CROMEMCO Computing [Parkison, Johnson, Borrill, others]

Third Floor, side. Started January 2001, updated April 2001, Oct. 2003, Jan 2004.

CROMEMCO Display Cabinet

Cromenco Model Z-3 [Parkison], with its `Glass teletype' Zenith display on top; An original Cromemco 3102 display is now next to it [donated by Andy Berg (groovelists at of Boston]. In front of it is a Cromemco Cyclops Video Camera, one of the original products of Cromemco [Roger Melen].


CROMEMCO was founded in 1974 by two Stanford Students, Roger Melen and Harry Garland, who lived and worked in the CROthers MEMorial Hall dormitory on the Stanford campus. Eventually CROMEMCO needed space to receive visitors, and moved out of the dorm to Mountain View, and incorporated in 1976.

Early products were a video camera and supporting cards for the S-100 bus, as used by the Altair and IMSAI computers, for instance an Analog-to-Digital converter -- soon to be in this exhibit. The sensor for that camera was a memory chip without its plastic covering case, which was sensitive to light. Such a camera, CYCLOPS and its interface is now shown in the display (Last photo below).

CROMEMCO's Computer products were based on the 1976 ZILOG-80 integrated 8-bit processor chip, a development of the architecture of the Intel-8008.

    Zilog produced the Zilog Z80 in July 1976 (also used by Sinclair in the ZX-80, ZX-81 and ZX Spectrum computers) and later the Zilog Z8000.
    Zilog was founded in 1974 by Franco Zefferelli and became a wholly owned subsidiary of Exxon Corp. by 1980. The company's management and employees purchased Zilog back from Exxon in 1989. Zilog became a publicly-held company in February, 1991. In March of 1998, Zilog was privatised, as a result of the merger and recapitalisation transaction by Texas Pacific Group (TPG).
    Zilog now (1998) produces a range of 8-bit microcontrollers, 8-, 16- and 32-bit microprocessors, and digital signal processors, covering the home entertainment, communications, and embedded systems markets.
    Address: 910 East Hamilton Avenue, Suite 110, Campbell, CA 95008, USA.

click click

CROMEMCO computers were extremely sturdy, encased in cast metal frame. The Cromemco Z-2 (not on dsplay) was built to address the needs of the industrial market, but did one better! The Z-2 line was the first commercially marketed microcomputer certified for use by the U.S. Navy for use aboard ships without major modification.
The unit featured a 22 slot S-100 bus, reinforced card cage with retaining bar, and a hefty power supply. Later versions, as the System Three on display, provided for the mounting of two 5 1/4 inch disk devices (floppies, or later a floppy and a hard drive) into the front panel.

In 1979 CROMEMCO produced the first multi-user operating system for a micro-computer: CROMIX, based on UNIX.

It appears that Steve Wozniak and Steve Jobs visited CROMEMCO in 1979(?), but were not welcomed by Garland.

In 1981 CROMEMCO produced systems using the Motorola 16-bit MC68000 processors.

In 1982 CROMEMCO Deutschland was founded to serve the European market out of Frankfurt am Main, Germany.

At is peak (~1983) CROMEMCO had approx. 500 employees and about $55 Million/year in revenue. By that time it had become the preferred supplier for micro-computers to the People's Republic of China. CROMEMCO was always a private corporation, wholly owned by Roger Melen and Harry Garland; it was never publicly traded.

In 1987 CROMEMCO was bought by Dynatech. Dynatech was a company that owned several electronics subsidiaries: Colorgraphics Weather Systems, Fuzz Buster, etc. The weather display systems produced by Dynatech, and used by many TV channels, depended on CROMEMCO Computers.

A successor company, structured from the European branch in 1990, CROMEMCO AG, operates today in Germany, Swizerland, and other European countries, with a.o., multi-processor, neural net, and document management products.

CROMEMCO C3 and Software Manuals

The C3 machine exhibited here was purchased in 1976(?) by Professor Ken Colby and used in the UCLA Psychiatry department, mainly for voice generation though a Votrax speech synthesizer, built by Jim Graham at the EE dept. of UCLA. Ken Colby had been on the faculty at Stanford, and participated in the AI Lab (SAIL) (see Ken's photo here).
The primary project which used the Votrax and Cromemco was for Intelligent Speech Prosthesis. Here the CROMEMCO C3 was used as the development platform for software to be downloaded onto a far-before-its-time notebook computer (built at UCLA), which was then carried by a person who could not speak. Another project was Speaking Parry, which allowed PARRY, a simulation of a paranoid patient, to give spoken as well as written responses to questions. [Donation courtesy of Roger Parkison, Stanford PhD 80].
(Ken Colby in 1968? at SAIL - photo courtesy of Bruce Baumgart).

