Last edit Aug 14 2011. This page (c) 2011 Herb Johnson except for quotes which are copyright by their authors.
Ed Roberts, the cofounder of MITS Inc. in 1969, and the designer of the MITS Altair 8800 computer first sold by MITS early in 1975, died April 1 2010. The importance of his early product was noted, in obituaries published in leading journals and newspapers.
The Altair is important because its expandable design was immediately adopted by dozens, and later well over 100, start-up and small microcomputer companies who produced "S-100" plug-compatible products and computers. Early Altair-compatibles ran the CP/M operating system from Digital Research Inc, which ran on the majority of 8080 and Z80 processor systems of the 1970's. Many of those manufacturers, and supporting programming and reselling companies and individuals, went on to support the IBM PC, as well as supporting those older systems and he 1980's. Among them was "Micro-soft" who produced Altair BASIC, Bill Gates and Paul Allen's first product for that company.
As "Micro Instrumentation and Telemetry Systems", MITS in the early 1970's produced everything electronic from clocks to calculators, to model rocket radios. But it was the January 1975 introduction of the Altair, and subsequent 8080 based microcomputers, that turned MITS into a multi-million dollar company. Early on, the 100-pin interconnection "Altair bus" became the springboard for bus-compatible "S-100" products by start-up companies, as part of the "microcomputer revolution" that put computers into offices and factories and onto desktops - several years before the IBM PC of 1981, and the Apple Mac of 1984. Meanwhile, Roberts sold MITS to drive maker Pertec; he entered medical school and became an M.D. in general practice, and then retired in Atlanta some years ago.
What follows the links below, are some extended exerpts from the Usenet CP/M support discussion group "comp.os.cpm". Thirty-five years on, members posted at length about their experiences with the Altair and their memories of the era and of Ed Roberts. One reason I created this page was to provide better access to what others have said in newsgroup posts, or subsequent emails to me, about Ed Roberts and MITS and his Altair. There are also links to my S-100 Web pages; I've supported S-100 on the Web for at least 13 years, and for many years prior to that. People build and rebuild S-100 systems, including the Altair when they can get one, to the present day.
- Herb Johnson
Here's a link to the front-page New York Time obituary of Ed Roberts "H. Edward Roberts, PC Pioneer, Dies at 68" for April 2 2010. Here's a snapshot of the cover. There's other obits and editorials in publications including The Huffington Post, the UK Register, and more. Wired magazine titles their obit as "Creator of First Personal Computer Dies". As the discussion below argues, it's more reasonable to describe Robert's Altair as the first personal computer to gain widespread acceptance and support.
This June 2010 editorial article by Joseph Desposito of Electronic Design magazine, who worked for Popular Electronics five years after the Altair article, acknowledges the Altair as "pav[ing] the way for other personal computers".
Jon Titus, developer of the Mark-8 8008 microcomputer offered months before the MITS Altair, and to the present day a tech columnist and digital consultant, comments on the Altair and MITS in an EDN (Electronic Design News) electronic trade magazine blog. The entry is titled "Ed Roberts, PC Pioneer (1941--2010)" for April 5, 2010.
Bill Gates and Paul Allen, who wrote Micro-Soft BASIC for the Altair as their first product, issued a joint statement on Gates' Web site.
MITS pages on my site:
a discussion about the Altair and other "first computers", with the Web site owner of a very good review of those computers. ALso, a Web page of frequent questions about the MITS Altair and its S-100 bus. I will add some notes about Dr. Robert's death and link it to this Web page. Finally, this Web link is to the home page of my collection of S-100 Web pages;that page lists over 140 companies that produced S-100 products in the 1970's and 80's.
- Herb Johnson
Allison Parent, an analog, digital and RF Engineer whose work began in the 1970's, said this:
There was a whole lot that went down at that time; while the Altair was important, it was a mere moment in a fast moving era when moments were significant.
