For this article I have relied upon personal records and my collection of early MITS documents, correspondence, manuals, products, and memorabilia. Furthermore, I have discussed by telephone most of the key points of the story with many of the people involved. I very much enjoyed putting this account together and hope it proves both useful and interesting.
Ed Roberts
Though MITS was officially founded by four partners, H. Edward Roberts was the company's driving force and its real founder. I first met Ed in the summer of 1968 when he was assigned to the Effects Branch of the Air Force Weapons Laboratory at Kirtland Air Force Base in Albuquerque, NM.
Though Ed was a brand-new second lieutenant when he arrived at the lab, he wasn't new to the Air Force. He had been an enlisted man for several years and had been commissioned after the Air Force sent him to Oklahoma State University where he received a degree in electrical engineering.
I remember well one of Ed's first assignments, to purchase for our branch Hewlett-Packard's new, state-of-the-art 9100 desktop computer. Ed was attracted to that machine like a magnet. Within a few days of its arrival he devised a program for calculating the parameters of transistor amplifiers. Even today, Ed recalls clearly the impact the desktop 9100 made on him.
As for computers, both Ed and I had prior experience as do-it-yourself computer hobbyists in the early 1960's. While a high school student, I had built a series of analog computers. My most advanced machine, which included a programmable analog memory array, could translate 20 words of Russian into English. Ed, too, had built analog computers. He had also built digital machines that used relay logic.
Reliance Engineering
Ed's Air Force career had not kept him from dabbling in free enterprise. While stationed at Lackland Air Force Base in San Antonio, TX, for example, he was the sole proprietor of two one-man companies, Reliance Engineering and Reliable Radio and TV. One of Reliance Engineering's biggest jobs was the assembly of the electronics that controlled the movements of the animated Christmas characters in the windows of the Joske's store across from the Alamo.
From the time I first met him, Ed often talked of placing Reliance Engineering back in operation. He was utterly confident his entrepreneurial gifts would allow him to fulfill his ambitions of earning a million dollars, learning to fly, owning his own airplane, living on a farm, and completing medical school.
Once he teamed with Glen Doughty, a captain in our branch, to design and build an infrared intrusion alarm for his uncle's fish farm in Florida. A few months later, he and Stan Cagle, a civilian electrical engineer whom Ed had known when they were both college students, worked together to design and build a regulated power supply they intended to sell. That project, which was never completed, soon led to the formation of MITS.
MITS
By 1969 the Effects Branch of the Weapons Lab had become part of the Lab's well-funded, highly classified Laser Division. Though I was working with state-of-the-art laser technology and super secret projects, I still found off-duty time to pursue my favorite hobby, model rocketry.
Several times I had mentioned to Ed the high level of interest among model rocketry enthusiasts for miniature light flashers for night launched rockets and economical telemetry transmitters. In the summer of 1969, we decided to meet with Stan and discuss the possibility of forming a company to design and sell telemetry gear for model rockets.
This first meeting took place in the kitchen of Ed's home in northeast Albuquerque. Besides Stan, Ed, and myself, Bob Zaller, another officer from the Weapons Lab, was present.
We spent most of the meeting discussing a proposed line of telemetry products. In retrospect, however, our most important action that night was to elect Ed president of our infant venture,
Our second meeting, like most others over the course of the next 18 months, was held in a spare front bedroom in Ed's home. A principle item on the agenda was what to call our company. Ed preferred Reliance Engineering, but I objected. Because the Massachusetts Institute of Technology was the center of model rocketry research, I suggested we form an acronym around the letters MIT. Perhaps, I suggested, we could call the company MIT Systems.
Stan and I then tossed out ideas for the acronym. I suggested micro for the M and telemetry for the T. Within a minute or so, Stan responded with Micro Instrumentation and Telemetry Systems.
Though I liked the MIT connection, Ed was unsure about the name. Wouldn't we be referred to as mits? I insisted people would refer to us as M-I-T-S, just as MIT is referred to as M-I-T.
Ed's second point was more valid. Reliance Engineering, he argued, was an existing company with an established credit rating. I offered a compromise: Why not designate MITS as a subsidiary of Ed's original company? This approach was acceptable to everyone.
Incidentally, my premise about the enunciation of our company name proved wishful thinking. Later, we even capitalized upon the name by labeling as MITS KITS some kits we sold.
