In 1983, the producers of a Hollywood thriller called War Games needed a personal computer for their teenage hacker’s bedroom. The film would star Matthew Broderick as David Lightman, a high school student in Seattle who accidentally dials into a military supercomputer and nearly starts a nuclear war. The computer on his desk had to look real, look serious, and look like something a sharp kid might actually own. They chose the IMSAI 8080, a machine built in San Leandro, California, seven years earlier by a company called IMS Associates.
They chose it for a reason no manufacturer would ever put in a press release: the company was dead. It had filed for bankruptcy in October 1979, four years before the cameras rolled, and a dead company cannot negotiate a product placement fee, demand script approval, or sue. And so the IMSAI 8080, a machine its own maker had abandoned, became the most recognized personal computer in the history of American cinema. Its front panel, a row of twenty-two alternating blue and red paddle switches set against brushed aluminum, became the image an entire generation associated with the idea of a computer in a bedroom.
The man who built IMS was William Millard. Before the personal computer, before the franchise empire, before the billion dollars and the lawsuits and the exile, he was a ditch digger, a truck driver, and a welder’s helper. Born in 1932, Millard never earned a college degree. In the California of the 1950s, where the aerospace and electronics industries drew on a pipeline of Stanford and Berkeley graduates, that absence was a wall.
But Millard had an instinct for systems and a relentlessness that compensated for every credential he lacked. He went to work for IBM, not as an engineer, but as a salesman who understood data processing well enough to talk his way into government contracts. By 1961, he had built the first computerized police information network for Alameda County, a system linking precincts across the East Bay to a central database. Cities like Oakland, Berkeley, Hayward, and Fremont could pull records from a single electronic source instead of calling each other on the phone.
It was real work delivered on deadline, and it proved that a man without a degree could build systems that institutions depended on. He left IBM convinced he could build his own company. In 1969, he co-founded a software firm called Systems Dynamics with his wife, chasing consulting contracts for mainframe users. The company overextended, burned through its investors’ goodwill, and went bankrupt in 1972.
Millard was forty years old, broke, and looking for something to believe in. He found it in Werner Erhard, the former car salesman turned self-help impresario who had created Erhard Seminars Training, the consciousness-raising program known simply as EST. The pitch was seductive, especially to a man who had just watched his first company die: if you take total responsibility for your circumstances, you become the creator of your life. Millard did not merely attend EST sessions.
He absorbed them. He would later require his employees to attend, structure his management philosophy around the belief that willpower and total commitment could override any material constraint, and pressure executives to go through EST seminars and mountaineering courses designed to push them past their fears. Former employees described a culture that ran simultaneously on exhilaration and anxiety. In May 1972, Millard began again, doing freelance computer consulting from home.
By 1973, he had incorporated the operation as IMS Associates, with IMS standing for Information Management Sciences. The office was in San Leandro, a working-class city tucked between Oakland and the salt flats of the East Bay. The company had no product and no hardware to sell. It had contracts and two part-time employees.
One was a programmer named Bruce Van Buren. The other was a physicist and electrical engineer named Joe Killian, whom Millard hired at $12. 50 an hour. Killian was quiet, methodical, and had the kind of engineering mind that solves problems by understanding the physics first and the market second.
Within eighteen months, he would design the machine that made IMS Associates famous. In 1974, IMS Associates had a contract to build a computer networking system for car dealerships. The work required small, affordable machines, and the available options were all built around minicomputers like the DEC PDP-11 systems that cost thousands of dollars and were designed for institutional budgets. The project was drifting toward failure.
Then the January 1975 issue of Popular Electronics appeared on newsstands with a blue and gray metal box on its cover. It was called the Altair 8800. Ed Roberts, a former Air Force officer running a struggling calculator company called MITS in Albuquerque, New Mexico, had designed it around the Intel 8080 microprocessor as a last-ditch effort to save his business after the calculator market collapsed. The kit cost $439.
It could do almost nothing out of the box. It had no keyboard, no screen, no storage. You programmed it by flipping toggle switches on the front panel and read the output by watching tiny red LEDs blink in binary. But it was a real computer, and orders overwhelmed MITS within weeks.
Millard and Killian saw the Altair and recognized the answer to their contract problem. A machine built around the same Intel 8080 chip could replace the expensive minicomputers their client needed at a fraction of the price. They tried to buy Altairs in volume, but MITS, drowning in its own backlog, offered neither quantity discounts nor reliable delivery dates. The decision that created the IMSAI 8080 was not born from a grand vision of the personal computer revolution.
It was born from impatience. If MITS would not ship them an Altair, they would build their own. Joe Killian went to work. He did not simply photocopy the Altair.
