🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: comp.sys.next.hardware
Article #1143

Re: DSP chips are NOT dinosaurs

#1143
From: Stephen A. Davis
Date: Thu, 01 Oct 1992 21:15
23 lines
1570 bytes
In article <2986@news.cerf.net> billabs@nic.cerf.net (Bill Romanowski) writes:
 
> SGI takes a more specific approach with it's DSP. It's used only as an audio
> resource for the system. No DSP coding support is provided (that I could  
find)
> for math or serial port hacking. But the audio is clean, fast and useable up  
to
> DAT rates.
> 
SGI will _never_ support user programming of the DSP.  Ask them and they'll  
tell you the same.  They think all DSP-type processing should be handled by  
the CPU but they've made modifications to the OS to support real-time  
scheduling.  The DSP is used as an I/O controller and runs a real-time kernel  
of its own to handle up to 4 channels of audio data in or out.  Possible  
future revisions will include a faster DSP for more channels but the apparent  
limit at this time is not the DSP but the DMA controller.  In truth, I just  
regret that you're stuck with only two channels on the Indigo.  Sigh.

stephen
--
sadman@ccrma.stanford.edu	
The Center for Computer Research in Music and Acoustics (CCRMA)
*** NeXT mail okay ***
Technical Details
Message-ID: <1992Oct1.211512.29049@leland.Stanford.EDU>
Path: arc1-ca.pugleaf.net!sparky!uunet!stanford.edu!leland.Stanford.EDU!news
References: <2986@news.cerf.net>

Show Full Headers
Complete Article Headers
Newsgroups: comp.sys.next.hardware
Path: arc1-ca.pugleaf.net!sparky!uunet!stanford.edu!leland.Stanford.EDU!news
From: sadman@cccrma.Stanford.EDU (Stephen A. Davis)
Subject: Re: DSP chips are NOT dinosaurs
Message-ID: <1992Oct1.211512.29049@leland.Stanford.EDU>
Sender: news@leland.Stanford.EDU (Mr News)
Reply-To: sadman@ccrma.Stanford.EDU (Stephen A. Davis)
Organization: DSO, Stanford University
References: <2986@news.cerf.net>
Date: Thu, 1 Oct 92 21:15:12 GMT
Lines: 23