Years later, Ed joked with me about the "hundreds" of times he had to explain how MITS got its name. In retrospect, he probably should have renamed the company after he introduced the Altair 8800, but that's getting ahead of the story.
Tooling Up
Though MITS's affair with model rocketry was to last but one year, it set the stage for the chain of events that led eventually to the Altair. Therefore the story is worth telling.
As resident model rocket fanatic and MITS marketing director, one of my responsibilities was to specify the various modules for our product line. My first magazine article, "A Transistorized Tracking Light for Night Launched Model Rockets," had been published in the September 1969 issue of Model Rocketry magazine, and I recommended the flasher as one of our first products.
In their capacity as MITS design engineers, Ed and Bob refined my design in Ed's garage workshop while Stan, our production engineer, laid out and made the etched circuit boards in his apartment. The TLF-1 light flasher soon followed.
Within a month, Ed, Stan, and Bob had completed work on two transmitters plus a variety of modules. In the meantime, I was hard at work writing "The Booklet of Model Rocketry Telemetry."
By October the circuit designs for the product line were finalized, and I wrote a press release and mailed it to Model Rocketry. It was published in the December 1969 issue.
While waiting for the release to appear, Ed, Stan, and Bob assembled hundreds of modules while I wrote operating instructions, designed an order form, and mimeographed big stacks of our self-published booklet. I also launched a series of rockets equipped with MITS transmitters and modules, all the time hoping for a photogenic crash that would demolish a rocket payload section while leaving the instrumentation unharmed.
In late 1969 we decided to incorporate. Each of us was given 950 shares of stock with the remaining 200 shares going to our attorney.
Each of us also made a contribution of cash and equipment to MITS. My $100 check was dated January 16, 1970. Ed had insisted that none of us become "silent partners," and, beside providing needed capital, the cash donations gave each of us a vested interest in the future of MITS.
In March 1970 the first MITS advertisement appeared in Model Rocketry.
The April issue of Model Rocketry included a second MITS news release. Also included was an attractive photograph showing one of our transmitters surrounded by six modules.
In spite of our countless hours of work, by May we had sold only a hundred or so transmitters and modules. As marketing director, I suggested we might increase sales greatly by cutting prices. How? By converting our line of preassembled modules into kits.
We decided to test this new approach by converting the TLF-1 light flasher into two kits, one with a dual flash rate and the other with an adjustable flash rate. The first ad for these MITS KITS appeared in the July 1970 Model Rocketry.
Popular Electronics Magazine
Big changes took place at MITS in the summer of 1970. As early as May we had recognized that our fortunes would never be made by selling model rocketry telemetry instruments to precocious teenagers and university professors. About this time, Bob Zaller, who was soon to be married, decided to leave MITS. (He later returned after the Altair was introduced.)
I also made a big change. In late 1969 I had decided to leave the Air Force upon completion of four years of service to become a freelance writer while continuing to work with MITS. I left the service on June 11, 1970 and immediately began work as the night attendant at the parking lot of Albuquerque's airport, the Sunport.
I took this night job thinking it would provide plenty of time for writing, and it did. The salary, however, was only a fourth what I had earned as an Air Force captain. Worse, I had to live with the pitiful looks on the faces of my former commanders and co-workers each time they drove out of the parking lot after returning from their frequent trips. They thought I was crazy to trade a position in state-of-the-art laser research and development for the parking lot.
In March, I sold my first article to Popular Electronics magazine, a feature about light emitting diodes. At one of our midnight meetings I suggested that we emulate Southwest Technical Products and develop a project article for Popular Electronics. The article would give us free advertising for the kit version of the project, and the magazine would even pay us for the privilege of printing it!
Ed had received a big bag of integrated operational amplifiers and comparators from a friend, so several times we seriously discussed using these chips to develop a kit analog computer. Since junior high school days I had been an abacus user, so we also discussed the possibility of making a solid state abacus, using red LEDs for the beads. Finally, I suggested we design an infrared voice communicator.
We decided upon the infrared communicator, so I contacted the magazine, and they agreed to consider the article. They also decided to hold my feature about LEDs and published the two articles as a pair. That summer, while we continued to fill model rocketry orders, Ed began designing what I called the Opticom.
