Jack’s effective control of the company by putting his sons in all the power,” recalls Russell. “You’ve got to realize, Sam was [already] running a big part of the company.” According to Bob Yannes, who was a friend of Russell’s at the time, Gould told Tramiel, “You are a detriment to the growth of the company.” Throughout the argument, Tramiel remained uncharacteristically calm. “Jack was trying to offer his case rather than attacking Irving,” says Russell. “He was arguing and Irving was going, ‘No.’ It was like seeing something blow up.” Russell, who was used to seeing Tramiel in full control of every situation, felt startled by Gould’s sudden domination. “I was more in shock that it was coming apart before my eyes,” he recalls. “Irving was like this guy who flew in with cute broads on his jet. It wasn’t like he was part of the company.” From the tone of the argument, Russell sensed the relationship between Gould and Tramiel was at an end. He was also about to lose his most valuable asset, the Jack card. “You played a Jack card and said, ‘If you don’t believe me, go ask Jack.’ All of a sudden, my whole source of power was walking out the door. It was pretty shocking to the people who were there.” The argument abruptly ended. Tramiel fled the hall as his life’s work disintegrated around him. Kit Spencer attended the show and ran into Tramiel as he left. “I was with him at CES when he quit and he said, ‘You’re looking good, Kit, from taking life easy. I’m going to do the same.’ And he took off around the world for about six weeks.” Tramiel entered the car his family gave him for his birthday. “Jack came up in the car and drove away, and wasn’t coming back. That was that,” says Russell. “His sons were kind of standing around there. Everybody asked, ‘Is Jack coming back?’ We said, ‘I don’t think so if you heard that conversation.’” [1]* Byte magazine, “Shakeout Predicted” (February 1983), p. 433. [2] Tape releases far outweighed cartridge releases, with 182 cartridge titles released and 428 tape based titles. There were no commercial disk-based releases for the VIC-20, owing to the limited quantities of disk drives produced by Commodore.

[3]Data Welt magazine (Data World) 1986 interview, translated from German. [4] Ibid. [5] Ibid. [6] Ibid. [7] Michael S. Malone, Infinite Loop (Currency, 1999), p. 102. [8] The C64 port used a software driven RS-232 implementation that was not reliable at speeds over 1200 bits per second.

CHAPTER 36

Dismissing the Founder 1984

On Friday, January 13, 1984, Jack Tramiel met with the board of directors in New York behind closed doors. Those who attended have not revealed what took place. “I know exactly what happened, but for personal reasons that I don’t really understand, neither my father nor Mr. Gould has ever been interested in documenting or revealing the actual events,” says Leonard Tramiel. “I wish there was some way to convince my dad to tell you the real story, but he won’t.” In an interview with a German magazine, Tramiel revealed, “To put it briefly, our philosophies were different. It came to the point where I said to him, ‘Either I lead the company my way, which I think is correct, or I must go.’ He [Irving] was completely friendly and said to me, ‘If you will not do what I want, then go.’ And I went.”[1] According to Bil Herd, “I think he had made a billion that year. The comment I heard at the time was they said, ‘Yeah, but you’re not the guy to take us to 10 billion.’” Bob Yannes agreed with the decision. “I think the market analysts were correct in that he was not the person to be running a $500 million or bigger company. You just can’t run it like your own little shop,” he says. Irving Gould’s main priority at the meeting was to rid Commodore of any influence from Tramiel by buying back his stock certificates. “It was announced that he was being ousted,” says Michael Tomczyk. “Fifty percent of his stock would be bought by Irving Gould at a fair price, and the rest of his stock he could sell on the market.”

