VCF-E 9.1 Exhibit, early 1970's microcomputing - COSMAC and Intel

Last produced Dec 3 2016, updated NOv 19 2024 Edited by Herb Johnson, (c) Herb Johnson Contact Herb at www.retrotechnology.com, an email address is on that page.

Index

Before the Altair and ELF...
Intel 4004, 4040, 8008
RCA COSMAC 1802 related and current 1802 products
The context of the exhibit, and consequences
contact

Before the Altair and ELF...

In the early 1970's, hobby electronics was about amateur radio, audio kits, and test instruments. They were made from digital logic chips - AND, OR, flip-flops, counters - as "wired logic" for specific tasks. Early microprocessors were seen as logic replacements, but too expensive for hobbyists. Engineers designed them into equipment with expensive development systems. Only a few visionaries, not big corporations, looked ahead at inexpensive microprocessor based computers for personal use. This exhibit shows some of the earliest use of microprocessors, from two early developers - Intel and RCA.

My exhibit shows some of the earliest Intel and RCA microprocessor based development tools, and products with those processors. Also, hobby electronics publications where early micro-computers were introduced. A chart of 1975 prices shows income and common costs in the era versus today, and a PolyPaks ad of 1975 shows costs of digital parts.

The exhibit was shown at the Vintage Computer Festival - East 9.1 held April 4-6 2014, at the Infoage Center in Wall, NJ USA, operated by the MARCH vintage computer club at Infoage. Thirty five exhibitors, 12 workshops, a dozen talks, and a vendor/consignment area. Hundreds attended. Here's a brief Web page about the event and exhibitors.

For more interpretation, see "The context of the exhibit, and consequences". For discussion of VCF-E exhibits after the event, see the Yahoo discussion group for the MARCH organization and posts under the topic "VCF thoughts" for 2014.

By 2020, the VCF-East Festival at InfoAge and the MARCH museum, were operated by the Vintage Computing Federation, Inc. and the discussion group moved to a private listserver with public access, called "vcf-midatlantic". Check with the vcfed.org Web site for further information. - Herb

I do a lot about restoring and preserving computers of the 70's and 80's.

Email me: Herb Johnson, retrotechnology.com

Intel 4004, 4040, 8008

The Intel 4004 was developed about 1972-3, and sold by Intel as a "programed logic" replacment for individual small-scale logic chips or "discrete logic" designs. The Intel 8008 was an 8-bit and more developed version. The 4040 was an advanced version of the 4004 chipset, produced about the time of the 8080. The early history of these microprocessors in development is described in detail elsewhere. What is not often described, is how these were actually used. This exhibit showed the 4004 and 4040 in use, and one 8008 card.

Vacuum Fluorescent Display w/4004

"Digital Electronics" designed a 4004 board to run a simple program to exercise a display. The 4004 board likely dates from 1976, the display from 1975 and may be from the same company. The character scheme is called "starburst" or 14-segment and produces readable text. These displays, and the 4004, were used in commercial scales, point-of-sale "cash" registers, gas pumps, and so on.

Intel 4004 driven fluorscent display control card, Digital Electronics brand, 1976
closeup of Intel 4004 chipset
fluorscent starburst display. Message scrolls left/right, blinks, etc.

Dranetz 606 AC line monitor w/4040

For decades, Dranetz has produced a series of AC power monitoring equipment to monitor and report on line faults, surges and spikes. They adopted the Intel 4040 for this 606 model, but also used other microprocessors when they offered price or performance benefits. The operating units have the same 4040 CPU board as the one on display.

Dranetz 660-3 system
Dranetz 660 control board with Intel 4040, early 1980 construction
Dranetz 660 control board on display, mid-1979 construction

Intel Intellec MCS 4/40 4040 development system

In 1973?, Intel introduced the MCS-4 development system for their 4004 processor. It consists of a collection of 100-pin boards with cross-connects (not a common "bus") to provide RAM, ROM, I/O, PROM programming and a control panel. They upgraded the CPU board to a 4040 probably in 1975 and called it a 4/40 or "4 mod 40".