CROMEMCO Poster, Shugart 8" floppy Drive (XEROX, 1970), Melen and Garland book

In the back, CROMEMCO advertisement posters, 1976 and 1978. Harry Garland, one of the founders of CROMEMCO, is standing in the 1978 poster. [Obtained from Ray Borrill, Bloomington IN, CROMEMCO dealer #7, by Gio Wiederhold.]

Roger Melen and Harry Garland had worked together earlier as students in Stanford's Electrical Engineering Department. They jointly published a book: Understanding IC Operational Amplifiers in 1971, displayed here. [Obtained from Ray Borrill, by Gio Wiederhold].

Also, in 1971, Harry Garland and Roger Melen received Altair number 0002. They had proposed in December to attach their Cyclops camera to the Altair, for use as a security camera.

Extract of Transcript from PBS Special "Triumph of the Nerds:

At the Bay Area Computer Club: "... The very uselessness of the Altair is what drove the hobbyists together. Roger Melen and Harry Garland started an early computer company. They came here to meet others and to figure out just what the heck could be done with this new toy -- a solution in search of a problem. There's no keyboard that I can see. The Altair was tedious to use. At first, the only way that data and instructions could be given to the computer was by flipping switches. Take something trivial like 2+2. Each 2 needed eight different switches to be flipped, then a ninth switch was used to load them all. 'And' required another nine switches. The answer 4 was if the third light from the left turned on. Eureka!
ROGER MELEN: So if you had a program that was a hundred bytes long you had to go this procedure a hundred times to load that in the memory.
Harry Garland: It took a long time.
BOB: I bet it did and what happened if you lost power or if you lost your way in the middle?
The Altair may have been frustrating, but it drove the nerds to experiment, finding real uses for the useless box, turning it from a curiosity to a computer.
LEE FELSENSTEIN: Steve Dumpier set up an Altair, then laboriously keyed a program into it. Somebody knocked a plug out of the wall and he had to do that all over again but nobody knew what this was about. After all, was it just going to sit and flash its lights? No.
ROGER MELEN: You put a little transistor radio next to the Altair and he, by manipulating the length of loops in the sofware - could play tunes.
LEE FELSENSTEIN: The radio began playing 'Fool on the Hill'....Da da da, da da da....and the tinny little tunes that you could tell were coming from the noise that the computer was generated being picked up by the radio. Everybody rose and applauded. I proposed that he receive the stripped Philips Screw Award for finding a use for something previously thought useless. But I think everybody was too busy applauding to even hear me.
ROGER MELEN: It was a very exciting thing, it was probably the first thing the Altair actually did.
Turning the Altair into a useful tool required a programming language so users could type their programs in rather than flipping switches. What it needed was a version of some big computer language like BASIC, only modified for the PC. This was called a BASIC interpreter, but it didn't yet exist because the experts all thought that not even BASIC was basic enough to fit inside the tiny Altair memory. Yet again the experts were wrong. Here comes the guy who solved the problem. Twenty years after finishing the first microcomputer BASIC, Paul Allen is returning to Albuquerque for a celebration of that event -- this time with his $15 million jet and three foot red carpet. At a time when I was killing brain cells, this guy was founding an empire. He has come to eat rubber chicken in honor of the Altair's 20th anniversary.
SPEAKER: I'd like to introduce to you - Paul Allen.
Allen co-founded Microsoft with his younger buddy from high school -- Bill Gates.
PAUL ALLEN: One day in Boston, I was in Harvard Square I saw a cover of Popular Electronics with this thing that looked like what I had been imagining, and so I grabbed it off the shelf, I looked at it and I bought it and I ran back to Bill's dorm, and I think he was probably playing poker that night and usually losing money at that point. One of the few times when that's been the case.
BILL GATES: Paul showed that to me and then okay, here was a company that would be needing software.
PAUL ALLEN: And he said OK we gotta call these guys up and see if this thing's for real.


Eight-inch Floppy Drive

The floppy drives, shown here, were first used to reload the micro-program into the cheap, but volatile processor memory of the IBM 360-20 in 1964. The floppy disks turned out to withstand a much higher duty cycle than originally envisaged and replaced small tapes (as seen in the home computer display) as an economical and removable medium. Such floppies were also used at Stanford in 1981 for the `Alphatext' photo typesetter, driven by Prof. Donald Knuth's TEX typesetting system.

Previous display Next display (Home computers) Prior floor (2) Next floor (4) All the Way Back