According to Commodore employees, Tramiel walked out in the middle of the meeting. “He was so pissed off he just stormed out of the meeting and he was on a plane back to California before the board meeting was even done,” says Tomczyk. Tramiel claimed spending issues were part of the disagreement. “Just because we were a billion-dollar company, we did not have to throw money out the window like a billion-dollar company. Because if you spend more, you must increase prices. The man I had worked for was of another opinion. When business was good, he wanted to spend more. That is one of the points in which we had a different opinion.”[2] Gould’s stubborn refusal to use publicly traded stock to raise money for Commodore aggravated Tramiel. “We were also in disagreement about the question of financing. I was of the view that we should have made a new stock offering when our shares were highly traded—particularly since we had never increased the number of shares since we had listed on the stock exchange in 1962. With the 120 million dollars, which we would have earned by offering two million new shares, we would have paid back all of our debts to the banks and strengthened the position of our company. We could have survived the storm, without having to rely on the banks. The man I worked for thought this would water down his portion of the company and he would lose influence—that was absolutely wrong.”[3] Although Russell clearly heard Tramiel and Gould arguing over his plans for his sons, Leonard Tramiel does not believe it was the central issue. “The most persistent story was that my father wanted his sons to be the leaders of the company and Mr. Gould wanted it to be professional management, and that the disagreement was the foundation of my father’s departure from the company,” he says. “That is 100% not true. That had absolutely nothing whatsoever to do with it.” Dave Haynie believes his sons were part of the issue. “As a worker bee at Commodore, we weren’t in the boardroom then, but neither were Sam or Leonard,” he says. “It’s impossible to say if that was ultimately the reason why Jack left, but I think it was definitely on the table.” Chuck Peddle also believes his sons were at least part of the issue, despite Leonard’s denials. “I’m not saying he’s wrong, because Leonard is a very

truthful guy,” he says. “But to some extent, Leonard is biased because he needs to protect his family position.” Tramiel clearly wanted to see his sons Commodore, talked about it often, and he was making it happen.[4] Peddle felt Tramiel always believed Commodore was his company, even though Gould owned more shares. “I think he just never got over the fact that it wasn’t his company anymore,” he says. Tramiel later said of his resignation, “Naturally, that was very, very hard for me. But if I cannot lead the company in the way I consider correct, then it is no longer my company.”[5] Within days, the Philadelphia Inquirer reported Tramiel’s resignation. A press release issued by Commodore stated it was due to personal reasons. It read, “He had been thinking about it for some time, and each landmark that Commodore achieved brought the day closer when Tramiel felt he could go on to other things.”[6] Tramiel was out of Commodore for good. Over the last 25 years, he had been all over the world on business trips, but he never had time to see the sights and take in his surroundings. Now, he planned to revisit those people and places by spending the next year touring the world on cruise ships with his wife Helen. * * * Following Tramiel’s departure, newspapers were buzzing with speculation. The day after the board of directors meeting, a New York Times headline read, “Founder of Commodore Resigns Unexpectedly.” A few days later, another headline from the Times read, “Rift Denied in Tramiel Departure.” Kit Spencer began to feel Commodore sinking the moment Tramiel left. “To me, that was the pivotal moment when it really started to go down. Jack had some great strengths and some great weaknesses, but he really was the key behind Commodore. He steered it for 25 years through some pretty tough markets: adding machines, electronic calculators, digital watches, and computers. After he went, it just didn’t have the same drive or direction again.” Gould knew he wanted to turn Commodore into a professional, business oriented company, however he was not sure how. “He never really thought about a 10year technology company. He never thought like an HP,” says Peddle. “He never

had to do that before and it wasn’t in his background.” Less than a week after Tramiel’s departure, the New York Times declared, “Commodore Fills Tramiel’s Position.” Gould chose 54-year-old steel industry executive Marshall Smith. Smith recently completed seventeen years at ThyssenBornemisza Incorporated, with eight years as CEO. Gould began making his presence felt at Commodore. “I didn’t really see Irving do anything until he brought the guy in from the steel industry after Jack left,” says Russell. Most Commodore employees doubted Smith’s ability to lead a computer company. “I think Marshal Smith had his heart in the right place, but he was not a person who should be in the computer business,” says Haynie. “He wasn’t used to an industry that moved that fast. He came from steel, which changes every generation, not every six months.” Curiously, Sam and Gary retained their positions with Commodore. “[Jack’s] sons stayed on,” recalls Haynie. “They were there for at least six more months.” * * * The 264 computer had a lackluster reception at CES, but Commodore management decided to continue with the rollout, despite protests from Neil Harris. “At that point, with the management change, the company said, ‘Okay, this is our next thing. We showed it at the trade show, now we’ll roll it out.’ And I was like, ‘This is going to be a turkey of a computer system! What are we doing here?’” Andy Finkel, who originally ed the 264 as a budget system, had reservations with the new pricing scheme. “The idea that you would have a computer that is a Magic Desk or a Logo machine, or a Plus/4 that fires up into a word processor or spreadsheet—that part was great. The bad part, though, was that the computer really had fewer capabilities than the C64. Nothing we could do could really hide or change that fact.” Commodore had shown multiple flavors of the 264 at CES and decided to cancel the Magic Desk and Logo versions. Instead, it would focus on the most popular versions: one with 64 kilobytes of memory and one with only 16 kilobytes. “They didn’t think that the market for instant-on Magic Desk or instant-on Logo was enough to make it worth producing,” says Finkel.

