The First Personal Computers

Simon and the Teaching Machines

In 1949 the mathematician Edmund Berkeley published Giant Brains, the first introduction to electronic computers for general readers. Besides the giants of the title it described a tiny one, Simon, named after the nursery-rhyme Simple Simon, essentially a relay-based Turing machine for teaching the principles. A year later Berkeley built it: a suitcase-sized machine that counted from 0 to 3, took data and its four instructions from paper tape or five keys, and showed results on five lamps. Radio-Electronics put it on the cover in October 1950 and explained it in a series of thirteen articles; Scientific American followed in November. At least one history website calls Simon the first personal computer, but it was closer to the construction kits and teaching computers that followed: the Geniac of 1955, the Minivac 601 of 1961, the Arkay CT-650 of 1969, West Germany’s Logikus (1968) and its East German cousin Piko dat (1969).

The phrase “personal computer” probably first appeared in print in the New Scientist of 27 February 1964, in an article by the IBM researcher Arthur Samuel, author of the 1956 checkers program, for a series imagining “computers in 1984.” By then the term could already describe a real machine. In 1965 the programmer Mary Wilkes was allowed to take a LINC, a laboratory minicomputer designed for research, home to her apartment in Baltimore and work on it there.

Desktop Machines, 1965 to 1975

Olivetti’s Programma 101, on sale from 1965, was a desktop machine with a printer, ten registers on a delay line, and magnetic cards that held 120 instructions, including a conditional jump; more than 35,000 were sold at about $3,200 to $3,500 (around 15,000 Deutschmarks in West Germany) (see Camillo and Adriano Olivetti). At the Leipzig autumn fair of 1965 the Dresden pioneer Nikolaus Joachim Lehmann showed the D4a, a transistor computer the size of a television set that managed 1,900 operations per second; it went into series production in modified form as the Cellatron 8205 in 1969 and sold 3,000 units, all of them, as far as is known, to factories, institutes and offices (see Computing in East Germany).

Over the next decade the gap closed from several directions at once. There were sophisticated desktop calculators (Wang 300 and 700, Hewlett-Packard’s 9100A and 9830), programmable office machines (the Wanderer Conti, the Diehl Combitron, and Heinz Nixdorf’s Logatronic of 1965, which became the Nixdorf 820), intelligent terminals, and stripped-down minicomputers (Wang 2200, IBM 5100, Xerox Alto). Nixdorf was still calling its LK 3000 a “personal computer” at the Hanover fair of 1980 because, in the words of Computerwoche, “you can use it personally.”

The Amateur Computer Society

On 5 May 1966 Stephen Gray, an editor at the magazine Electronics who had spent years failing to build a computer in his spare time, sent ten technical magazines an “invitation to computer builders”: anyone constructing a machine that could multiply and divide was asked to get in touch. Five magazines printed it, 160 readers answered, all of them men, and 110 joined his Amateur Computer Society, most from the United States, especially California, some from Canada, Italy, Japan and Switzerland. Members paid three dollars. From August 1966 Gray typed and copied a newsletter every two months.

Newsletter number 8, of January 1968, described two finished amateur computers, both completed in 1965: Jim Sutherland’s ECHO IV, now in the Computer History Museum, and the EL-65 of the young Swiss Hans Ellenberger. Hal Chamberlin’s HAL-4096 was running by 1968. From November 1975 to December 1976 the newsletter listed 67 new microcomputer systems; the Apple I was number 52. The cheap microprocessor kits ended the era of the lone builder, and Gray closed the society with a farewell letter in February 1977.

The Kenbak-1

In September 1971 an advertisement in Scientific American offered the Kenbak-1 for $750, a price “that even private individuals and small schools can afford.” Its designer, John Blankenbaker, born in Oklahoma in 1929, had read about the UNIVAC’s binary arithmetic as a student in 1949, looked after the SEAC during a summer job at the National Bureau of Standards in 1951 and worked in industry in the 1960s; in 1970 he took time off to build a small computer from integrated circuits. The machine, 49 by 29 centimetres and 6.5 kilograms, used no microprocessor; it worked with eight-bit data, stored 256 bytes in shift registers, took input from push buttons and showed output on lamps. Blankenbaker meant it for schools, but sold only about forty to schools and a dozen to private buyers before stopping production in 1973. The Computer Museum in Boston named it the first personal computer in 1986; in 2015 the Deutsches Museum in Munich bought one at auction for €34,000.

Terminals That Were Computers

The Datapoint 2200 of the Computer Terminal Corporation (CTC) in San Antonio, Texas, was announced in June 1970 and delivered from March 1971 for $7,800. It was sold as a programmable terminal the size of an IBM Selectric typewriter, with a twelve-line screen and two tape cassette drives, one for the terminal program and one for data, and two to eight kilobytes of internal memory; Victor Poor and Harry Pyle designed the electronics. A sales manager, Dave Gust, saw that it was a general-purpose computer and sold five to Pillsbury, where four of them did the payroll for chicken farms in April 1971. By the end of 1972 CTC, renamed Datapoint, was advertising it as a computer with disk drives and an operating system called DOS. CTC had given its processor design to Texas Instruments and Intel, but did not use the single-chip versions they produced; Intel sold its version from April 1972 as the 8008, ancestor of the 8080 and the x86 family (see Federico Faggin and the Microprocessor).