In late July, before the Opticom was ready, I received a call from Leslie Solomon, technical editor of Popular Electronics. Les was coming to Albuquerque with his wife and daughter. Could he stop by for a visit?
I was elated. By then I was writing a monthly column for Model Rocketry, but this would be a chance to meet an editor from a magazine with considerably more clout. Furthermore, this would provide the opportunity to introduce Les to Ed and Stan and discuss our kit ideas.
The Solomon family arrived at my mobile home on Monday, August 3, 1970. Les and I retired to my tiny electronics workshop while our wives spent the afternoon visiting. That evening, we met Ed and Joan for dinner at the Beef and Bourbon, a steakhouse on North San Mateo Street several blocks from Ed's house. Stan had to work that evening and couldn't join us.
A summer thunderstorm brought welcome relief from the afternoon heat. While it thundered outside, Les poked fun at the restaurant's decor while Ed and I silently wondered how we would ever manage to pay the bill. After the meal, our wives visited and the children fidgeted while Les, Ed, and I spent a couple of hours discussing MITS and our proposed Opticom project. Les seemed enthusiastic about the project and encouraged us to get it completed as soon as possible.
We discussed other matters as well. Like how many kits could we expect to sell. (Who knows? Maybe a few hundred, maybe a thousand.) We also talked about writing as a profession. Detecting my optimism about becoming a full time freelancer, Les volunteered that it was next to impossible to make a living from writing outside New York.
Les Solomon's visit provided all the motivation we needed to finalize the Opticom. When the first transmitter and receiver pair were completed, I field tested the units and wrote the construction article. Since the design was Ed's, I bylined the article with both our names. Popular Electronics soon paid for the piece with a check for $400 which I deposited in the meager MITS checking account.
In the meantime, Ed arranged to finance the Opticom kits by borrowing a few thousand dollars from an Air Force friend. We also decided to move our Opticom kit production line into a building.
Since I worked nights, I was assigned the daylight task of renting a building. I began looking on September 30 and by October 9 narrowed the search to a former snack bar called The Enchanted Sandwich Shop. I rented the small brick building for about $100 a month.
My LED feature and the Opticom article were featured on the cover of the November 1970 issue of Popular Electronics. When the magazine appeared in late October, we began receiving as many as a dozen orders a day. But within a few weeks, the surge slowed to a trickle. We eventually shipped a little over a hundred Opticoms--far fewer than we had hoped.
The Split
Shortly before the Opticom article appeared, Ed and Stan had begun work on a desktop digital calculator. As Stan recalls it, Ed was so intrigued over the prospect of building a calculator he was willing to use TTL logic chips. Fortunately, Stan happened to see in Electronics magazine an ad for a calculator chip set manufactured by Electronic Arrays, Inc.
Ed wanted to move directly from the Opticom to a calculator kit before bigger companies became involved, but Stan and I held back. Stan wanted to use up our remaining Opticom parts and lenses by continuing our plan to develop an infrared intrusion alarm kit, which was nearly ready, and a solid state laser. Remembering the competition we faced soon after introducing the telemetry line, I felt the calculator venture was very risky.
The difference in opinion over which path to take led directly to a permanent split. One night in early November, Stan visited the parking lot and suggested he and I offer to buy Ed's stock. But on my salary, I could barely put food on the table.
Though we disagreed with Ed's timing of the calculator idea, Stan and I realized Ed would go ahead with the calculator project with or without us. As things developed, Ed offered to buy our stock.
Stan, Ed, and I held our final meeting as MITS partners on November 10, 1970 in the parking attendant's booth at the Sunport. With help from his Air Force friend, Ed offered to buy our stock for $300 cash, $300 by the following March, and $350 in equipment. I took my equipment in the form of unsold model rocket telemetry gear.
Stan and I had both agonized over our decision to leave MITS. Even though we had never paid ourselves a salary or a bonus (there simply wasn't the money to do so), for me the decision to leave MITS was harder than resigning from the Air Force.
On the other hand, the excitement of seeing my first two articles featured as cover stories in Popular Electronics was still fresh in my mind. Leaving MITS would provide much more time to develop my budding writing career.
MITS Enters the Calculator Business
Ed's uncanny ability to recruit engineers, technicians and financial backers has always served him well. By the time Stan and I signed the papers transferring our stock to Ed, he had teamed up with Bill Yates, a young second lieutenant from the Laser Division at the Weapons Lab. He also secured additional financial help from another officer.