However, Commodore would still release Magic Desk and Logo as conventional software for the computers. “You could still get the cartridge with Magic Desk or Logo,” says Finkel. The 16-kilobyte version of the 264, now called the C116, represented Herd’s original Sinclair killer. Commodore was about to repeat an old mistake by using a low quality keyboard similar to the original PET-2001. “It was tiny, but not quite as bad as a Timex-Sinclair membrane keyboard,” says Haynie. “It was more like a calculator keyboard.”

Comparing the C116 (top) with the Sinclair ZX-Spectrum.

The Plus/4—complete with Sinclair-style case and full-sized keyboard— included 64 kilobytes of memory and the built-in Trilogy software suite. Commodore also wanted to produce another 16-kilobyte version of the computer with a full-sized keyboard, called the C16. “The C16 came after CES,” says an obviously disgusted Herd. “The C116 became a C16 abomination. They stuck it in a C64 case.” The C16 looked remarkably like a VIC-20, except the case was black with light-gray keys. The most fanciful computer in the TED line, the talking 364, did not make it very far after CES. “I was in charge of that for about two weeks between the time it was handed to me and the time it was cancelled,” says Haynie. Commodore now faced the huge cost of developing manuals, packaging, advertising, semiconductors at MOS Technology and even ports of their best software titles. It was a massive risk, and Commodore banked on large sales in order to make the release worthwhile. Commodore marketing scheduled the release of the TED computers for May 1984, but met with inevitable delays due to RAM shortages and problems with their supplier, Micron. “We had to wait on the 16 kilobyte DRAMs, which were brand new back then,” says Herd. “All of this wouldn’t have mattered as it waited on the TED chip.” Much like the VIC-20, the final production design for the C116 went to Asia, where Sam Tramiel resided. Irving Gould must have worried about employing the son of a disgruntled company founder within his company. “A lot of the different models came out of the Japan office, which is where Sam was working at the time,” says Haynie. Bil Herd helped along the completion of the C116 design. “They had flown me to Japan to get it ready, which is how I found out some of the things that were going on,” he recalls.

While in Japan, Herd met the formidable Tony Tokai, whom he thought of as a Samurai warrior. “There’s two types of Japanese businessmen—there’s what I call the Samurai and then there’s the other people,” he recalls. “He was Jack’s hit man in the Asian market.” Similarities between Tramiel and Tokai were apparent. “He was born and bred in Jack’s shadow, as far as his management style,” says Herd. “He was a rough person. You didn’t cross him unless you had some pull of your own.” Herd played a game of dominance with Tokai. “I’m a Samurai too. I wasn’t raised that way, but I was born that way. When Tony would smack a chair, that was his way of telling his underlings where to sit,” he recalls. “I’m in his office in Japan and he smacked a chair cushion with his hand. I looked at him after he did that and I put my feet up on the table. By putting my feet on the table, what I said was, ‘I’m an American, I’m Bil Herd, and that shit is not going to work with me.’ So then, Tony and I started getting along a little bit. It was like two alphadogs.” It was up to Herd to ensure the TED computers ed FCC requirements. “I brought some FCC knowledge with me [from Pennsylvania Scale Company], so I was able to get a handle on that,” he explains. “It flew through FCC, which was a first.” Herd was determined to improve his communications with his fellow Japanese engineers. “I learned some Japanese because that was a way to become a better engineer,” he says. “Pretty soon I was going over there and moving around at will.” Herd memorized a few key phrases to help him in meetings. “I spoke enough Japanese that when my boss was being an idiot, I could say in front of a room full of people, ‘Ignore him, he’s being an idiot.’ They would all laugh and he would look at me and go, ‘What are you doing?’ I’d say, ‘He’s going to be dead soon. I don’t care,’ which meant he was going to be fired.” Herd encountered some inexplicable mistakes while in Japan. “We had some mighty funny things happen to the Plus/4 during the time when Jack was no longer with us, but his sons were,” he recalls. The first anomaly had to do with the plastic material used for the case. “Suddenly the case, which was supposed to be made of fire-retardant plastic, had