He studied Ed Roberts’s design, kept what worked, and fixed what did not. The Altair’s central innovation was its bus, a set of 100-pin edge connectors that allowed expansion cards to slot into the motherboard. Roberts had sourced the connectors cheaply, and the architecture they created would eventually become the S-100 standard, an industry specification that powered an entire ecosystem of compatible hardware for a decade. Killian adopted the bus wholesale.
His machine would be electrically compatible with the Altair, which meant every peripheral card, every memory board, and every piece of software being developed for Roberts’s machine would also work on the IMSAI. But Killian made improvements that went far beyond compatibility. He designed a single 22-slot motherboard where the Altair used a segmented board requiring hand wiring between sections. He rotated the card slots 90 degrees so the front panel became just another card plugged into the bus, eliminating hours of labor-intensive point-to-point wiring.
And he built a power supply that delivered 20 to 30 amps at 5 volts where the Altair supply struggled at 8. In practice, that was the difference between a machine that ran reliably with a full complement of memory and peripheral cards and a machine that browned out and fried its own voltage regulators when pushed. The IMSAI 8080 was the first clone in the history of personal computing. It was better than the original.
Then there was the front panel. Where the Altair used uniform red toggle switches, Killian chose alternating blue and red paddle switches, wider and more substantial, set into a brushed aluminum faceplate. The choice was cosmetic, but it gave the IMSAI an identity the Altair, for all its historical priority, never achieved. The Altair looked like a piece of test equipment on a lab bench.
The IMSAI looked like it meant business. IMS Associates advertised the IMSAI 8080 in the back pages of Popular Electronics beginning in October 1975. The kit was priced at $439; an assembled unit cost $621. To a family in 1975, $439 was two weeks of take-home pay for a median household.
The first fifty kits shipped on December 16, 1975, less than a year after the Altair had ignited the market. By April 1976, the company was generating hundreds of thousands of dollars a month. The orders came faster than the team in San Leandro could assemble machines. Millard’s EST-fueled philosophy found its natural application: total commitment, total hours, no division between the mission and the life.
Around this time, Millard brought in Seymour Rubinstein as director of marketing. A programmer named Rob Barnaby joined around the same time. Neither man would stay long. Both would leave to found MicroPro International, where Barnaby, working from specifications Rubinstein wrote, created WordStar, the word processor that turned the personal computer from a hobbyist’s curiosity into a tool for writers, lawyers, and businesses worldwide.
Two of the most important people who passed through IMS left to build something that mattered more than any single machine, and the company that had given them their start never saw a cent of it. In 1976, Millard partnered with a former IBM manager named Ed Faber to open a retail computer store in Hayward, California, a few miles from the IMS factory. They called it Computer Shack. The concept was sound: the personal computer was too complicated for most buyers to order by mail, assemble from a kit, and debug alone.
What the market needed was a storefront where a customer could walk in, see the machines running, ask questions, and leave with a working system and someone to call when it broke. The name lasted until Radio Shack’s lawyers sent a letter. It became Computerland. Millard turned it into a franchise.
Each franchisee paid an initial fee and agreed to source the vast majority of inventory through Computerland’s central purchasing operation, a model that generated revenue from every store. By the end of 1977, there were Computerland stores in a dozen cities. By early 1978, the count was accelerating toward 100. From the moment the franchise model proved it could scale, Millard began to see IMS Associates not as a computer manufacturer with a retail sideline, but as a retail operation with a factory it might not need much longer.
The timing of that instinct was shaped by forces larger than any one company. 1977 was the year the personal computer stopped being a hobbyist pursuit and became a consumer product. Apple released the Apple II, a sleek, integrated system with a built-in keyboard and color graphics. Tandy launched the TRS-80 through its nationwide chain of Radio Shack stores.
Commodore shipped the PET, an all-in-one unit with a built-in monitor. These were not kits. They did not require soldering. They booted to a prompt a human could read.
The hobbyist era, the era of front panel switches and blinking LEDs, was no longer the center of the industry. By the end of 1977, Apple had sold tens of thousands of Apple IIs. Radio Shack sold over 10,000 TRS-80s in its first month and a half. The IMSAI 8080, with its lifetime production of roughly 20,000 units spread over three years, was being outpaced by consumer machines in a matter of weeks.
The IMSAI 8080’s golden age was short. At its peak, between 1976 and 1978, IMS Associates shipped between 17,000 and 20,000 units. Every IMSAI that left San Leandro went to someone who knew what a bus was, who understood why a 30-amp power supply mattered, who could open that steel cabinet and slot a new card into one of those 22 S-100 connectors. If you bought the kit, you built the machine yourself, connection by connection, board by board.
