Allen Baum was one of the bright young university graduates snared by Hewlett-Packard recruiters, and he immediately suggested that the company interview an older friend who designed computers, Stephen Wozniak. So when, in 1973, Hewlett-Packard offered Wozniak a job as an associate engineer in its Advanced Products Division, he jumped at the chance. The division made pocket calculators, which for Hewlett-Packard—with its traditional affinity for high-quality, low-volume electronic equipment—was a bold departure. The success brought by the arrival in 1972 of the HP 35, the first desk-top calculator to pack the punch of a slide rule, brought a blush to the entire division. For a time it was a whizzy place to work, and as competitors slashed the prices of their calculators, Hewlett-Packard concentrated on adding features to the HP 35 and giving them model numbers like HP 45 and HP 60. Six months after joining the company Wozniak was given his epaulettes and became a full-blown engineer. He, in turn, managed to persuade HP to hire his neighborhood chum Bill Fernandez as a lab technician.
Wozniak found the world of calculators and the problems they presented far removed from his exploits with minicomputers. He was assigned to work on a project to make refinements to the HP 35 and suffered the fate of many engineers who work for large companies when, after eighteen months of effort, the project was canceled. Myron Tuttle, an engineer who worked with Wozniak on the project that was code-named Road Runner, recalled, “I don’t think anyone in the lab was thought of as exceptional. Wozniak was one of the few people without degrees. He didn’t stand out. He was nothing out of the ordinary. He was a competent engineer.” Wozniak was intrigued by rumors of a hand-held terminal for the handicapped that was being designed in the company’s research laboratories; he applied for a transfer but was turned down. “They decided I didn’t have enough education.”
Aside from the repeated comments about his thin formal training, Wozniak enjoyed the way Hewlett-Packard allowed his mind to wander. He liked the doughnuts and coffee trundled around on a cart every morning, the regular paychecks, the attention paid to engineers (which included leave to appeal to the president if faced with dismissal), the manner in which the company enforced across-the-board salary cuts rather than resorting to layoffs, and the stock rooms that were open territory for engineers working on their own projects. From stock-room parts, Wozniak built Allen Baum an HP 45, converted Elmer Baum’s HP 35 into an HP 45 (and attached a company label that gave warranty information in Japanese), worked out a way of solving square roots on the less powerful HP 35, and also challenged Fernandez to see who was “the fastest square root in the West.”
While he worked at Hewlett-Packard Wozniak took lunchtime jaunts in light aircraft that belonged to his workmates. He conducted an eccentric private life from an apartment in Cupertino that resembled a bachelor’s version of the Bronx Zoo. Pet mice roamed around the calculator and computer manuals, and there were boxes filled with videotape players that a group of HP engineers had bought in bulk. The one substantial piece of furniture was a sofa that could be converted into a pool table, while the bedroom was furnished with a mattress and the sink was usually piled with dirty dishes. Apart from a fancy stereo system the center of Wozniak’s existence was still the telephone. He appropriated a used phone number for what he boasted was the Bay Area’s first dial-a-joke. Each day he recorded a new message on his answering machine selecting most of them from a book of two thousand Polish jokes along the lines of “When did the Polack die drinking milk? When the cow sat down.” Sometimes, after returning from work, Wozniak answered the phone, introduced himself as Stanley Zeber Zenskanitsky, and read jokes. After receiving angry letters from the Polish American Congress, he switched countries and made Italians the butt of his humor, though he still used the same accent and continued to introduce himself as Stanley.
When Pacific Telephone’s answering machine broke under the strain, Wozniak rigged up his own and asked callers to ring the phone company and complain about the tardy service. The phone company, monitoring the heavy traffic and pestered by a local store that had the misfortune of having a number similar to Wozniak’s finally gave him one of the lines from “the radio station bank” that were reserved for heavy use. One of the callers was Alice Robertson, a plump San Jose high-school student with long hair, wide eyes and a hefty laugh. Wozniak answered her call, chatted for a couple of minutes, abruptly announced, “I can hang up faster than you can,” and slammed down the receiver. That peculiar exchange marked the start of a nervous series of calls that eventually culminated in a date.