been re-specified to be a flameful plastic,” says Herd. “We were like, ‘Well, we normally keep very good track of these changes to these drawings and yet this one seems to be erased and penciled in.’” Early on, Herd decided to use the Commodore 64 power supply for the TED computers. Once again, he found another anomaly. “The one that really got us was that we had laid out the board for a circular DIN connector so we could use the Commodore 64 power supply,” he recalls. “The case drawings were property of Japan. They’re going straight to plastics and it had a square hole. I’m like, ‘That doesn’t fit the board.’ Out comes this prototype board, they said, ‘No, no. We’re going to redo the board with a square connector.’ I said, ‘We haven’t tested with the FCC with that, but I can put a square one back on the master one if you want.’ I think we ended up with a square one for a while and went back to circular.” Herd had a third and final surprise when his power supplies arrived. “When we got the power supplies, the round connector was plastic. In other words, it didn’t conduct electricity,” he says. “So we couldn’t get through FCC again, plus I’m dubious it would not have been breaking a lot. So we had to quickly get power supplies with metal collars on them. That was just the kind of thing that was happening.” Engineer Dave Haynie believes the changes could have been attempted product improvements, but does not discount the possibility of sabotage. “It’s very possible, but things are always being cost reduced,” he says. “Anything that is changed is supposed to go through engineering. We had a complete system for that where a responsible engineer signs off on everything after presumably ensuring it works. Things like that shouldn’t get through. So if something like that happens and we find it by surprise, well, there’s probably been some monkey business somewhere.” * * * Advanced press from Commodore marketing stated the Plus/4 was more highend than the Commodore 64, which falsely built up expectations. Commodore came surprisingly close to producing a product the market might have desired. If they succeeded in producing a functional graphical interface for the lowcost Plus/4, they might have had a popular product to compete with the overpriced Macintosh, which debuted at the same time.

At the June 1984 CES show, Commodore displayed the Plus/4, C116, and C16 computers. “You can see what a bad CES booth looked like because we didn’t go all out at the summer CES shows,” says Herd. Commodore was confident about the success of the Plus/4 and began preparations for manufacturing 600,000 units. “Let’s just step back for one second and talk about how Jack tended to do business,” says Neil Harris. “There would be a lot of products developed that would be shown at major trade shows. Jack’s favorite form of market research was what he called the cashbox survey. You put a product out there and if people bought it then you would make more of them. And if people didn’t buy it then you would make less or none of them. We would show products at a trade show and if it seemed that people didn’t like it, we would just simply never ship it.” Without Tramiel, Commodore no longer functioned properly. “What should have happened if Jack Tramiel hadn’t left at an untimely moment is that the Plus/4 would have gone away, never seen the light of day, and the company would have moved onto the C128, which would have kept the momentum moving forward,” says Harris. “Jack would have put a bullet in those products and moved on, but the new guys felt that they had announced it so they had to move forward.” During the design phase, Herd concentrated on keeping his target of under $50. “I actually knew the prices in my head of a 7407 gate on Hong Kong that week,” he says. “That’s how sensitive we were to prices.” Unfortunately, Herd’s dedication would amount to nothing. Herd was disappointed with the new software. “We designed it to cost 99 bucks, and we succeeded. Our cost was $49,” he says. “Then the software management, Sig and those guys, got hold of it, and they put that real bad Plus/4 software in it, and then they left for Atari. I think they were sabotaging it with software.” With the new software additions, the Plus/4 retail price soared. “They made the cost $299,” says Herd. “Well I’m sorry, but it was designed to cost $99 and it would have been a good computer for $99.” Dave Haynie believes the addition of software might have been part of a larger plan. “In retrospect, it seems like that was an ideal way to make Commodore all confused about what they ought to be doing while Jack set up his operations,” speculates Haynie. “I don’t think you could ever prove it, but it seemed that way

