Talk:PARC (company)/Archives/2015

The Digital Music Sampler

A Xerox PARC engineer took me there one evening in 1975 to show me the various wonderful things. I saw the mouse and GUI, but as a music enthusiast, those were far overshadowed by the astonishing Digital Music Sampler. There was a small room, like a closet, with a full music keyboard connected to a computer and a stereo with speakers. When you played the keyboard, you heard the sound of a huge multi-story pipe organ. At the time, it was astonishing. Looking through Wikipedia, it appears that EMS had invented this in 1969, and it is unclear whether there was any connection between the three of them and Xerox PARC, or whether PARC had just come up with the idea separately.

I have not added this to the actual wikipedia page, because of the absurd rule that no first hand information is allowed - so I would have to put the paragraph on some web page out there, and then link to that as a "reference" (oooh!) and then that makes it acceptable (rolls eyes). — Preceding unsigned comment added by 64.166.167.10 (talk) 00:07, 7 July 2012 (UTC)

(From Alan Kay) The "Digital Music Sampler" was part of the work of my research group (The Learning Research Group). The pioneer of digital music was Max Matthews of Bell Labs going back into the 60s and the techniques were well known -- but were not done digitally in real-time. Ralph Deutsch of North American Rockwell had made a chip ca 1971 that Allen Organ used for real-time additive synthesis. The first real-time system at PARC was done by me using a Data General Nova and a "long-sample" scheme (that is like what is now used in the organ simulator Hauptwerk). We got 3 simultaneous voices on the Nova (and they sounded great). This provided benchmarks for then being designed Alto, which wound up being able to do 12 polytimbral simultaneous voices, and to be controlled by a two manual and pedal organ keyboards that the Alto could read. In 1974 we did a Parc-version of the FM synthesis that John Chowning at Stanford had been experimenting with (at about 100 times slower than real-time) and were able to find a way for the Alto to do 8 real-time polytimbral voices (despite not having a multiplier). Steve Saunders did the important parts of the work for this project, and a few of the rest of us kibitzed. There is a photo of the setup at http://alexshye.files.wordpress.com/2013/12/alan-kay.jpg — Preceding unsigned comment added by 172.248.33.123 (talk) 00:11, 22 January 2015 (UTC)