On 18 January 1971 George Cogar, a former UNIVAC engineer, presented the Cogar System 4 in Herkimer, New York: a desktop machine for data entry with a keyboard, a small screen, two tape drives and sixteen kilobytes of semiconductor memory, which his company mentioned could also serve as a “stand-alone mini computer.” Cogar sold more than 6,000 in five years before Singer bought the company in 1975.

The First Microprocessor Computers

The first computer sold with a microprocessor came from France. In 1972 Alain Perrier of the agricultural research institute INRA needed a cheap computer to measure evaporation from fields, and the small firm R2E near Paris built one around Intel’s 8008. François Gernelle and three colleagues finished it by the end of the year; Perrier received it in January 1973, and in February R2E put it on sale as the Micral N for 8,500 francs (see France’s Tech Industry).

In Canada Mers Kutt, who knew Robert Noyce and had heard of Intel’s eight-bit chip before its release, founded Kutt Systems in Toronto in December 1971 and built the MCM/70 around the 8008. Presented on 25 September 1973 and delivered from the summer of 1974 for $3,500 to $9,800, it had a keyboard, a one-line plasma display, cassette drives and the programming language APL built in (see APL and Array Programming). Fewer than a thousand were sold.

In the United States the Scelbi-8H, offered from March 1974 for about $500 as a kit or assembled, also used the 8008 and sold poorly. In July 1974 Radio-Electronics put Jon Titus’s Mark-8 on its cover as “your personal minicomputer”; readers could buy the six circuit boards for $47.50 and had to source the chips themselves, and about four hundred sets were sold. Its rival Popular Electronics answered in its January 1975 issue, on the stands in December 1974, with the Altair 8800. Utah’s Sphere Corporation announced the Sphere 1 in July 1975: a Motorola 6800 machine with keyboard and screen, $650 as a kit, of which 1,300 were built before the company failed in the spring of 1977.

The MCS-40

Germany’s first microprocessor computer was a diploma project. From 1974 Ulrich Klotz, an engineering student at the Technical University of Berlin, built a teaching computer around Intel’s four-bit 4040 chip and its MCS-40 chip set, which gave the machine its name. The hardware was ready in the spring of 1975 and the operating system ran in December 1975; software was entered bit by bit through toggle switches on the front panel, and a Teletype served as terminal. It could show step by step what went on inside the processor and control simple experiments, such as holding a steel ball suspended under an electromagnet. The machine later taught students in Hamburg and is now in the collection of the Heinz Nixdorf MuseumsForum. Klotz went on to Nixdorf’s Berlin research centre and spent most of his career as innovation adviser to the IG Metall trade union.

What “Microcomputer” Meant

In the mid-1970s the word “microcomputer” did not yet mean a small computer. Popular Electronics introduced the Altair as a “minicomputer kit”; in May 1975 Scientific American defined the microcomputer as “a set of microelectronic ‘chips’” serving the functions of a computer, and Adam Osborne’s 1976 textbook An Introduction to Microcomputers told hobbyists that they built their computers “out of a microcomputer.” Apple’s 1977 advertising called the Apple II a “personal computer,” and only gradually did both words come to mean the machine on the desk.

📚 Sources

• Heinz Nixdorf MuseumsForum blog, “Der Computer wird persönlich” (18 February 2016)
• Heinz Nixdorf MuseumsForum blog, “Wir bauen uns einen Computer” (3 May 2016)
• Heinz Nixdorf MuseumsForum blog, “Kenbak-1 – ein Computer macht Schule” (7 September 2021)
• Heinz Nixdorf MuseumsForum blog, “Datapoint 2200: der erste Kleincomputer” (5 June 2020)
• Heinz Nixdorf MuseumsForum blog, “Cogar System 4 – klein aber Computer” (19 January 2021)
• Heinz Nixdorf MuseumsForum blog, “Klein und französisch” (6 April 2018)
• Heinz Nixdorf MuseumsForum blog, “MCM/70 – der kleine Kanadier” (18 May 2021)
• Heinz Nixdorf MuseumsForum blog, “Der letzte Vorläufer – Mark-8” (12 July 2019)
• Heinz Nixdorf MuseumsForum blog, “Mikrocomputer aus Utah – Sphere 1” (9 January 2024)
• Heinz Nixdorf MuseumsForum blog, “MCS-40: der erste deutsche Mikrocomputer” (13 January 2026)
• Heinz Nixdorf MuseumsForum blog, “Netzwerk für Jahrtausendmenschen” (4 June 2016)
• Heinz Nixdorf MuseumsForum blog, “Was ist eigentlich ein Mikrocomputer?” (25 September 2015)