to a bunch of us. We have this product that nobody wants.” Robert Russell believes it was just poor marketing. “I think it was one of those things where the marketing and sales organizations wanted to run with something because people were starting to make comments that the Commodore 64 was aging,” he says. Neil Harris blames Commodore management squarely for releasing the Plus/4. “I think those decisions were made by group of executives who didn’t understand the process,” he says. “They were built for a very specific reason. Even though it became invalid, the new management went forward anyway, which was a big mistake. Jack’s way of doing things was very unusual. The new guys didn’t know it, and they said, ‘We’ve announced it so by God we’re going to ship it, and how do we make this the best we can?’” The difference between Jack Tramiel and Marshal Smith was evident. Tramiel brought products to CES, observed the crowd’s reaction and reacted accordingly. He had done this before with the Max Machine and other computers. The new management continued forward, ignoring the lousy press. Commodore’s VP of marketing Myrddin L. Jones was interviewed about the new computers (called the 264 at the time). It was painfully obvious he did not grasp the importance of software and computers. When Compute!’s assistant editor Selby Bateman asked, “Why buy a 264 instead of a 64 that has a word processor and, say, a Simon’s BASIC? It would be the equivalent of the 264 for less money.” Jones replied, “The difference is that the market is changing because the consumer is far more knowledgeable.”[7] It was a non sequitur. Irving Gould issued similar unsure statements in the May/June 1984 issue of Commodore Magazine. Commodore sold the C16 “abomination,” as Herd referred to it, in Canada and parts of Europe. Commodore bundled the computer with a datasette and marketed the product as a learning computer. It sold in supermarkets and discount stores. Commodore released the C116 computers in Europe to compete against the ZX Spectrum. Unfortunately, Sinclair had upgraded the keyboard in the ZX Spectrum+ around the same time. Despite the similar look and price, the ZX Spectrum dominated over the C116 in the UK. However, the C64 continued its

domination of the ZX Spectrum, making the C116 redundant. The Plus/4 was the flagship product in the series, which Commodore attempted to sell worldwide. Compute!’s Gazette called the Plus/4 software, “mediocre compared to the best commercial software for the C64.” Even worse, it cost more than the C64. Predictably, it fared poorly. It should have been apparent by early 1984 that there was no reason to release the 264 line due to the continued success of the C64. If anything, the computers would cut into C64 sales and confuse customers. “It was the same problem that everybody else who had gone after the Commodore 64 had run into,” recalls Haynie. “The C64 was too popular.” Rather than wasting their time working on the doomed Plus/4, engineers could have concentrated on making the next generation Commodore 64, the D128. Bil Herd receives solace knowing that his computer sold well when retailers later dropped the price. “When they sold them through some of the other mail order companies, they couldn’t keep them in stock at 99 bucks,” he says. “You can think whatever you want, but it was a great $99 computer.” The Plus/4 received almost no software from developers. Al Charpentier blames Tramiel for the oversight. “Jack didn’t understand the power of software,” he explains. “It’s a software game, and to win this game, you have to have lots and lots of people writing software for your product. He didn’t understand that you had to feed the market and you would have to pay for software to be done up front. His feeling was, ‘Let me put a lot of boxes on the shelf and the software guys should be privileged and happy to write stuff for my boxes’. That wasn’t realistic back then.” According to Haynie, there was no competition in consumer’s minds. “You had ten thousand programs for the C64 and not one of them would run on the Plus/4. There was no way it could compete,” he says.[8] Commodore took a technological step back and expected interest in their higher priced product. “The sound wasn’t as good, the graphics were a little bit better, but the C64 had better sprites and you needed that for games at the time,” explains Haynie. “What was better wasn’t better enough to offset anything else.” Commodore’s reputation suffered in the aftermath. Even though Tramiel instigated the projects, many thought it was a sign that Commodore could not

