Atransistor computer, now often called asecond-generation computer,[1] is acomputer which uses discretetransistors instead ofvacuum tubes. The first generation of electronic computers used vacuum tubes, which generated large amounts of heat, were bulky and unreliable. A second-generation computer, through the late 1950s and 1960s featuredcircuit boards filled with individual transistors andmagnetic-core memory. These machines remained the mainstream design into the late 1960s, whenintegrated circuits started appearing and led to thethird-generation computer.
TheUniversity of Manchester's experimentalTransistor Computer was first operational in November 1953 and it is widely believed to be the first transistor computer to come into operation anywhere in the world. There were two versions of the Transistor Computer, the prototype, operational in 1953, and the full-size version, commissioned in April 1955.[2] The 1953 machine had 92point-contact transistors and 550diodes, manufactured bySTC. It had a48-bit machine word. The 1955 machine had a total of 200 point-contact transistors and 1,300 point diodes,[3] which resulted in a power consumption of 150 watts. There were considerable reliability problems with the early batches of transistors and the average error-free run in 1955 was only 1.5 hours. The Transistor Computer also used a small number of tubes in its clock generator, so it was not the firstfully transistorized machine.[4]
The design of a full-size Transistor Computer was subsequently adopted by theManchester firm ofMetropolitan-Vickers, who changed all the circuits to use more reliablejunction transistors. The production version was known as theMetrovick 950 and was built from 1956 to the extent of six[3] or seven machines, which were "used commercially within the company"[5] or "mainly for internal use".[3]
During the mid-1950s a series of similar machines appeared. These included theBell LaboratoriesTRADIC, completed in January 1954, which used a single high-power output vacuum-tube amplifier to supply its 1-MHz clock power.[6]
The first fully transistorized computer was either theHarwell CADET, which first operated in February 1955, although the price paid for this was that it operated only at the slow speed of 58 kHz,[citation needed] or the prototypeIBM 604 transistor calculator. TheBurroughs Corporation claimed theSM-65 AtlasICBM / THOR ABLE guidance computer (MOD 1) that it delivered to the US Air Force at theCape Canaveral missile range in June 1957 was "the world's first operational transistorized computer".[citation needed]MIT'sLincoln Laboratory started developing a transistorized computer theTX-0 in 1956.
Further transistorized computers became operational in Japan (ETL Mark III, July 1956), in Canada (DRTE Computer, 1957), and in Austria, (Mailüfterl, May 1958),[7] these being the first transistorized computers in Asia, Canada and mainland Europe respectively.
In April 1955,[8]IBM announced theIBM 608 transistor calculator, which was first shipped in December 1957.[9] IBM and several historians thus consider the IBM 608 the first allsolid-state computing machine commercially marketed.[8][10][11][12] The development of the 608 was preceded by the prototyping of an experimental all-transistor version of the 604. This was built and demonstrated in October 1954, but was not commercialized.[9][11][13]
The Philco Transac models S-1000 scientific computer and S-2000 electronic data processing computer were early commercially produced large-scale all-transistor computers; they were announced in 1957 but did not ship until sometime after the fall of 1958. The Philco computer name "Transac" stands for Transistor-Automatic-Computer. Both of thesePhilco computer models used the surface-barrier transistor in their circuitry designs, the world's first high-frequency transistor suitable for high-speed computers.[14][15][16] Thesurface-barrier transistor was developed by Philco in 1953.[17]
RCA shipped theRCA 501 its first all-transistor computer in 1958.[18]
In Italy,Olivetti's first commercial fully transistorized computer was theOlivetti Elea 9003, sold from 1959.[19]
IBM, which dominated the data processing industry through most of the 20th century, introduced its first commercial transistorized computers beginning in 1958, with theIBM 7070, a ten-digit-word decimal machine.[20] It was followed in 1959 by theIBM 7090, a36-bit scientific machine, the highly popularIBM 1401 designed to replace punched cardtabulating machines, and the desk-sized1620, a variable length decimal machine. IBM's7000 and1400 series included many variants on these designs, with different data formats, instruction sets and even different character encodings, but all were built using the same series of electronics modules, the IBM Standard Modular System (SMS).[21]
Developers of the TX-0 left to form theDigital Equipment Corporation in 1957. Transistorized from the beginning, early DEC products included thePDP-1,PDP-6,PDP-7 and earlyPDP-8s, the last starting theminicomputer revolution. Later models of the PDP-8 beginning with PDP-8I[22] in 1968 used integrated circuits making them third generation computers
In 1964, IBM announced itsSystem/360, a collection of computers covering a wide range of capabilities and prices with a unified architecture, to replace its earlier computers. Unwilling to bet the company on the immature monolithic IC technology of the early 1960s, IBM built the S/360 series usingIBM's Solid Logic Technology (SLT) modules. SLT could package several individual transistors and individual diodes with deposited resistors and interconnections in a module one-half inch square, roughly the equivalent logic of the earlierIBM Standard Modular System card, but unlike monolithic IC manufacturing, the diodes and transistors in an SLT module were individually placed and connected at the end of each module's assembly.[21]
First generation computers were largely out of reach of schools and hobbyists who wished to build their own, largely because of the cost of the large number of vacuum tubes required (though relay-based computer projectswere undertaken[23]). The fourth generation (VLSI) was also largely out of reach, too, due to most of the design work being inside the integrated circuit package (though this barrier, too, was later removed[24]). So, second and third generation computer design (transistors and MSI) were perhaps the best suited to being undertaken by schools and hobbyists.[25]
{{cite journal}}: CS1 maint: DOI inactive as of July 2025 (link)