About this Item
Wilkes, Maurice Vincent (1913- ). A method of solving second order simultaneous linear differential equations using the Mallock machine. Offprint from Proceedings of the Cambridge Philosophical Society 36, part II (April 1940). Original buff printed wrappers. Signed by Wilkes on the front wrapper. [1], 204-208 [2]pp. 257 x 184 mm. Provenance: Maurice Wilkes. First edition, offprint issue. Wilkes directed the design and construction of EDSAC, the first readily usable, full-scale stored-program computer. EDSAC was preceded in operation by the Manchester "Baby"prototype stored-program machine which ran for only a short time in 1948; in America, BINAC was probably running programs about the same time, but it too was a very short-lived machine. In addition to developing EDSAC, Wilkes was responsible for a number of programming innovations, such as labels, macros, and microprogramming, that became standard in the computer industry. He studied physics at Cambridge University, where he received his Ph.D. in 1936 with a thesis based on work performed in the Cavendish Laboratory on the propagation of very long radio waves in the ionosphere. While engaged in postgraduate research on this topic, he was allowed to use Cambridge University's model differential analyzer to solve a difficult equation. This machine, which Wilkes found "irresistible" (Wilkes 1985, 25), inspired an abiding interest in automatic computing. At the end of 1936, Wilkes was put in charge of Cambridge's model differential analyzer, and the following year he joined the staff of the university's newly founded Mathematical Laboratory, becoming its director after the close of World War II. The Mathematical Laboratory (renamed the Computing Laboratory in the 1960s) played a critical role in the development of the electronic digital computer. One of Wilkes's tasks during 1937 was to gather information about the special-purpose calculating machine invented by R. R. M. Mallock (of Cambridge's Engineering Department), which the Mathematical Laboratory was interested in purchasing. Wilkes described the machine as follows: The Mallock machine was an analogue device and was capable of solving ten simultaneous linear equations in ten unknowns. It was based on the use of tapped transformers with the windings connected to form a network. The accuracy obtainable from such an arrangement might be expected to be very low because of losses in the transformers. What made the Mallock machine give a useful accuracy-one part in 1000 in favourable cases-was the use of a highly ingenious feedback circuit, known as a compensator, associated with each transformer. As a piece of electronics, this was well ahead of its time. . . . In order to solve a set of simultaneous equations, one had first to set the coefficients on an array of digital switches. The roots were then obtained by adjusting another switch until a galvanometer showed zero. This had to be done for each root in turn. When I got to know him, Mallock was experimenting with a device based on the use of telephone relays for performing this operation automatically and printing the result on a paper strip. Although this gear came with the Mallock machine, it was not fully developed and we made no attempt to use it. However, it gave me my first introduction to the use of telephone relays in computing, or rather control, circuits and to some of the tricks one can play with them (Wilkes 1985, 29). Wilkes 1999, no. 6. Origins of Cyberspace 1014 .
Seller Inventory # 39743
Contact seller
Report this item