As he embarked on his first major amorous adventure, Wozniak also had to contend with nagging obligations to Alex Kamradt and Computer Conversor. Kamradt had engaged several other engineers who spent months trying to unravel Wozniak’s design. To help turn a prototype and some schematics into a product, Kamradt looked to the person who had accompanied Wozniak on several of his trips to Computer Conversor, Steven Jobs.
According to Kamradt, Jobs promised to take charge of guiding production on the terminal in return for salary and stock. Kamradt recalled, “He resented me having money. He was somewhat unscrupulous and he wanted to get as much as he could, but I liked his assertiveness.” Wozniak, who scarcely visited Call Computer’s one-room headquarters, was unaware of the scope of Jobs’s interest. “Steve listened to Alex. He was very attentive. He listened to what Alex said a terminal could do for his business.”
For several months Jobs worked with Robert Way, the head of a small engineering company that provided design services for electronic companies. Jobs monitored the layout of the printed circuit board and the design of a vacu-form case. Together with Way he drew up a bill of materials and a parts numbering system and also acquired a license from Atari for a video circuit that would hook the terminal to a television. Way found Jobs a hard taskmaster. “Nothing was ever good enough for him. He was the rejector.” Way also observed the division of responsibilities. “Every check I ever received was signed by Kamradt. The responsibility for seeing the design got done was Jobs’s.” After some months Way, bemused by Kamradt’s perennial optimism, threw up his hands and ducked out of the project. “They were the weirdest group of people I ever met in my life.”
While Kamradt worried whether he would ever get the terminal to work, Wozniak, stimulated by the Homebrew meetings, was working on his own computer. He subjected some of the new microprocessors to minute inspection and quickly found out that they hadn’t changed the essence of a computer. “I was surprised that they were like the minicomputers I had been used to.” Though microprocessors hadn’t changed the nature of the enterprise, diehards still looked to the early mainframes, when computer design was attacked by large teams, and hailed them as the grand old days when men were men. Yet even in the forties and fifties the challenge facing engineers was one of minimal design—though they were trying to limit the size of a computer to a room.
For microcomputer designers like Wozniak, the challenge was still to squeeze the maximum performance out of the minimum number of parts. A compact machine not only kept the cost down but was also the source of substantial pride. The size of the new components, the fact that a computer could be squeezed into a case the size of a bread box rather than manhandled into an office building, also made it possible for one person to exercise control over an entire machine. “In microcomputer design,” one of the Homebrew regulars remarked, “you could express yourself in a way that hadn’t been possible in the entire history of electronic computing.”
Microprocessors did, however, bring a change of focus. With a computer’s central processing unit reduced to a chip, engineers like Wozniak and Baum felt that some of the broader issues of computer design had evaporated. Instead, they were forced to focus on the best ways of linking the computer on a chip to a board of memory chips, to a television screen or printer, and to a typewriter keyboard. The data sheets accompanying the microprocessors prescribed the rules that bounded the microcomputer designer and left some of the purists feeling hamstrung. Allen Ba
um complained, “You’re stuck with what you have and you’ve got to make it work right. If something doesn’t work right, you cannot redesign it. It’s a lot less fun.”
If the problems of size had been eliminated, cost was still an issue for threadbare engineers. In 1975 microprocessors like Intel’s 8080 were selling for $179 and Wozniak couldn’t afford them. Baum heard that Hewlett-Packard’s Colorado division was experimenting with the Motorola 6800, a microprocessor introduced about a year after the Intel 8080, which along with a few accessory chips was being offered to employees at a steep discount. Wozniak placed his order while his workmate Myron Tuttle scampered out to buy a technical manual that explained the intricacies of the chip. The choice of microprocessor was the most important decision that a computer hobbyist could make. It became the cause of frustration and exasperation, the source of pleasure and satisfaction, and also shaped the slant of his entire machine. Wozniak’s choice of microprocessor ran counter to fashion in the summer of 1975.
