Pilot ACE
The Pilot ACE was a prototype of the full Automatic Computing Engine (ACE). It too was designed by Alan Turing personally, with the assistance of J. H. Wilkinson. Like the ACE, its main memory also was mercury delay lines, and its whole architecture was adapted to their characteristics, to a degree which today is hard to imagine without seeing the details (for instance, in the use of using optimum programming, to give a simple example). For its day, as a result, it was thus a very fast machine indeed, although difficult to program.
After Turing did the ACE design in late 1945, the NPL moved very slowly on doing anything (although in fairness, resources at NPL were short, compared with what Turing had become used to at Bletchley Park). In January, 1947, Harry Huskey joined the group, and pushed for action, and the group started work on a machine called the Test Assembly, to prototype Turing's ideas. Turing, apparently unhappy at the delays in building the ACE, left NPL at the end of September, 1947, eventually ending up at Manchester University.
The NPL had by then abandoned the Test Assembly, which was a marginal design, and instead moved on to a scaled-down prototype of the ACE, the Pilot ACE. Construction of that started in early 1949, and it was initially operational by May, 1950; a public demonstration was held at the end of November, 1950. It contained about 800 vacuum tubes.
It was so successful that English Electric, which had assisted on the building of the Pilot ACE, produced a copy of it as a product, the DEUCE, of which no less than 33 were built. The Bendix G-15, done by Huskey, was also a descendant of the Pilot ACE. The original Pilot ACE was shut down in June, 1956.
Technical details
In addition to 11 (at the end of its development) long delay lines, each of which had a capacity of 32 32-bit words, the Pilot ACE also had 2 double-word lines (for double-length operands), and 5 single-word lines (the reduced size gave lower access times). The shorter ones had the role of registers in most computers. The time required for one word (the whole machine was serial) was termed a 'minor cycle'; a complete cycle of a long line was a 'major cycle'.
The Pilot ACE's instructions were also 32 bits. They contained 7 fields:
| Field | Width | Description |
|---|---|---|
| Next Instruction Source | 3 | Next instruction line |
| Source | 5 | Line holding source operand |
| Destination | 5 | Line where result is to be stored |
| Characteristic | 2 | Number of minor cycles for this instruction |
| Wait | 5 | Minor cycle in which to start instruction |
| Timing | 5 | Next instruction minor cycle |
| Go | 1 | Actuated by console key |
Five bits were unused. For sources in long lines, the minor cycle used was that selected by the 'Wait' field. Instructions contained 3 addresses; a source, a destination, and the address of the next instruction. (The Pilot ACE had no Program Counter.) Note that there is no operation code in the instruction; instead, many of the sources and destinations had side-effects.
It could do an addition in 64 μsec. The machine underwent a number of changes during the period 1946-49; the information above represents its final configuration. A hardware multiplier was added in 1951.
Input/output used punched cards; in February, 1954 the Pilot ACE was given a drum (holding 32 tracks initially, eventually 128; each containing 32 words - the same size as the large delay lines); it used a moving group of 16 heads.
Further reading
- Alan Turing, B. E. Carpenter (editor), R. W. Doran (editor), A. M. Turing's ACE report of 1946 and other papers, MIT Press/Tomash, Cambridge/Los Angeles, 1986 - the Pilot ACE is covered in the last document re-printed herein
- David M. Yates, Turing's Legacy: A History of Computing at the National Physical Laboratory 1945-1995, Science Museum, London, 1997 - the environment which produced the Pilot ACE is covered in Section 2.3, in some detail; the Pilot ACE itself is covered in Section 2.4; descendants are covered on pp. 40-46
- B. Jack Copeland (editor), Alan Turing's Automatic Computing Engine: The Master Codebreaker's Struggle to Build the Modern Computer, Oxford University, Oxford, 2005 - essays from a number of people, including a lot of material on various aspects of the Pilot ACE
- Harry D. Huskey, From ACE to the G-15, Annals of the History of Computing, Vol. 6, No. 4, Oct0ber, 1984, pp. 350-371
External links
- Architecture of the English Electric DEUCE computer - also covers the Pilot ACE in some detail
- Evolution of the ACE
- ACE hardware
- Photographs of the Pilot Model ACE
