Ludgate's Analytical Machine
1909 · Ludgate Design · Work 501 of 1,871
A printed page of a scientific proceedings volume of 1909, dense typeset paragraphs in two columns reading as grey texture, a heading rule across the top, cream paper laid flat on a grey studio table, even flat light from directly above.
- What it was used for
- Percy Ludgate (1883-1922) was an auditor with Kevans and Son in Dame Street, Dublin, and designed a program-controlled calculating machine in his spare time; his niece said he used to work at it nightly, until the small hours of the morning. The design appeared in "On a proposed analytical machine", Scientific Proceedings of the Royal Dublin Society 12, no. 9 (1909), pp. 77-91, read on 23 February and published on 28 April 1909. It provided storage for 192 variables of twenty figures each. Ludgate gives about 10 seconds for the multiplication of two 20-figure numbers, about three seconds for addition or subtraction, a probable maximum of one and a half minutes for a division and two minutes for a logarithm.
- What was new
- Ludgate did not multiply by repeated addition as Babbage did, but through index numbers that Professor Boys called Irish logarithms. Control came from a single perforated formula paper; Randell writes that this sequencing has more in common with the Harvard Mk. 1 of nearly forty years later than with Babbage's separate card strings. The numbers sat in shuttles independent of the machine, so that new variables could be introduced at any time - for Randell a foreshadowing of replaceable storage media. The barb: the machine was never built, and the drawings are gone. Whether they were among Ludgate's effects in 1922 is unknown; at his brother Alfred's death in 1953 no such papers were found among Alfred's effects. Of a machine 26 by 24 by 20 inches, two papers remain.