That summer at the Homebrew Club the Intel 8080 formed the center of the universe. The Altair was built around the 8080 and its early popularity spawned a cottage industry of small companies that either made machines that would run programs written for the Altair or made attachments that would plug into the computer. The private peculiarities of microprocessors meant that a program or device designed for one would not work on another. The junction of these peripheral devices for the Altair was known as the S-100 bus because it used one hundred signal lines. Disciples of the 8080 formed religious attachments to the 8080 and S-100 even though they readily admitted that the latter was poorly designed. The people who wrote programs or built peripherals for 8080 computers thought that later, competing microprocessors were doomed. The sheer weight of the programs and the choice of peripherals, so the argument went, would make it more useful to more users and more profitable for more companies. The 8080, they liked to say, had critical mass which was sufficient to consign anything else to oblivion. Lee Felsenstein had plenty of companions who shared his belief that “the 6800 was another world. It wasn’t worth any attention.”
Wozniak bucked the trend and chose the 6800. His interest in the Motorola chip was shaped almost entirely by price but he also thought that it was more like his favorite minicomputers than the 8080. The signals that emerged from the 6800, for example, were synchrononous (and so bore a conceptual resemblance to the architecture of the Data General Nova) while the signals on the 8080 were less predictable. Wozniak started spending his time at Hewlett-Packard delving into the properties of the 6800: finding out how much memory it could cope with, the voltage that it needed, the speed with which it executed instructions, and the pattern of its signals. On paper he plotted out a design of a computer built around the 6800. The design was an enhancement of the prototype he had built for Computer Conversor. “I designed it just for fun. I could do a whole bunch of things I’d wanted to do five years before and didn’t have the money to do.”
The economics of the semiconductor industry were also in Wozniak’s favor. Chips seldom sold long at their introductory price. Competing devices from the dozen or so major semiconductor manufacturers usually ensured that prices would fall fast and dramatically. In the fall of 1975 the laws of the industry held true and played havoc with the pricing of eight-bit microprocessors. Wozniak first stumbled on the change when he and Baum traipsed to an electronics trade show in San Francisco and spotted a new microprocessor, the MOS Technology 6502, made by a Costa Mesa, California, company. The men at MOS Technology were aiming the 6502 at high-volume markets like copiers, printers, traffic signals, and pinball machines rather than the small computer-hobbyist market. The 6502 was almost identical to the Motorola 6800 and the MOS Technology salesmen pointedly stated that their company was trying to make a smaller, simpler version of the older chip. The similarities were so blatant that they eventually became the subject of a lawsuit between the two companies but for Wozniak and other hobbyists, legal squabbles were a distant blur. The critical issue was price. The Motorola 6800 cost $175. The MOS Technology cost $25. Wozniak fished a 6502 out of a large glass bowl brimming with microprocessors and immediately modified his plans. He abandoned the 6800 and decided to write a version of the computer language BASIC that would run on the 6502.
His decision to write the language and then to build the machine was tacit recognition of the importance of software. He envisaged using the computer to play the sort of games he had run across on larger machines, which consisted of bursts of typewritten commands and retorts that appeared on a teletype or television screen. One of the more popular games had the lovely name Hunt the Wumpus which let players journey through a maze filled with monsters. At all the club meetings BASIC had proved to be the most popular language on the Altair and the 8080 microprocessor. “At the club all we talked was BASIC. I had a chance to have the first BASIC for the 6502. I wanted to demo the machine quickly.”