Just think of the time from the Mark-8 to Apple, TRS-80, SWTP 6800 and others. I started college in fall of '72; by 1975 two generations of micros (8008 and 8080) occurred and small computers went from about 4 times the price of a new car to 1/4 the price of a new car. That was as big a change as from my second year in high school, where communications gear was predominately tubes with some transistors. when I entered college, IC's and solid state were the norm for everything but high power. As a result I have an appreciation of digital technology, from when RTL gates were first available, to current IC tech and all the things around it.
The whole process was building up from the advent of TTL and "cheap memory" [1101s in 1972 were about $6 or about two hours pay] but actually cheap compared to core. By 1974 2102s were about the same price, 4 times the capacity and easier to use. Without memory the rest of the computer was doable then but memory was tough and seen as required. Everything that went before, even Mircal, were all memory starved or where it existed it was painfully costly.
In essence, by the time of the introduction of the MITS Altair, many factors reached critical mass and any one person could have integrate all the parts in a box - and Ed Roberts did. He wasn't magic, but he was a business man who felt it was all there, waiting to be done.
As for the issue of the "first" personal computer, first and breakthrough are not always the same thing. Just like Tandy's TRS-80 wasn't the first but they were the first to sell more personal computers in one year, than IBM had ever made to that date. First year's sales were estimated at 250,000.
Meanwhile, before the MITS Altair, total microcomputer sales were likely tens of thousands total, including development hardware. In this case 8008 chips, as used in traffic light controllers and such, don't count. After the 8080 based Altair, I'll bet that 10,000 microcomputers per month was the average for the year for that first year, including all the other entries.
To put the $1000 in 1975 in perspective. a new car at that time was about $3200 for middle of the road and something really fancy was maybe $4400. So $1000 was still serious dollars but the next best choice was a base PDP-8/E system for maybe $10K. Also 4K of core then was well over $2000.
To put that in further perspective: For example DEC's PDP-5 sold only a few hundred, PDP-8 was for its lifetime 35,000 (not including the CMOS DECmate word processors). PDP-11 didn't crack 50,000 sales until 1980 (mostly due to LSI-11). Yet, we have MITS, SWTP, Apple, NorthStar, Comodore, Tandy TRS-80, and many others that collectively from Jan 1975 to 1980 put many more than 400 to 500 thousand machines into the market.
Some have said the MITS Altair was a "hobby" machine or a "commercial" machine, not a "personal" computer. Well, I was there and actually bounght one. It was the first computer to be owned by many hobbyists, as well as other people not directly involved with companies using computers or in the computer business.
The MITS Altair system I bought had an 8800 cpu, case, 4KB static ram, 88 ACR serial card, and 88PIO parallel card was $1,234 1974 dollars. That's the total of a check I'd written on December 16, 1974 to MITS Inc. I also bought from SWTP a SWTP CT1024 video terminal board and a matching keyboard .
At the time I was still in College getting my EE degree and making about $7800 (at that time that was good pay for a student) working in commercial mobile communications field. (FYI I had designed a 8008 machine for the company a year earlier).
In short I was both a hobbyist, and in the industry. And my PC at the time was a Cincinatti Milacron CP2100 minicomputer - the company owned it but I was the sole user.
The Altair or S-100 bus had varients due to lack of clear specification and some vendors trying to make theirs vendor specific. Additionally it suffered from being the 8080 architecture. But it would grow into a bus that would see many cpus and even things like multimaster (multiple intelligent controlers on the same bus)- none of which were foreseen. However it was not an untenable situation at the time, and those vendors that chose to be incompatable by design or error either disappeared or corrected their designs.
It wasn't perfect but it was serviceable.
As for the many prior computers built by individuals: A lot of people did that, it was apparenly the norm for universities and colleges where the then current cost of a computer was out of reach for a department level activity.
