home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.atari.st.tech
- Path: sparky!uunet!stanford.edu!ames!elroy.jpl.nasa.gov!hanauma.jpl.nasa.gov!hyc
- From: hyc@hanauma.jpl.nasa.gov (Howard Chu)
- Subject: Re: Falcon BUS... DSP
- Message-ID: <1992Nov19.041528.16856@elroy.jpl.nasa.gov>
- Keywords: SoUnD
- Sender: news@elroy.jpl.nasa.gov (Usenet)
- Nntp-Posting-Host: hanauma.jpl.nasa.gov
- Organization: SAR Systems Development & Processing, JPL
- References: <11065@uqcspe.cs.uq.oz.au> <2681@ulysse.enst.fr>
- Date: Thu, 19 Nov 1992 04:15:28 GMT
- Lines: 31
-
- In article <2681@ulysse.enst.fr> elendir@inf.enst.fr writes:
- >Warwick Allison writes
- >] Also, with the DSP running at 32MHz, it has time for quite a few
- >] instructions for fiddling with those values too, besides adding the
- >] sounds together. Of course, when processing 90 voices at once, you'd
- >] be pushing it pretty hard :-)
- >
- > A 32 MHz DSP has a 16 MHz instruction clock rate. A 1 Mbyte/sec throughput
- >corresponds to 300 KDSPwords/sec, or one interrupt every 3 us. You have time
- >to execute 48 instructions then between each interrupt. Enough to do some
- >pretty effects : Surround, Echo, Reverb, Basic IIR filters and much more.
- >
- > If you want a demo of the sound capabilities of the DSP, watch out for the
- >Audio Fun Machine demo at the Comdex (Personnal pub).
-
- I just got home from Comdex. Yeah, the AFM is a nice little toy. (All kinds
- of sound effects in realtime.)
-
- As for sound channels, I think Warwick has it right on the money. (Although
- trying to implement an actual orchestra, and then only going to 11khz sampling
- is going to put a real damper on your instruments, eh? }-) My company is
- currently showing a single-line voice-mail system on the Falcon (most of the
- group is still in Vegas right now) but it would be trivial to bump it up to
- 6 or 12 phone-lines per sound-track, giving a maximum capacity of handling 96
- phone lines simultaneously. (I suppose we would have to be nice and leave
- at least two channels free for system audio, though, unless the machine was
- used as a dedicated automated response system...)
- --
- -- Howard Chu @ Jet Propulsion Laboratory, Pasadena, CA
-
- All true wisdom is conveyed in one-line witticisms.
-