Content deleted Content added
Line 198:
:: So the trail from EIA RS-274 to X-code to MIT is as follows:
:: RS-274-C, which was approved in April 1974, unified two different EIA standard: RS-273 (Straight Cut) and RS-274 (Contour Cut), both of which had initial publication in January 1963 (see Library of Congress copyright filings).
:: The EIA RS standards, which governed members of the EIA, were essentially copies of earlier AIA National Aerospace Standards: NAS 943 (Straight Cut) and NAS 955 (Contour Cut). NAS 943 was approved in 1960 (as per the AIA Annual Report, available online). The EIA governed Electronics manufacturers,
:: Those standards are predominately based on the two leading Numerically Controlled mills during the APT work: the Giddings & Lewis mill (trademarked as the Numericord) which used a combination of General Electric and Concord Controls controllers. The second system was the Kearney & Trecker mill, which used a Bendix controller. Both machines had detailed papers published at the Eastern Joint Computer Conference in December 1957 (proceedings available online). The Bendix solution used "R-codes" to describe operational modes (essentially G-codes) and another set to describe auxiliary on/off functions (essentially M-codes). The Giddings & Lewis system used "X-codes" which also described operational modes and auxiliary on/off functions. These are the X-codes described in the APT flowcharts. The assertion that these were the two leading NC mills in the mid-50s can be found in <i>Forces of Production</i> by David F. Noble. The same book contains significant content as to why Bendix became the preferred solution during the standardization process, which was driven as a requirement by the US Air Force.
:: The X-codes are better documented in US patent 2963137 "Controls for a typewriter and associated apparatus" (the patent title is somewhat misleading - this is a numerical control patent). The patent assignee is Giddings & Lewis, but the four inventors (McDonough, Susskind, Grossimon, and Lee) were all MIT staff attached to the Servo Lab. Giddings & Lewis poached the entire MIT numerical control staff (with the exception of Susskind) to form Concord Controls at the beginning of 1956. Note that the same team worked on the original MIT NC Mill, which was finalized in 1952. Bendix licensed the two patents that were generated from the 1952 mill. It's unknown if the idea for X-Code occurred while the group was at Concord Controls or while at MIT. They would have been influenced by activities in the field of Automatic Programming occurring on the MIT Whirlwind computer in 1954 (Susskind was on the Whirlwind staff). Two Automatic Programming activities, well-known in Whirlwind circles, were the "CS" and "SS" programs, which both utilized labels that were called "mnemonic alphadecimals", which consisted of a single alphabetic character, followed by up to 3 decimal digits, the same as the X-Code format. (documented in MIT Report R-233, "The MIT Systems of Automatic Coding", C.W. Adams, available online)
| |||