In 1987, IBM launched the Personal System/2, or PS/2, with a single strategic goal: to reclaim the personal computer standard from the clone manufacturers who had copied IBM’s own design and undercut the company’s dominance. The machine was a technical triumph, introducing innovations like VGA graphics, the 3. 5-inch floppy drive, and the now-iconic round PS/2 keyboard and mouse port. But the strategy failed.
The line was discontinued in 1995, and IBM eventually sold its entire PC division to Lenovo in 2004 for $1. 75 billion. The little purple port, however, outlived the machine, the division, and the idea that any one company could own the personal computer. To understand why IBM built the PS/2, one must recognize what the company was before the PC existed.
In 1980, International Business Machines was the most valuable technology corporation on Earth, with a market capitalization approaching $90 billion by mid-decade, roughly $270 billion in today’s money. It employed over 400,000 people, and its mainframe computers ran the banking, airline, defense, and government systems of the Western world. The phrase “Nobody ever got fired for buying IBM” was not a joke; it was a description of market gravity. Annual revenue exceeded $50 billion, surpassing Exxon and General Motors.
But by the late 1970s, personal computers from Apple, Commodore, and Radio Shack were appearing on desks, and IBM had nothing to compete. The company’s internal development cycle lasted four years, while the personal computer market moved in months. In July 1980, a Boca Raton lab director named William Lowe walked into IBM’s corporate management committee with a radical proposal: do not build it the IBM way. Buy the processor from Intel, license the operating system from a small Seattle software company called Microsoft, publish the architecture to attract outside developers, and ship the machine in twelve months.
Lowe assembled a small team under the codename Project Chess, housed away from IBM’s layers of oversight. Day-to-day leadership went to an engineer named Don Estridge, an informal man in a company that measured hierarchy by the whiteness of a shirt. Estridge built the machine in a year, using Intel’s 8088 processor, Microsoft’s disk operating system, and an open expansion bus anyone could design for, called the Industry Standard Architecture, or ISA. Every signal and pin was documented and public.
In IBM’s mainframe world, this was unthinkable. In the personal computer world Estridge was building, it was the strategy. On August 12, 1981, IBM unveiled the PC at the Waldorf Astoria Hotel in New York. The company expected to sell perhaps 200,000 units over three years.
Within the second year, customers were buying 200,000 a month. First-year revenue reached $1 billion, and by 1983 IBM held roughly 80% of the personal computer market. But Estridge had built an open door that could not be closed. In February 1982, three engineers who had left Texas Instruments—Rod Canion, Jim Harris, and Bill Murto—sketched a business plan on a placemat at a Houston pie shop.
The PC’s architecture was open, and Intel sold the same processor to anyone. The only proprietary component was the BIOS, the basic input-output system that handled low-level hardware communication. Copyright protected expression, not function. If you could write a BIOS that performed every function without copying a line of IBM’s code, you could build a machine that ran every IBM program and owed IBM nothing.
Compaq used a clean-room process, splitting engineers into two teams: one studied the IBM BIOS and documented its behavior without writing code; the second, isolated team wrote a new BIOS from scratch. The result was 95% compatible with IBM’s software library. The Compaq Portable shipped in November 1982, and by 1985 annual revenue reached $53 million. In 1984, Phoenix Technologies industrialized the process, selling compatible BIOS to any manufacturer.
Dell, AST, Epson, Zenith, Tandy, and dozens of others began shipping machines that ran every IBM program at lower prices. By 1986, IBM’s share of the PC market it had created collapsed from 76% to 26% in three years. The clones had become IBM at two-thirds of the price. The open architecture chosen to get the PC built in twelve months had become the open door through which the profit margin walked out.
On August 2, 1985, Don Estridge boarded Delta Air Lines Flight 191 in Fort Worth with his wife Maryanne. The Lockheed L-1011 hit a microburst on approach to Dallas/Fort Worth and crashed short of the runway. All 137 people aboard died. Estridge was 48 years old.
At the time of his death, the Entry Systems Division he had built from a twelve-person team employed nearly 10,000 people. A middle school in Boca Raton was later named after him: Don Estridge High-Tech Middle School. With Estridge gone, the bureaucracy he had held at arm’s length closed around his division. The team grew from 4,000 to 10,000.
Development cycles lengthened. The PCjr, IBM’s attempt at a home computer, was built through the company’s standard committee-driven process at an estimated cost of $250 million. It shipped with a rubber keyboard and software incompatibilities, and was killed within two years. The PC/AT, launched in 1984 with Intel’s 286 processor, was a commercial success, but clones appeared within months at two-thirds of the price.
John Akers, who became chief executive in 1985, was a mainframe man. He had built his career where IBM held 70% market share and wrote its own terms. The PC market, where a competitor could replicate your hardware in a single product cycle, ran against every instinct of the mainframe business. By 1986, the consensus inside IBM’s Armonk headquarters was clear: the openness Estridge built into the PC was the wound.
The cure was a new machine, with a new bus, behind a wall of patents and licensing fees. The engineer who designed the proprietary barrier was Chet Heath. The machine was called the Personal System/2. On April 2, 1987, IBM unveiled the PS/2 in Miami.
Four models launched: the entry-level Model 30 at $1,695, the Model 50 at $3,595, the Model 60 in a floor-standing tower at $5,295, and the flagship Model 80 running Intel’s 386 starting at $6,995, climbing past $10,000 with options. The machines were redesigned from the ground up with smaller cases, 3. 5-inch floppy drives, integrated controllers, and a new video standard called VGA that pushed resolution to 640×480 pixels in 16 colors. The 15-pin analog monitor connector would become the standard for two decades.