compete without him. Somehow, everyone knew the Plus/4 was a bad idea except for Commodore’s management and marketing teams. According to Chuck Peddle, Gould blamed Tramiel for the failure. “He went out and designed this follow-up to the C64, and he built up all this inventory, assuming he was going to go out and get everybody to upgrade their machines,” he says. “The product never went anywhere. If you talked to Irving after that, he said, ‘Jack almost broke the company. I left him alone and didn’t watch him.’” Gould clearly misunderstood Tramiel’s strategy of market-testing products before committing to a release. * * * The failure of the Plus/4 line of computers slowed Commodore’s remarkable upward trajectory. “After the Commodore 64, they took a wrong turn, so they lost almost two to three years where they could have been competing with IBM and Apple, in of coming up with a more powerful computer,” says Charpentier. “That was a significant loss of time. In the computer world, if you miss a generation, you can die pretty quickly.” Only the Commodore 64 allowed the company to continue. Thankfully, 1984 was one of its strongest years. It won awards as home computer of the year and dominated sales of every other system. Eventually, it became perhaps the most important system for colonizing homes with computers. Wired magazine compares the C64 to the Ford Model T, as the first home computer for the masses. The designer of the C64, Bob Yannes, witnessed the success of his creation from afar. “I felt good about it because it was something I designed and obviously a lot of people really liked,” he says. “Over the years, there’s been a lot of people who say, ‘That was my first computer and I really learned a lot.’ I meet people that are presidents of high-tech companies and they say, ‘The reason I got into computers was because I got a Commodore 64.’ I find that very satisfying.” Michael Tomczyk puts much of the success of Commodore on vertical integration. “Number one, Commodore vertically integrated the company almost by accident,” he says. “They needed to vertically integrate to stay in the digital calculator business. As part of this strategy, Jack bought MOS Technology. And when he bought MOS Technology, he accidently inherited the first self-

contained desktop computer developed by Chuck Peddle, which set the company on a course to become a home computer pioneer.” Tomczyk credits Tramiel’s vision of selling computers to everyone as key to Commodore’s early success. “The second thing Commodore did right was resist the temptation to try to become a business computer company—in order to fulfill Jack Tramiel’s dream to design computers for the masses, not the classes. They wound up resonating in the marketplace, which was extremely hungry for home computers. I don’t think anybody realized how hungry the market was at that time. Commodore capitalized on that hunger.” However, Chuck Peddle believes Tramiel should have focused on the business sector. “The problem is, the consumer computer didn’t have legs,” he says. “It was a great idea, it made tons of money, but it didn’t have legs. The business computer market, as you can see, had big legs.” Peddle believes Tramiel made an early misstep by not focusing enough on longterm research and development. “Breaking up Moorpark was the death knell for Commodore,” he says. “It effectively put him out of the computer business.” After January 1984, Tramiels shocking departure left his loyal employees wondering what to do. “You could just smell all the energy went out of the building when Jack left,” says Neil Harris. “You know, Jack was a challenge to work with, but he kept things moving.” Many doubted Gould and Smith could maintain the same aggressive approach to product marketing. “By the time Jack left, we established a remarkable strategy of developing an entirely new generation of computer every 12 months,” says Tomczyk. “Whether it’s a memory expansion, new applications, or different architecture, we were committed to having a new computer every year at the CES show.” Without Jack Tramiel at the helm, Commodore employees began to have doubts about the future of the company. “At the end of the day, you had a company that was built around one really strong, really smart guy—that being Jack Tramiel obviously,” says Harris. “There was no real succession plan and the next layer of management was generally not that strong. You get a guy like that leaving the company and the vacuum was huge.” Tomczyk compares Tramiel to the celebrated WWII general of the US Army.