Wozniak made every technical decision to satisfy his own interests and made an art of the homily “Adequacy is sufficient.” The deadlines, pressure, and spur were imposed by the fortnightly Homebrew meetings and also by the prospect of his wedding to Alice Robertson. After some weeks of dithering, Wozniak had finally decided to get engaged after tossing three dimes in the air and waiting until they all landed heads up. As he started on the software he also developed asthma and wheezed loud enough for the neighbors to hear through the flimsy plasterboard walls. Frightened that fluid would fill his lungs while he slept, Wozniak took to writing programming code until the small hours.
Wozniak found writing software a more arduous exercise than designing hardware. The shape and style of his first major piece of software were dictated by necessity. He spent several weeks studying the grammatical rules for BASIC and found they were similar to the rules for FORTRAN with which he was familiar. Faced with a choice between two versions of BASIC, Wozniak chose the simpler. He wrote the programs in pencil on paper and a colleague at Hewlett-Packard wrote a program that simulated the behavior of the 6502 and which ran on a Hewlett-Packard minicomputer. The Hewlett-Packard computer was used to test some of Wozniak’s programs. Wozniak conceded, “Fortunately I’d spent a lot of time in my math classes not doing math but trying to write compilers in assembly language when I didn’t have a machine. I had gone off in directions which I had no way of knowing whether they were good or bad.”
After he completed the code, he set about designing a computer and reverted to the schematics he had drawn for Motorola’s 6800 microprocessor. He compared the features of the 6800 with the MOS Technology 6502 and its slightly cheaper brother, the 6501. Wozniak found that with a couple of alterations to some of the electronic signals that affected the chips’ timing, his previous design needed no alteration—“I didn’t have to change a single wire or pin on my design.”
He used some of the techniques he had used in designing the Computer Conversor terminal to make significant advances over earlier designs like the Cream Soda Computer. The most significant difference was, of course, the inclusion of the microprocessor. But there were other advances that also helped to make the computer easier to use. Instead of using switches to toggle instructions to the computer, Wozniak attached a typewriter keyboard. He also used some chips called PROMS (programmable read only memories), which stored instructions that previously had to be entered into the computer every time it was switched on.
He was precise about the way in which the chips for his computer should be laid out on the bread board. He spent hours working out where the chips should be placed before plugging the sockets, the cradles for the semiconductors, into the board. Wozniak was more meticulous than most engineers when it came to making the wire connections between the semiconductor pins. He disliked the popular “wire-wrapping,” which tended to wreath boards in a spaghetti jungle of wires, and favored “point-to-point” wiring, which required laboriously snipping and soldering lengths of wire between pins. The fastidious approach paid off when it came to trou
bleshooting and made it far easier to find troublesome pins and spot faulty connections.
Wozniak’s private interests consumed more and more of his time. He carted his prototype to work and spent much of his time at the lab bench making further refinements. Especially after Hewlett-Packard announced that the calculator division would be moved to Oregon, Tuttle said, “We spent half our time working on our pet projects.” Tuttle had also bought a 6502 and was also burning the midnight oil, taking his prototype home and trying another approach. With their prototypes built, Tuttle, Wozniak, and another colleague approached their lab manager to suggest that Hewlett-Packard consider making microcomputers. Tuttle recalled, “It was one of those informal meetings. It wasn’t a big deal. We just sort of asked for five minutes and showed Woz’s board. We were told, ‘HP doesn’t want to be in that kind of a market.’”
When Wozniak took his unnamed computer to the Homebrew Club it received another cool reception. That wasn’t surprising since a poll at one of the club’s meetings in October 1975 showed that of the thirty-eight computers belonging to members, twenty-five were either Altairs or used an 8080 while only one used a 6502. Wozniak hooked his computer to a black-and-white television, connected a board of 4K bytes of memory chips that Myron Tuttle had lent him, and patiently typed in the BASIC. There was a certain amount of surprise that BASIC would run on a machine with so few chips, but most of the club members didn’t even bother to inspect the computer. Wozniak passed out schematics to the few who were interested and later put his new machine in perspective. “It wasn’t as difficult as some of the other computers I had designed.”
Return to the Little Kingdom Page 14