And, not just individuals built computers one at a time. Start with the LINC computer in 1963-4 as a machine that could be made with purchaseable modules. A link to LINC is this site:
BASIC, from MITS [as written by Microsoft], and more importantly as freely published software like TinyBasic, and later LLL Basic, Processor Technology's FOCAL, and their 5K basic - all had a significant impact. However MITS also sold the Programming Package (editor, assembler debugger) and arguably that was a precursor tool to the next level, a growing base of available software. Also other vendors of software like PT also had software development tools. There was a lot of building going on but always first was to obtain a platform to run it on. That would repeat itself for every generation of machine, past and future.
[Some] assert that without software it would be an anchor. Entirely correct save for some saw software as the potential market. It was the volume of 8080 or S100 machines that triggered that need for software and also provided the platform to create it. MITS (or more correctly Gates/Allen) created only a small piece of that software base. Basic was a tool not the end game. The key being people didn't want to run Basic they wanted to run Basic programs that would do games and useful work. Since Basic had existed on other older machines there was a base of programs written in Basic that were both available and to the user important.
I'd argue that Altair and other machines before (for example: PDP-8) like it were pivot points where hardware begat software that lead to further software becuase the platform became ubiquitous. This in itself would be repeated in that every new generation of hardware would trigger further software development.
The computer industry was once described to me as the most competitive and incestuous of all. Seems often the people competing were often using their competitors' tools to create their own better mouse trap. The world of hardware to run all that on, was similar.
- Allison Parent
Like Allison, Dave has been a software and hardware developer for decades. Here's his take on MITS and Ed Roberts:
FWIW, this is what I wrote about the MITS Altair 8800, in the introduction on my Web site to my Altair simulation files:
This package is my attempt to preserve a detailed record of my Altair 8800 computer. For those who are not familiar with the name "Altair", the Altair 8800 is considered to be the first "personal" computer - Although this title has been the subject of some debate and controversy (Several other small and functional computers did exist prior to the Altair), I believe the title is deserved because: - It was the first computer of substantial capability which was marketed to hobbiests and small business, and was affordable by people of modest means. For this reason: - It was the first computer to be owned by many hobbiests and other people not directly involved with companies using computers or in the computer business. - It was openly documented and expandable. With the release of the Altair, MITS created the "Roberts" (S-100) bus, which was the industry standard small system bus until the IBM PC appeared. Before the Altair, expansion hardware for a computer was available only from the machines manufacturer. The Altair opened up the idea that "off the shelf" components could be made which would work in any manufacturers machine. Many small companies were formed because this bus opened up a new market in computer add-on's.
I have been involved in the computer industry since the mid 1970's, and have always had a keen interest in "personal" machines. In the early years I built several small homebrew designs, and knew a few other people doing the same using such early microprocessor chips as the Intel 8008 and the RCA 1802. I know of at least one individual who built a functional CPU from TTL logic! A clear memory of mine is that the Altair was the first commercial machine I ever saw in someone's living room.
prior to the Altair, purchasing a computer was beyond the budget of most people - nobody ever said the Altair was the first computer that could be purchased.
I don't have an actual price list handy, however the Feb-75 issue of PE which has the last 1/2 of the Altair article mentions that you can expand the 256-word memory card included with the Altair with up to 3 more 256-word chunks at $34/each). A 4K dynamic card is listed at $198. A MITS advertisement in the same issue lists the Altair system at $750 assembled, and $495 in kit form, with a $100 discount if you ordered before March 1/75 - So if you were willing to put it together yourself (which most hobbiests did), you could have an Altair with 8K(+256) bytes of memory for under $1000 - not small-chance in 1975 but low enough that a lot of people jumped into it.
My Altair started with 256 bytes of memory - pretty much only enough to play with it from the front panel - but it "grew" over the years :-) I had a number of homebuilt cards, and every now and then I'd find a spot, stick on another pair of sockets and put in two more 2114's (1K of memory). Eventually I got a 16K card, and later a 64K card (which is still in it).
The S-100 bus was originally called the "Altair" or "Roberts" bus - the S-100 name didn't get used until later (when other companies didn't want to include reference to MITS or Altair in their products).