Keyboard and mouse plugged into new six-pin mini-DIN connectors. The strategic core was inside the expansion slots. Every model above the entry-level used Micro Channel architecture, a new proprietary bus designed by Chet Heath. Micro Channel was technically superior: 32-bit data paths, bus mastering, and software-based card configuration.
But it was deliberately incompatible with every ISA expansion card ever produced. Peripheral manufacturers would need to design new products and pay IBM a royalty of 1 to 5% of revenue. IBM had fenced the design with patents and made clear it would enforce them. The world the PS/2 launched into had already moved past the point where a single company could reclaim it.
The PC revolution was a structural change from vertical integration to horizontal layers, from proprietary control to open commodity. Intel made the chips, Microsoft made the operating system, third parties made peripherals. The assembler, whether IBM or Compaq or Dell, was just one replaceable layer. In 1987, the combined share of the nine largest clone manufacturers exceeded 33% of the PC market, while IBM’s share had fallen to 27%.
For the first time, the clones collectively outsold the original. Intel’s processor revenue from clone makers was growing faster than from IBM. Microsoft’s licensing income from the compatible world already exceeded what IBM paid. In September 1988, seventeen months after the PS/2 launch, Compaq led a coalition of nine PC manufacturers in announcing a competing bus standard called EISA, the Extended Industry Standard Architecture.
The gang of nine included Compaq, AST Research, Epson, Hewlett-Packard, NEC, Olivetti, Tandy, Wyse, and Zenith Data Systems. EISA matched most of Micro Channel’s features but was backward compatible with every existing ISA card, and it was licensed royalty-free. The message was blunt: IBM would not own the standard. The PS/2 sold respectably at first.
Three million units shipped by September 1988, and IBM’s factories produced nearly 4,000 machines a day by mid-1987. The corporate market bought the machines in quantity, and IBM spent over $500 million a year on advertising. But the Micro Channel bus split the peripheral market into two incompatible worlds, and the smaller world was IBM. Third-party manufacturers followed ISA and EISA, where the volume was.
Compaq’s sales jumped 47% in the quarter the PS/2 launched. The innovations that had nothing to do with the bus war were universally adopted. VGA became the standard display connection for more than 25 years. The 101-key keyboard layout became the template every PC keyboard has followed since.
The six-pin mini-DIN ports carried the PS/2 name into the next century. The 72-pin memory module became the RAM standard for the entire 486 and Pentium generation. The clone makers took every innovation that was not locked behind a patent wall and left the Micro Channel bus on the other side. The market did not reject IBM’s engineering.
It rejected IBM’s control. Microsoft then delivered the blow the hardware rivals could not. IBM commissioned Microsoft to build OS/2, a protected-mode multitasking operating system meant to be the software half of the lockdown. But OS/2 1.
0, released in late 1987, was text-only, unstable, and demanded expensive memory upgrades. Meanwhile, Microsoft built Windows. Windows 3. 0 launched in May 1990 and ran beautifully on cheap ISA clones that cost half as much as a PS/2.
Microsoft sold 2 million copies in the first six months. The IBM-Microsoft partnership came apart publicly in 1992. OS/2 peaked at roughly 5 to 6% of the operating system market. Windows reached 90.
The PS/2 did not collapse in a single quarter. It eroded over years. IBM, which had employed 407,000 people at its peak, began cutting. In 1992, IBM posted a net loss of $5 billion.
In 1993, it posted an $8 billion loss, at the time the largest annual loss in American corporate history. John Akers was forced to resign on January 26, 1993, and the board hired an outsider: Lou Gerstner, a former RJR Nabisco executive with no technology background. The PS/2 line was quietly discontinued in July 1995. IBM reverted to building machines on the industry-standard ISA architecture under new names, including the ThinkPad laptops, which became the single bright spot in the wreckage.
In December 2004, IBM announced the sale of its entire personal computer division, including the ThinkPad line, to Lenovo for $1. 75 billion. A company that once held 80% of the PC market sold the entire operation for less than 2% of its peak market capitalization. IBM did not die.
It reinvented itself as a services, software, and consulting firm and remains a profitable corporation today. But the personal computer left IBM’s hands for good. The machine is gone, but the things it introduced are not. The VGA connector shipped on monitors and projectors for more than 25 years after the computer was discontinued.
The 3. 5-inch floppy format survived into the 2000s. The 72-pin memory module became the RAM standard for a generation. The 101-key keyboard layout is the one the world types on today.
And the six-pin connector, the PS/2 port, is still manufactured and soldered onto gaming motherboards by companies like ASUS, MSI, and Gigabyte, still used by competitive players who value hardware-interrupt input over USB polling. Nearly four decades after the computer it was named for shipped, the PS/2 port remains one of the oldest active interface standards in personal computing. The Boca Raton campus where Don Estridge’s team built the original PC and where the PS/2 was designed was sold and renamed the Boca Raton Innovation Campus. It now houses 36 companies.
Most of the people who work there have no idea what was built in those rooms. The PS/2 was built to close a door that should never have been opened, by a company that could not accept that the door was the whole point. IBM invented the open personal computer, watched the open standard escape, and built the PS/2 to take it back. The industry IBM created looked at the proprietary bus, the licensing fees, and the patent wall and said no.
IBM did not lose the personal computer because it lacked engineering. It lost because it tried to own what had already become everyone’s. The small round port that still ships on motherboards forty years later is the proof.
The market kept exactly what it wanted from the PS/2 and discarded the rest, including IBM itself.