“Patton, for all his faults, was very similar to Jack Tramiel,” he says. “He got the job done. A lot of people who criticized his management style, either in business publications or people who worked for him who didn’t like what he did, they also didn’t recognize that the army loves fighting for a winner.” Throughout 1984, Commodore’s share price fell from $60 to around $20. After the failure of the Plus/4, it was up to Marshall Smith and Irving Gould to find a next-generation machine capable of capturing a large market, otherwise Gould’s shares in Commodore would continue their downward plunge. The PET and CBM line of computers were in their twilight years and Gould needed a replacement. He wanted a powerful product to fight IBM. Meanwhile, Jack and Helen Tramiel continued touring the world. They only got as far as Sri Lanka and Hong Kong before Tramiel realized it was not the life for him. In mid-March, Tramiel ended his worldwide tour and returned home to get back to doing what he loved best: business. He was determined to crush Commodore.

(L to R) Gary, Sam, Jack and Leonard Tramiel.

[1]* Data Welt magazine (Data World) 1986 interview, translated from German. [2] Ibid. [3] Ibid. [4] After Tramiel left Commodore, he and his sons ran their next company together. This might be the best evidence of his intentions. [5] Ibid. [6]RUN magazine, (March 1984), p. 6. [7]Compute!’s Gazette magazine, “Some Answers from Commodore” (April 1984), p. 58. [8] Years later, Commodore dumped their remaining inventory of Plus/4 computers on the Eastern European market, primarily Hungary. The computer actually outsold the C64 in Hungary and received respectable software from local developers.

Epilogue

Christopher “Kit” Spencer, the man who launched the VIC-20 and Commodore 64 to major success in North America, remained semi-retired in the Bahamas, writing Commodore’s company newsletter and playing tennis. According to Spencer, “I think we’ve got it wrong. What you should do is to try to retire at 40 and then go back to work when you are 65.” As of this writing, he remains in the Bahamas and enjoys an active lifestyle. “I enjoy hiking and recently for my 60th birthday I trekked up to the Everest base camp at 18,500 feet in the Himalayas. I like to do things like that.” * * * Albert Charpentier (VIC, VIC-II chip designer), Robert Yannes (SID chip designer, C64 system designer), and Charles Winterble (VIC-II, SID, C64 project manager) received one million dollars from Atari for My First Computer. The product languished at Atari and by 1984, it was no longer commercially viable for release. Profits from the Atari venture were supposed to fund a new computer. “When we left Commodore, our plan was to design the next great computer with better graphics and better sound,” says Yannes. “We had a real good plan. We were going to talk to the game companies like Electronic Arts and find out what they were really looking for. [It was going to be] a computer that could play games, with graphics acceleration and a bit blitter. We were going to do sampled sound —a lot of the stuff that ultimately ended up in the Amiga.” The computer they previously created derailed their ambitious plans. According to Yannes, “We were well on our way to doing that, but the Commodore 64 became such a steamroller that no one was interested in financing a company that was trying to compete against it.” Instead, the trio started a sound company called Ensoniq. “We ended up going into the music business because we were very far along on the design on the

sound chip,” says Yannes. Commodore engineer Bruce Crockett also ed Ensoniq. The trio also invited Andy Finkel, but he remained at Commodore. “Sometimes I regret that,” he says. The resulting chip found its way into a rival computer. “At one point we were trying to raise some money for the company and we ed Steve Wozniak directly and wondered if he would be interested in this chip,” says Yannes. “He ended up telling Apple they should look into this. So our sound chip ended up going into the Apple IIGS.” In 1997, Singapore based Creative Technology bought Ensoniq for $77 million, making the trio multimillionaires. Al Charpentier is still an active entrepreneur and engineer. While interviewed for this book, he was marketing a bodyscanning device called Intellifit. R&D Magazine named the device the most technologically significant product of 2004. Yannes believes the media has treated Commodore badly. “I’m not speaking for myself here, but I think Commodore the company deserves more credit than it gets,” he says. “Based on the people I’ve met in the last 20 years, the Commodore 64 and the VIC-20 had a huge impact on people that it really isn’t given the credit for. It’s not like tens of millions of people bought them and threw them in the closet. They played with them for a while. It really helped a lot of people get comfortable with computers, to enjoy and appreciate computers. I’m not sure that the PC market could exist as it does now if something like Commodore hadn’t paved the way for it.” * * * Jack Tramiel’s second in command, Dick Sanford (VP of Finance and Executive Vice President of Commodore International) started his own company after leaving Commodore. He approached Tramiel with his plan, hoping for financing. According to Sanford, “He said, ‘Nope, won’t do it.’ It was just part of the lesson, you know. ‘You leave, I cut the cord.’ He’s a tough guy. He did not venture my company. I’ve seen him only a couple of times since.” Sanford believes Tramiel wanted to teach him a lesson. “It was really, ‘Do it on your own, son. You want to do this? Go do it.’ He knew I would do it. I knew I would do it too. It’s just another lesson learned, that’s all.”