My 4/40 system was obtained several years ago, but only restored to use in 2014 thanks to Kyle Owen's encouragements. This system is mostly operational and runs a terminal base ROM monitor to read and dump memory to paper tape. Assembly is done by hand, or with cross-assemblers in FORTRAN run on a mainframe. Thanks to Kyle Owen for supporting my repairs of this system. - Herb Johnson

Working Intel Intellec 4/40 development system
Intel 4/40 4040 CPU board 1974 chipset
current-loop to simple RS-232 interface, 1200 baud
My Web page on my 4/40

An additional working 4/40 was displayed by Kyle Owen, in the adjacent exhibit by the Atlanta Historical Computer Society. Thanks to Kyle and Flash Corliss for monitoring my exhibit.

Pro-Log Corporation 4004, 4040

Pro-Log introduced microprocessor development tools and control products in 1973 with the Intel 4004. Shown are Pro-Log 4040 boards and tools, sold for logic control at a biology lab at a university. The lab bought the ProLog series 90 model 900 PROM programmer to store the hand-assembled 4040 control program into 256-byte 1702 PROMS. There's a Pro-Log single-board 4040, and a three-board 4040-ROM-I/O set. More info is on this linked Web page. Prolog later created the STDBus standard with similar boards. STDbus became a standard and ProLog and others produced a variety of boards for many microprocessors. I offer some STDbus cards and docs.

A ProLog 4040-based controller board set in card cage
ProLog model 4410, with 4040 CPU, a single-board 4040. missing are 4002-1 and 4002-2 RAM chips

The ProLog series 90 model 900 PROM programmer, has a 4004 microprocessor and 1702 PROMs. Also displayed is a 4040 controller for a ProLog 980 PROM programmer which has 2716 PROMs.

Prolog model 900 PROM programmer, uses Intel 4004
Prolog 90 M900 PROM programmer brochure PDF
Prolog M980 PROM programmer CPU board with Intel 4040 probably made 1980

Beehive brand SuperBee terminal - Intel 8008

Beehive brand, Super Bee terminal CPU board with Intel 8008: late 1973
Thanks to Rick Bensene, of The Old Calculator Museum at oldcalculatormuseum.com for identifying this in Dec 2016. In Mar 2020, I found a Super Bee manual online at oclc.org and confirmed the details of this board.- Herb

RCA COSMAC 1802 related and current 1802 products

The COSMAC microprocessor was first developed as a conceptual design in discrete (small scale) TTL logic in the early 1970's by Joseph Weisbecker and several of his RCA colleagues. A two-chip version was developed by RCA's Solid State Division in 1973-4, and a single chip version soon thereafter. I cover early COSMAC development on another Web page. This exhibit shows some of the earliest use of the COSMAC, by RCA in various development tools, in commercial use; and as the latest of hobby microcomputer "tribute" kits based on the early COSMAC ELF.

Chrysler Spark Control System w/COSMAC:

In Nov 2024 I expanded this description of the Lean-Burn system to this linked Web page which details the Chrysler products, analog and COSMAC digital, and confirms the RCA COSMAC was an embedded auto emissions controller in the late 1970's. - Herb

Developers of the RCA COSMAC looked for products for their microcontroller, which was suited for hostile enviroments. Chrysler was looking for cheaper ways to reduce emissions in their cars to meet 1976 EPA standards. The COSMAC was sold as a "programmable logic device" where a one-chip change supported each car model and engine. The "Lean-Burn" system controlled sparkplug firing based on engine timing, throttle position, air temperature and other values.

This product is one of the earliest production, real-time, embedded microcontrollers. It got a bad reputation because it was mounted on the air cleaner and suffered from heat, vibrations and resulting poor connections. Other COSMACS were used in 1970's amateur and research satellites, and the 1990's NASA/JPL Galileo spaceprobe to Jupiter.

Chrysler Spark Control System first produced in 1976
SCC functional diagram of digital control of ignition spark timing
SCC board showing 1802 This is a 1982 dated chipset
analog and discrete component version of SCC

The COSMAC in space

COSMAC Microkit

This 1974-5 product of RCA was a COSMAC development system based on a number of small boards in a bus-like card cage, for expansion by engineers to develop COSMAC designs. Note the two-chip CPU board with "house numbers" instead of 1801 labels. Expanded versions supported a floppy disk operating system; the earliest version was ROM and Teletype based for paper-tape files; later it supported cassete-tape files. Loaned courtesy of Bill Degnan Some of these photos are his.

pre-1801 Microkit from 1974-5
Microkit top view showing card cage
Microkit CPU chipset 1845, 8682, 8335, 2192
Microkit ROMs
See my Web page about the MicroKit ROM monitor.

