Talk:Expensive Desk Calculator
 | This article is rated Start-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects: | |||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||
PDP-1 software
Suggestion to merge with PDP-1 software had no visible support in December 2006. Removing merge (archive). -Susanlesch 06:42, 2 January 2007 (UTC)
Sources needed
Hello. The following unsourced contribution from November 6, 2007 is copied here. -Susanlesch 21:10, 6 November 2007 (UTC)
| “ | One of the best features of Expensive Desk Calculator, which one would only discover on reading the code, was its ability to be assembled for various precisions. Specifically, the source code provided a parameter which could be set to any number from 1-63. Through a very clever usage of nested macros, this number determined the multiplicity of precision, six bits at a time, that would be provided by the arithmetic routines. This made EDC on the PDP-1 a valuable resource for data analysis, even for students with access to the big IBM iron which dominated MIT computing at the time (ca. 1961-63). | ” |
My apologies for any confusion regarding sourcing. I was an MIT Junior Physics major in early 1962 and used the PDP-1 for several class projects. Among other things, I was working on the Musical Acoustics Project, which was led by Prof. Melville Clark of the Nuclear Engineering Department. (Really!) We had recordings of every instrument of the orchestra, played in the anechoic chamber by members of the Boston Symphony. They played the entire chromatic range of their instrument in many different moods. My project was to create strip charts showing the waveform envelope of each note played. We measured the peak amplitudes and correlated them with the note played, for each mood of each instrument.
Without going into detail, we needed a way to quickly write up and run some polynomial fits and get quality estimates. I proposed using the PDP-1 and (by extending the precision of the assembled code as had been planned by the author) was able to get results by using EDC on the PDP-1 very easily. The alternative would have been to get time on the IBM 709 and code it up in FORTRAN. How boring.
So the source is "personal testimony" I guess. Yes, I knew Alan Kotok. Didn't everybody? Ldmjr 06:22, 7 November 2007 (UTC)