From owner-freebsd-multimedia  Tue Aug 26 13:54:00 1997
Return-Path: <owner-freebsd-multimedia>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id NAA07150
          for multimedia-outgoing; Tue, 26 Aug 1997 13:54:00 -0700 (PDT)
Received: from labinfo.iet.unipi.it (labinfo.iet.unipi.it [131.114.9.5])
          by hub.freebsd.org (8.8.7/8.8.7) with SMTP id NAA07133
          for <freebsd-multimedia@FreeBSD.ORG>; Tue, 26 Aug 1997 13:53:50 -0700 (PDT)
Received: from localhost (luigi@localhost) by labinfo.iet.unipi.it (8.6.5/8.6.5) id VAA00302; Tue, 26 Aug 1997 21:37:55 +0200
From: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
Message-Id: <199708261937.VAA00302@labinfo.iet.unipi.it>
Subject: Re: Regard to phone call about FreeBSD drivers
To: hasty@rah.star-gate.com (Amancio Hasty)
Date: Tue, 26 Aug 1997 21:37:55 +0200 (MET DST)
Cc: tinguely@plains.NoDak.edu, freebsd-multimedia@FreeBSD.ORG, sales@omt.com
In-Reply-To: <199708261725.KAA05572@rah.star-gate.com> from "Amancio Hasty" at Aug 26, 97 10:25:04 am
X-Mailer: ELM [version 2.4 PL23]
Content-Type: text
Sender: owner-freebsd-multimedia@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

> >From The Desk Of Luigi Rizzo :
> > make it $20-$40 for a soundcard. The OPTI931 has a street price of $20
> > here in italy, and CS4236 (assuming you can find them) are little more
> > expensive. For conferencing purposes they are both more than ok
> > (Crystal much better since it is better documented and less buggy...)
> 
> Well, it depends the clock they use and the quality of the analog
> parts. Conferencing is one thing however I think that you may 
> want to use the soundcard for other things like playing back
> audio mpeg files. By far most PC soundcards fall short on 
> the clock part . Interestingly, the old sparcs used to have a
> a cs4231 and the accuracy of their dsp subsystem has always
> been better than your typical PC soundcard -- the primary difference
> from an accuracy stand point of view is just the clock part.

the opti931 boards I have take the clock from the 14.3xxx clock on the
ISA bus, but the internal division do not yeield the exact frequencies
(44.1KHz, 8KHz, etc.) one would want, there is some 0.15% or so of
difference. I doubt one would notice this (well unless it uses several
machines  with different clocks at the same time to simulate an
orchestra, each machine playing a different set of instruments...).

Also, if there are no short term drifts I believe the
differences can be compensated (e.g. in software by adding/dropping
samples every now and then).

Apart from clock stability, I suppose that dma overruns/underruns might
also make the clock appear to have a different frequency (sorry for the
convoluted sentence, I am italian...).

	Cheers
	Luigi