COSMAC MicroTutor

This was a 1974-5 low-cost kit offered by RCA for COSMAC microprocessor development. There's no ROM, only RAM and toggle-in programming. Joseph Weisbecker,COSMAC processor developer (and game designer) tried to make microcomputers assessable to hobbyists and educators, as an exercise in programming, logic, games and puzzles. A later version was the COSMAC ELF, published in Popular Electronics in Aug 1976. Loaned for my VCF-E exhibit courtesy of The Sarnoff Collection, The College of New Jersey

RCA COSMAC Microtutor kit
My Web page about the earliest COSMAC processors.
Contact me about a copy of the MicroTutor manual.

COSMAC Elf and ELF2K, Atlanta

Courtesy of Kyle Owen and Flash Corliss of the Atlanta based Vintage Computer Festival - Southeast, I shared a table of COSMAC items with them. Photographed are two of their items. In the back is a replica of the COSMAC ELF as displayed in the 1976 Popular Electronics article. In the front of the photo, is Spare Time Gizmo's ELF2K product of several years ago, with added solid-state drive card to run ELFOS on the ELF2K.

Lee Hart and the COSMAC "Membership Card"

Lee Hart was a digital engineer in the 1970's when a COSMAC MicroKit and MicroTutor were shown to him by a RCA "field rep". His big-company bosses wouldn't buy COSMACs. But the low-power noise-resistant CMOS product held his interest. In 1978 he started his own digital design company, to make a COSMAC controller.

Four decades later, COSMAC enthusiasts (and me) knudged him to produce a modern ELF (and Microtutor) called the Membership Card. Five years later, they are still available today. The model on display is an early version with a ROM/RAM upgrade: the latest version has both.

For fun, Lee Hart designed a simpler COSMAC board, just to randomly run a program and display "results" as an LED face - called Face Card.

When Lee saw my "microprocessors as logic" exhibit, he wired up in a day, a pure-logic "nametag" display board. NO PROCESSOR! just a ROM and counter and clock. It simply displays 7 bits of ROM as 7-segment letters.

Membership Card kit - modern version of 1976 COSMAC 1802 ELF kit
Order the Membership card from Lee Hart at TMSI
Support Web pages for 1802 Membership card product

Face Card - 1802 powered, simple blinky-light board, executes random 1802 instructions Photo by Lee Hart
COSMAC 1802 executes random instructions to create facial expressions with LEDs
available as simple to assemble kit

nametag - simple design - ROM, counter, clock, display
Designed & built by Lee Hart in honor of VCF-E exhibit. If you want one, let me know!

Books and magazines of the early 1970's

These books and magazines represent both "discrete logic" hobby activities of the early 1970's, and the earliest microcomputing activities. They were not displayed at the actual exhibit, for lack of space and because visitors had no time to look at them. I'll provide portions of these documents on this Web site, later.

73 Magazine for Aug 1975 selected pages
73 Mag for Apr 1975 selected pages
Radio Electronics for Apr 1975 PolyPaks ad for Aug 1975
The Bugbook I - Logic and Memory Experiments, 1975
TTL Cookbook, Don Lancaster, 1976
An Introduction to Microcomputers, Adam Osborne, 1976

Prices in 1975

$1 in 1975 = $4 in 2014

a 1975 Chevy Nova sedan? about $3K-$4K
gasoline? 30 to 50 cents a gallon
a color TV? maybe $750
a term at state college, room & food? $5-600
Jan 1975 average US home price? about $40K
median household income? about $10K year
Engineering starting salaries? $12K-18K year

8080 processor - $175-$250
8008 processor - $30-$50
1K bit RAM - $4-$5
256 byte PROM - $20
TTL chips - 2 to 5 for $1
typical logic kits - $hundreds
8008 kits - $400
* MITS Altair kit - $439 *

The context of the exhibit, and consequences

What was this exhibit about?

