The University of Manchester has played an important role in the development of computer science. In the 1930s Douglas R. Hartree (1897-1958), professor of physics, constructed a differential analyser, a mechanical calculating machine, based upon the designs of the American engineer, Vannevar Bush. This was an analogue device. After the Second World War these machines were overtaken by electronic stored-program digital computers.
In 1946 Professor (Sir) F.C. Williams (1911-1977) and (Professor) Tom Kilburn (1921-2001) began work at the University with the intention of developing a novel form of computer storage using cathode ray tubes. The system, which involved the use of the "Williams tube" to store "bits" of information, was perfected during 1947, with Kilburn publishing the results, together with the outline design for a hypothetical computer, in December of that year. The team was also joined by G.C. Tootill, who, like Williams and Kilburn, had previously worked at the Telecommunications Research Establishment at Malvern. A prototype machine, (the forerunner of the Manchester Mark I), was built and on 21 June 1948 it became the world's first stored-program computer to operate. The successful running of this first program, recorded Williams, "was the breakthrough and sparks flew in all directions".
The Manchester group doubled its size in 1948 by taking on two research students (D.B.G. Edwards and G.E. Thomas). The team was a remarkably talented one: besides Williams and Kilburn, who provided the electrical engineering skills, Professor M.H.A. (Max) Newman (1897-1984) and Alan Turing (the latter having joined the Mathematics Department in 1948), gave theoretical expertise. Turing, for example, designed, with Edwards and Thomas, the paper tape input/output system and wrote a programming manual.
The prototype machine was then reworked from late 1948, and redesigned versions were implemented in April 1949, which carried out work on Mersenne Primes, and in October 1949, which was used to investigate the Riemann hypothesis and calculations in optics. This version of the Mark I worked until the summer of 1950. In 1948 the attention of Sir Ben Lockspeiser, the then Government Chief Scientist, was drawn to the Mark I. The result was a Government contract with Ferranti Ltd to make a production version of the machine "to Professor Williams' specification". The first Ferranti Mark I was installed at the University of Manchester in February 1951, thereby becoming the world's first commercially available computer to be delivered. Ferranti marketed this computer as the "Manchester Electronic Computer"). The Government's involvement with the University of Manchester proved worthwhile: royalties from the Williams' patents (the first of which had been filed on 11 December 1946 by the Ministry of Supply) gave an important boost to the National Research Development Corporation (NRDC), a Government body, set up in 1949 to advise on, and support, developments in British industry.
The Ferranti Mark 1 was used extensively by other universities, research bodies and businesses; fees raised were used in part for developing new computers. Tony Brooker of the Computing Machine Laboratory developed the Mark 1 Autocode system as an accessible programming language (Brooker also developed autocode languages for Atlas and Mercury).
The Department's next project was the MEG or Mark II Computer (Meg was a shortening of megacycle engine), developed between 1951 and 1954. Meg was intended to be a faster version of the Mark I, and included floating point arithmetic facilities (this promoted faster processing by using approximations of real numbers). Meg also had a reduced number of thermionic tubes. Meg ran its first program in May 1954 and was later developed commercially by Ferranti as the Mercury. The University acquired a Mercury computer in 1957, paid for by Mark I funds and a UGC grant (the Ferranti Mark 1 machine was decommissioned in December 1958).
Running at the same time as the MEG was the Transistor computer. This was a small-scale machine, which used transistors instead of thermionic tubes. It was developed by R L Grimsdale and D C Webb, with assistance from some of the Department’s research students. Two prototype transistor computers were developed in 1953 and 1955; the 1953 machine is considered to be the first transistor computer.
By the late 1950s there was growing concern about the lack of fast computers in the UK. In response, Kilburn developed the Muse project (the word is derived from microsecond). This aimed to develop a computer with much greater immediate access storage capacity. The project attracted interest from industry and research bodies, but was extremely ambitious in scope and funding was uncertain. the department developed a more basic version based on work by Dai Edwards and Tony Brooker. In 1958, Ferranti became involved with the project, and support was also provided by the National Research Development Corporation. This collaboration eventually resulted in the ATLAS computer, an ambitious project that pioneered many concepts in storage and addressing which are in common use today. On its official inauguration on 7 December 1962 it was considered to be the most powerful computer in the world. Atlas was used at Jodrell Bank Observatory and the Science Research Council's Atlas Computer laboratory at Harwell. It was also used in the development of the Cambridge Titan machines. The University's Atlas computer was officially decommissioned in September 1971.
The University was not the only centre in Britain or the world involved in pioneering computer technology. Later in the 1960s, developments elsewhere, particularly in the USA, had overtaken Manchester. New computer software as well as new hardware, such as transistors and silicon chips, eroded Manchester University's lead in computers. Nevertheless, it is worth recording that besides building the world's first stored-program computer, as well as the world's first commercially available computer, Kilburn and his group can be credited with building the first proper transistor computer in 1953 (the Metropolitan Vickers Company later built a commercial version of the design).
Both in commercial and technical terms the 1950s and 1960s were the ' vintage' years of the Manchester Department (it had officially separated from the Department of Electrical Engineering to become the Department of Computer Science). The Department's last major 'big' computer project was the MU5. Planning for this began in 1966 and it was intended to be 20 times faster than Atlas. MU5 aimed to run high level language programs more efficiently. The project received funding from the SRC and ICL (a British company created from a merger of the major computer manufacturers) assisted with production facilities (mainly at its West Gorton plant). An ICL 1905E computer was used for the MU5 logic simulator and test-bed for its associative store. The MU5 design team included Kilburn, Edwards, Sumner, D Aspinall and J S Rohl. There was a great emphasis on software development and by 1974, when the MU5 became fully operational, it was using Algol and Fortran. ICL developed a successful range of commercial computers bass on the MU5 (the 2900 series).
By the late 1970s, the Department's involvement in building big processors was changing. Kilburn retired in 1981, symbolically bringing an end to this phase in the Department's development. From the late 1940s to the late 1970s, Manchester’s computing group had arguably been the most significant university research team in computing in the world. The five major Manchester computer projects made a major contribution to the development of computer hardware and software. Finally, all these developments established an important link between the University and the computer industry, which has lasted through various projects to the present day.