In 1981, he started Intelligent Electronics, a direct competitor to the ComputerLand franchise. “I ventured that by myself and risked everything to build it,” he says. The company outlasted ComputerLand and eventually grew to a $3.7 billion dollar wholesale distributor. Sanford retired in 1998 and founded Operation Warm, which delivers winter jackets to inner city youth. * * * Chuck Peddle (6502 Inventor, PET Designer), Bill Seiler (PET, TOI, VIC-20 Systems Engineer), and his team of engineers went on to produce the state of the art Victor 9000. The machine turned out to be remarkably similar to the IBM PC released a year later. The Victor 9000 dominated business sales in Europe for a short period before the IBM PC overtook them in the late eighties. Peddle believes if Commodore had produced and marketed his computer, it might have dominated the business market. “If we had done a similar product to what we did, I’m not sure the IBM PC would have ever gotten off the ground,” he says. “We did pretty big damage to IBM with no resources. Having Commodore’s resources and the ability to raise stock money, I think it would have been a totally different world today.” * * * Decades later, Bill Seiler lives and works in California as an engineer. He cut his hair short and no longer surfs. “I do a lot of sidewalk astronomy. That’s my fun thing,” he says. “I set up on street corners in front of the Monterey Bay Aquarium or near the lighthouse down on Capitola Beach in Santa Cruz. I show people what I can see just for fun. I meet a lot of crazies that way.” He also named his dog GUI. * * * Years later, Peddle attended an event commemorating the anniversary of the personal computer. The event gave Peddle some perspective on his experiences with Jack Tramiel. “I got to make the world have personal computers because of Jack Tramiel,” he says. “He destroyed me, he destroyed my family, he did all

kinds of terrible things, but he gave me a chance to do something nobody else would give me necessarily.” Peddle eventually met with Tramiel at a company named JTS. “I asked for an audience with him privately,” he reveals. “I basically told him, ‘I think you cheated me, I think you did some bad things to me, but you gave me a chance to be somebody that nobody else could be, and I appreciate that.’” Bob Yannes, the Commodore engineer who was partly responsible for Peddle’s downfall, says, “I respect what he did. He did a lot of work on the 6502, which is a great processor, and he was responsible for the PET. Commodore wouldn’t have been in the computer business if it wasn’t for the PET.” Although Apple revisionists try to paint their founders as the innovators, it is clear Peddle and his 6502 had much more to do with the personal computer revolution. Steve Wozniak built his computer by assembling it from parts available on the market, but Chuck Peddle built his machines starting with raw sand. “He’s one of the founders of the microcomputer industry,” says Sanford. “He just is. He did a tremendous service for Commodore and for the industry.” Byte magazine, which documented the history of personal computers as it happened, wrote, “More than any other person, Chuck Peddle deserves to be called the founder of the personal computer industry.” Although Peddle’s experiences at Commodore were devastating, he feels it was worth it. “I got to live a life that nobody else has gotten to live,” he says. As of this writing, Peddle lives in California with his girlfriend and spends time with his kids and grandkids. He is a running enthusiast, a cat aficionado, enjoys reading science fiction (including works by former Byte magazine contributor Jerry Pournelle) and visits CES every year. He occasionally works with Bill Mensch at the Western Design Center and he has his own company THSTYME (pronounced “this time”), where he designs next generation USB and solid-state drive controllers. He manufactures his products in a new factory in Chennai, India. “Everybody keeps looking at me and saying, ‘You made all these people rich. Aren’t you rich?’ The answer is I’m not.”

Continued in…

Commodore

the amiga years

ISBN: 978-0-9738649-9-1 Their story will be told. 2012