Ed moved the MITS assembly line from The Enchanted Sandwich Shop back to his garage for the next several months. He then moved MITS to a rented house at 2016 San Mateo, N.E. In the meantime, he had acquired a chip set from Electronic arrays and began work in earnest on the calculator project. The first crude prototype, however, failed to work. The expensive chips had been installed in their sockets backwards!
In August 1971 I left the parking lot to become a full time freelancer. I had just sold a pair of articles about semiconductor lasers to Popular Electronics, one of which described a solid state laser transmitter and receiver.
The laser project had been on the agenda before Stan and I left MITS, and Ed agreed to sell a kit to use up unused Opticom parts and lenses. In return for a royalty from MITS, I agreed to write assembly manuals for the transmitter and receiver plus a manual of kit assembly hints. Consequently, I spent a good deal of time at MITS during the final stages of the development of the 816.
Since we had both built Heathkits, I told Ed the laser manuals would meet Heath's standards. He must have liked them, for after the manuals were completed Ed asked me to write the assembly manual for the calculator project. In return, he would give me a calculator.
To borrow a phrase from Ed's lexicon, the calculator project wasn't trivial. Indeed, it is fair to say that in many ways it was more complex than the Altair.
In the last few days before the 816 calculator appeared on the cover of Popular Electronics, the MITS operation at 2016 San Mateo was a beehive of activity. I spent two very busy weeks writing the 64-page assembly manual. The laser project had been published in the October 1971 Popular Electronics, but the hundred or so orders that came in caused only a small ripple in the onrushing press to complete the calculator.
Ed's article about the MITS 816 calculator was featured on the cover of the November 1971 issue of Popular Electronics. An accompanying editor's note described the 816 as "an exciting new breakthrough construction project--a modern, high-speed 16-digit calculator."
MITS offered a kit version of the 816 for $179 and an assembled machine for $275. Subsections of the machine and the circuit boards could also be purchased. For only $2, ambitious do-it-yourselfers could purchase complete mechanicals, circuits diagrams, and foil patterns.
The 816 calculator was a major success, and, for the first time, MITS earned a profit. Moreover, the 816 article marked an important turning point in hobby electronics, for it was a portent of the eventual arrival of low cost personal computing.
Consider this: Even before the 816 design was completed, Ed had designed a 32-step programming unit that would transform the machine into a programmable calculator. Complete interfacing terminals for the programmer, which was scheduled for introduction in April 1972, were included on the original 816 CPU circuit board.
Besides being too small for the burgeoning company, the house MITS occupied on San Mateo was scheduled for demolition so the street could be widened. Therefore, in 1972 Ed moved MITS to a larger building at 5404 Coal Avenue S.E. Eventually, MITS settled into a series of adjacent storefronts at 6328 Linn Avenue S.E., just a few blocks from my mobile home. There a wave soldering machine and an efficient assembly line were set up.
At the Linn Avenue operation Ed added a technical writing staff and even a receptionist. Though I wasn't needed for manual production, he and I collaborated on a series of magazine articles about digital logic and one of the first published calculator books.
MITS eventually introduced a line of compact calculators with LED displays. While MITS was selling many thousands of calculators, the big companies began their move into the field. Eventually, MITS was forced out of the market, and by 1974 the company was some $200,000 in debt. Discouraged but not down, Ed decided to leapfrog the calculator industry by developing an even more powerful product.
The Altair 8800
The defunct MITS calculator line had evolved from a mail order magazine project for electronics hobbyists into a bonafide consumer business. For his new product, Ed was to return to the marketing strategy that had served MITS best.
The new product was his most ambitious yet, an affordable microcomputer designed around Intel's new 8080 8-bit microprocessor. The project would fulfill Ed's lifelong ambition to design a working digital computer. And, if successful, it would save his company from bankruptcy.
Though he was fully prepared to sell the computer he planned by means of ads in electronics magazines, the method he had used to sell calculators, the Popular Electronics connection intervened. It so happened that Arthur Salsberg, the magazine's editorial director, had been actively searching for a computer project since early 1974.