I don't know how much "invention" there was (from looking at the bus diagrams, it seems he mostly just brought each signal on his CPU card down to the closest pin) ... but Ed Roberts certainly created the bus. Equally important to the connector is the pinout - Ed put the two together and documented a bus that he incorporated into his computer, and that others could make peripheral cards for. The key word in that sentence is "documented" - this enabled people to build and sell cards which would work in any Altair (ok - if you "were there" you know that it wasn't quite that simple... but for the most part it's true).
I had a friend who built a machine with a TTL CPU - and I built several small machines prior to getting my Altair ... but we were never able to buy peripheral cards or software for any of those machines - for some reason nobody ever thought it would be economically viable to producing add-on's for our one-of-a-kind custom designed systems that they had never heard about.
Anyway - enough - I for the most part agree with the view [of the Altair as a "first" personal computer], but I also don't think of that as a "cut and dried" statement - Ed and his Altair played a significant role in the establishment of the early personal computer industry - how's that?
H. Edward Roberts, the creator of the Altair says its even simpler than that: "We coined the phrase Personal Computer and it was first applied to the Altair, i.e., by definition the first personal computer." ... "The beginning of the personal computer industry started without question at MITS with the Altair."
That quote from Ed Roberts is taken directly from correspondance with him. (If I hunt around I can probably find the letter). I'm very sorry to hear of his passing - Ed did a lot for the early personal computer industry but seems to be largely unknown by anyone younger than 40 these days.
Barry Watzman was a Heath engineer, and later a Product Line Director for all computers and systems software at both Heathkit and Zenith Data systems. He worked on the Heath/Zenith Z-100 series of S-100 computers. He was a designer during the period of the MITS Altair and knew or worked with many other microcomputer designers.
...has been discussed so many times before. [Here's] a few comments:
First, the statement made is absolutely true. The statement was:
"the Altair is widely credited as the first personal computer"
Note that this is not the same as saying that the Altair WAS, in fact, the first personal computer. That statement would be debatable. But saying that the Altair "is WIDELY CREDITED as the first personal computer" is beyond debate. It is. Period. [Note that "widely credited" does not even require that most people consider it to have been the first personal computer. Only that "a lot" of people do. "Widely" does not mean "primarily" or "mostly" or "predominantly" or "usually". It just means "often" or "frequently".]
Second, as has been discussed many times, it all comes down to the definition of what constitutes a "personal computer". And that itself is a question to which you can get hundreds of answers. - Barry
Herb Johnson was an EE student when the MITS Altair was announced. He designed or installed Intel based systems for many years. For decades, he's offered support and documentation for S-100 systems, through today on his Web site.
In any event, the TOPIC which started this discusion was the sad loss of the developer of the MITS Altair 8800, and a founder of MITS itself, namely Dr. Ed Roberts.
There should be no doubt that the MITS Altair was the beginning of a decade of production, and two decades of use, of a series of microcomputers which used a bus initially based on his designs. By my count, over 140 companies produced S-100 compatible products or systems.
Furthermore, the individuals and companies which produced those systems, not only had an impact in the era; many of them went on to create designs and products and start companies which supported subsequent generations of personal computers. Bill Gates and Paul Allen are two of many, many such people.
Those collective facts distinguish the MITS Altair 8800 from many other early computers which in general did NOT enjoy much commercial success, did NOT lead to either plug-compatible products, and only involved a handful of persons.
One cannot deny there were computers which were earlier and which were "personal" in many ways; they and their designers deserve credit. But one also cannot deny that the MITS Altair 8800 had an impact, and that the loss of the founder of MITS this month is an occasion where that impact is worth remembering. There's some comfort in reading in the comp.os.cpm newsgroup, some of the tributes to Roberts and to his era. Those remarks acknowledge his contribution to early personal computing in the 1970's. Much of that decade of microcomputing is simply ignored in many "historic" accounts today.
- Herb Johnson
Copyright © 2011 Herb Johnson