The point of my exhibit was to show the limits and perceptions of the earliest use (early 1970's) of microprocessors, namely as "programmed logic". That's how Intel 4004's and COSMAC 1802 were sold, to engineers and manufacturers, to replace mechanical logic or process-control equipment. A 40-year-old power meter with a 4040, still printed warnings when the line voltage dropped, when another exhibitor ran his mark-sense card reader (with 1/4 horsepower AC motor. So it was an "interactive" and "networked" exhibit. ;)

I chose the processors from the earliest available.

The COSMAC has always drawn interest from people who want to work with binary-level programming and gaming. Generations of binary-level ELF computers were, and are, in hobby production. And a subset of people who want to put modern "tech" on vintage micros, also put them on COSMACs. I showed current versions of the ELF too.

The Intel 4-bit processors are of "classic" interest, because they led to the 8-bit 8008 and 8080 (and Z80), and then to Intel's other processors. "Firsts" of popular products attract interest.

Why was this exhibit relevant to anyone else?

It answers the question "what did people do with that Altair 8800 after they built it?", a topic disucssed by many in 21st century. As the Altair had no graphic displays, no I/O devices, what could it do? A common demonstration, is to play tunes on an AM radio, using program loops to generate RF signals. Clever, attractive (like a bumble bee) but useless. The same question applies to owners of the first KIMs and other single-board micros. It still applies to new owners of Arduinos and Raspberry Pi's today.

The direct answer - binary interfaces to equipment available in the 1970's - was was not in my exhibit. I ran out of space, time and effort. Hobby devices interfaced at the time included Teletypes, older terminals and printers, and lights and relays to run other things. Cassette recorders became mass storage devices via audio tones. TV's became displays too; there was a random-logic TV "typewriter" in 1973.

Howe were these early microprocessor applications done, then?

You built other devices for them to control and measure, and you wrote the programs to do that. You wrote the programs in binary, or hand-assembled code. Or you could "play" simple games, or treat programming as a game. The MicroTutor manual on display had some textual examples, toggle-in programs. Few people could program a prom in 1975. But with storage, on paper tape or cassettes, you could store programs and data. With a terminal and more memory, you could run like a then-conventional computer could, but without punched cards and waiting a day for results!

ABut how was it done first? By the engineers and techs, who had these tools, on their company's budgets or in their college labs. They had the earliest microprocessors and the austere development environments for them. Or none! They programmed in binary! They hand-assembled! Stored data in ROMs and punched tape, or just toggled it in.

During the exhibit, a have a photo of one of the exhibitors, using my 1702 PROM programmer, to fix ROMs for his 8080 single-board on exhibit. In hex, byte by byte. Gettin' it done. This was preplanned and not accidental, but it demonstrates that these techniques are not dead. He used a cross assembler on a laptop, but that's OK, in the era one may have used a mainframe running a crossassembler too.

Were there other exhibits of this vintage? with similar technology?

Many exhibits were of video gaming systems, or computers with such features. An argument for these was, this attracts public interest. No doubt, visual/audio gaming attracts attention. Go figure! Lights and sounds! Touch this, and that happens. No surprise that "people" find that attractive. And cats.

My exhibit had one game with no rules: a device that turned random COSMAC 1802 instructions into a face display made from LEDs. It got a share of attention. Most people attracted to my exhibit were of two groups. 60 year olds who used it first, and 20 year olds who like restoring and playing with really old computing technology.

I actually thought that through, asked my colleagues about it. I decided that amusing my age-cohorts was sufficient. If others "got it", so much the better. But in fact, I ended up with multiple collaborators, who added their stuff to my exhibit.

The consequence of collective work in the exhibit was, that other people benefited and they also helped me. The public, or some of it, was amused. Mission accomplished.

Wasn't personal computing a result of video games? Not binary stuff?

Lots of talk about "gaming drove personal computing". Well, lots of things "drove" personal computing, at different points in time or at the same time. Like blind people around an elephant, each "sees" it as like a snake, like a tree trunk, like a curtain.....

- Herb Johnson =========

contact:
This page and edited content is copyright Herb Johnson (c) 2024. Contact Herb at www.retrotechnology.com, an email address is available on that page..