Art's interest had been stimulated by an ASCII keyboard and encoder project designed by Don Lancaster of TV typewriter fame. Don's project, which was available as a kit from Southwest Technical Products, was the cover story of the April 1974 issue of Popular Electronics.
Art discussed the possibility of a computer project with Les Solomon. They eventually located a microcomputer trainer project by Jerry Ogdin. Art scheduled the trainer for the December 1974 issue of Popular Electronics.
In those days there was a healthy rivalry between Art's magazine and Radio-Electronics. As Art recently wrote in a letter recalling that early period, Jonathan Titus's Mark-8 8008-based computer in the July 1974 issue of Radio-Electronics caused Ogdin's microcomputer trainer project to be placed on hold. "I felt as if the rug was pulled out from under me," Art wrote. He very much wanted to "top their article."
Art asked Les if the knew of a more advanced computer project, particularly one using Intel's new 8080 microprocessor. Les was aware of Ed's project, so Art asked him to call MITS to see if Ed could deliver an article in time for a winter issue. "Tell him that he's got to have an attractive cabinet in order for it to be a cover story," Art recalls telling Les.
Soon Les raced into Art's office to tell him Ed could deliver a computer project in time for the January issue. "January is always the best-selling newsstand issue we've got," Art observed.
A few weeks later Ed called Art to inform him the computer would be housed in an attractive, multi-colored Optima cabinet with a shadow box design. Art postponed Ogdin's project, slating it for use as a backup in the event MITS didn't come through.
In the meantime, Ed, Bill Yates and a few others left over from the post-calculator bust were hard to work preparing the prototype computer. Ed designed the interface logic for the 8080, a 256-byte RAM memory, a 2MHz clock, and the front panel logic for the 25 control/input switches and 36 indicator LEDs on the machine. Bill Yates laid out the foil patterns for the circuit boards.
Ed also made what was to prove a momentous decision: He included provisions for an open bus so additional memory and peripheral cards could be added later. The oversize Optima cabinet could accommodate up to 16 additional cards. Therefore, Ed designed a hefty 8-ampere power supply for the machine, having no idea that even this much power would later prove inadequate for the dedicated computer fanatics who stuffed their blue and gray cabinets with peripheral cards.
Ed shipped the completed prototype via REA to Popular Electronics and then flew to New York to demonstrate it for the editorial staff. Alas, the machine never arrived. It was apparently lost or stolen at Kennedy Airport.
Nevertheless, Ed spread out the circuit diagrams and explained the operation of the machine. He then accompanied some of the editorial staff to dinner at an Italian restaurant before leaving for Albuquerque.
Ed recently recalled how troubled he was during the flight home. He had managed to borrow an additional $65,000 to float the computer project, but in spite of the magazine's assurances he had no firm agreement they would publish the project. "What really bothered me," he later told me, "was that Les Solomon said 'I think we're casting our pearls before swine on this one.'"
Art needed a computer right away for front cover photography, so Bill Yates put together a non-functional mock-up and shipped it to New York. Later, they also built a second prototype and shipped it to the magazine.
Then, there was the matter of giving the machine a name. David Bunnell, vice president of marketing and advertising manager for MITS, the jobs I once held, came up with three pages of suggested names. among his favorites was Little Brother.
Ed eventually called the computer the PE-8, but Les Solomon felt that was a rather dull name for such a powerful and momentous product. Les discussed the matter with associate editor Alexander Burawa and assistant technical editor John McVeigh. Al later remembered saying. "It's a stellar event, so let's name it after a star." Within a few minutes, John McVeigh said "Altair."
Les called Ed to try out the new name, but Ed's concerns were elsewhere. He told Les he didn't care what they called the computer so long as MITS could break even by selling 200 of them.
The Altair was featured on the front cover of the January 1975 issue of Popular Electronics as a "Project Breakthrough! World's First Minicomputer Kit to Rival Commercial Models." The magazine appeared on newsstands a week before Christmas of 1974.
In the accompanying article by Ed Roberts and Bill Yates, MITS offered a complete kit version of the machine for an incredible $397. A fully assembled version was available for $498.
Art Salsberg titled his editorial "The Home Computer is Here!" He wrote, "we were determined not to present a digital computer demonstrator with blinking LEDs that would simply be fun to build and watch, but suffer from limited usefulness... What we wanted for our readers was a state-of-the-art minicomputer whose capabilities would match those of currently available units at a mere fraction of the cost."