When you flipped the power switch for the first time and the LEDs on the front panel lit up, the machine was something you had made with your own hands. That bond between builder and machine lasted decades. But the machine Killian built was already being orphaned by the company that sold it. Computerland was working.
By late 1977 there were dozens of stores. By 1978, over a hundred. The stores sold IMSAI machines, but they also sold the Apple II, and the Apple II was the machine customers actually wanted. Resources that should have funded the next generation of IMSAI hardware were redirected to opening more stores.
The company tried to build a successor, the VDP-80, an all-in-one machine with a built-in terminal and floppy drives. It was unreliable, shipped late, and sold poorly. The factory that built the IMSAI 8080 was being hollowed out to feed a retail chain that sold other companies’ products. In October 1979, IMS Associates filed for bankruptcy.
The IMSAI 8080 was four years old. Two former employees acquired the IMSAI trademark and continued manufacturing for the devoted community that still wanted the machine. Computerland, the thing Millard chose over the machine, became enormous. In 1981, IBM selected Computerland as one of three retail outlets authorized to sell the IBM PC.
By 1982, annual sales exceeded $400 million. By 1984, they surpassed $1 billion. At its peak, Computerland operated roughly 800 stores. Bill Millard’s personal stake was valued at over $1 billion.
The ditch digger who never finished college was, on paper, one of the richest men in American technology. Then the lawsuits arrived. In the early days of IMS Associates, Millard had pledged 20 percent of the company’s stock as a convertible note to an investment firm called Mariner and Company in exchange for $250,000. The resulting litigation ended with a $141 million judgment against Millard.
He lost control. In 1987, he sold Computerland to the private equity firm Warburg Pincus for approximately $200 million. The empire he had built by abandoning the IMSAI sold for less than the cost of the judgment. The name Computerland was sold again in 1993, and the corporate entity ceased operations in February 1999.
Most of the 800 stores had closed by 1990. Millard had left the country, moving first to Saipan, then to the Cayman Islands. In 2013, he and his wife filed for bankruptcy protection there, owing $118 million in unpaid taxes and interest. The man who had built IMS and Computerland ended as a tax exile in the Caribbean, bankrupt for the second time in his life, while the $439 kit computer he abandoned sat in museums on three continents.
In June 1983, War Games opened in American theaters on a budget of $12 million. It earned over $124 million worldwide, making it the fifth highest grossing film of the year. The IMSAI 8080 was not a background prop. It was the vehicle of the plot.
The machine on screen was fitted with IMSAI dual 8-inch floppy drives, an IMSAI keyboard, and an acoustic coupler modem. But the most telling detail of the IMSAI’s Hollywood career is one the audience never knew. The text that scrolled across the monitor as Broderick typed was not generated by the IMSAI 8080. A faster CompuPro system sat hidden out of frame, triggered by keystrokes on the IMSAI’s programmable keyboard, producing the display in real time.
The most famous personal computer in film history was not even running during its most famous scene. What audiences took away was real enough. War Games told Americans that a teenager with a computer and a phone line could reach into systems that were supposed to be secure. President Ronald Reagan watched War Games at Camp David and, by multiple accounts, was visibly shaken.
He raised the film’s scenario with his national security advisers and the Joint Chiefs of Staff. They confirmed the premise was technically plausible. Eighteen months later, Reagan signed NSDD 145, the first presidential directive on computer security in American history. The film also helped catalyze the Computer Fraud and Abuse Act of 1986, the first federal law criminalizing unauthorized computer access.
A movie about a teenager hacking a military supercomputer with an IMSAI 8080 changed the law of the United States. The machine’s maker was bankrupt. The machine’s maker had abandoned it. And the machine was reshaping American policy from a teenager’s bedroom on a Hollywood sound stage.
Today, working IMSAI 8080s sell at auction for $2,000 to $5,000 and sometimes considerably more. The original War Games prop was appraised in 2010 at well over $25,000. The S-100 community is still active, trading boards on forums, restoring dead machines to working condition, and writing new software for hardware that has not been manufactured in decades. At the Smithsonian’s National Museum of American History, an IMSAI 8080 sits in the permanent collection.
At the Computer History Museum in Mountain View, the documentation is archived. The Center for Computing History in Cambridge, England, has a restored unit on display. The company that built the IMSAI is gone. The man who founded it abandoned it for a franchise empire that made him a billionaire and then collapsed into lawsuits.
The retail empire is gone. The billion is gone. But the machine is still here. It boots.
It runs. Collectors keep it alive. It sits behind glass in three countries.
And for anyone who has ever seen War Games, it is the image of what a computer looked like when computers were still strange and personal in the way that only a machine you built with your own hands can be.