Art ended this momentous editorial by promising "There'll be more coverage on the subject in future issues. Meanwhile, the home computer age is here--finally."
Well, maybe. While tens of thousands of readers eagerly read every letter and comma in the Altair article, Art was questioned about the wisdom of the piece by one of his superiors. How could he justify the Altair project when no computer companies advertised in the magazine? This concern was not neutralized when MITS bought a full page ad in the February issue. The ad ran across from the second installment of the Altair article.
Meanwhile, back in Albuquerque, orders came flooding in. The response was overwhelming. Already backlogged with orders, Ed didn't even have an operator's manual for the Altair. He called in early January and said, "I'm going to make you an offer you can't refuse."
I bicycled over to Ed's office where he offered me an assembled Altair in return for a quick job on the operator's manual. He was right, it was an offer I couldn't refuse.
The World Altair Computer Convention (WACC)
The flood of Altair orders soon had Ed hiring more people and looking for bigger quarters. MITS organized an Altair user's group, and in April 1975 several Mits marketing people drove the MITS mobile, a motor home equipped with Altair equipment, on a tour through Texas. The MITS mobile team was eventually to visit many cities across the country, giving seminars, staging slide shows, and distributing literature, catalogs, and door prizes.
By June 1975, David Bunnell was editing a monthly tabloid called Computer Notes, a Publication of the Altair Users Group. In the November/December issue of Computer Notes, Bunnell announced in a banner headline ALTAIR CONVENTION.
The meeting, which was Bunnell's brainchild, was officially called the MITS 1st World Altair Computer Convention. It was scheduled for March 26-28, 1976 to coincide with the completion of the move to the new MITS headquarters in a brand new building adjacent to the Albuquerque Sunport.
The WACC deserves credit as the first major microcomputer convention. But the WACC was responsible for an even more momentous development, the formal arrival of competition. The lobby of the Airport Marina hotel where the WACC was headquartered buzzed with rumors about some people from Processor Technology who had rented an upstairs suite.
I made my way through the crowd and peered over the heads of the curious onlookers and saw the future: memory boards priced cheaper than those sold by MITS. The Altair's open bus had paved the way for the arrival of a microcomputer industry.
The Legacy of the Altair
An era ended in 1977 when MITS was sold to Pertec Computer Corporation. Ed stayed with MITS for a while, but eventually moved to a 900-acre farm in Georgia. He is now in his third year of medical school at Mercer University. He also heads a new company called Georgia Medical Electronics.
Today, comparatively few users of personal computers have ever heard of MITS and the Altair 8800, much less Ed Roberts. This is unfortunate, for Ed did for computing what George Eastman did for photography.
No, Ed did not invent the microcomputer. That credit belongs to the brilliant engineers who designed the early microprocessor and calculator chips.
Nor was the Altair a perfect machine. Ed himself admits the infamous 4K memory board was a major mistake. MITS made other mistakes as well, some of which it candidly admitted in Computer Notes (which was subsequently acquired by Creative Computing).
As for those who criticized the limited capacity of the 8-ampere power supply, how was Ed to know an industry would spring up almost overnight with the sole purpose of supplying peripheral boards for his Altair? Some of these boards, like the Godbout Electronics 4K static RAM, consumed nearly two amperes!
Sure, there were problems, but consider what MITS accomplished! The first computer stores anywhere were set up to sell Altairs. The open Altair bus paved the way for a microcomputer revolution. And everyone who uses Microsoft Basic can thank Ed Roberts for the decision to select the version of Basic developed by Paul Allen and Bill Gates for the Altair over the other languages he considered.
In the final analysis, MITS pioneered nearly every aspect of today's microcomputer industry. Computer shows, users' groups, newsletters, seminars, software exchanges, peripherals, software products, quality documentation, and cheap computers, all commonplace today, were first pioneered for the personal computer market by MITS.
Recently Ed and I lamented the fact so many of today's computer users think Apple, Radio Shack, or even IBM invented personal computing. We also wondered about the distorted versions of the early days at MITS written by some of today's computer journalists.
I'll have much more to say about the early days at MITS and the Altair in a forthcoming book [Siliconnections, McGraw-Hill, 1986] . In the meantime, I hope this preliminary account has helped set the record straight.
