From alsa-devel-owner@alsa.jcu.cz  Thu Feb  4 13:38:53 1999
Received: from keg.cs.vu.nl (root@keg.cs.vu.nl [130.37.24.4])
	by marvin.jcu.cz (8.9.1a/8.9.1) with ESMTP id NAA17772
	for <alsa-devel@alsa.jcu.cz>; Thu, 4 Feb 1999 13:38:05 +0100
Received: from localhost by keg.cs.vu.nl with esmtp
	(Smail3.1.28.1 #49) id m108O2l-0007ELC; Thu, 4 Feb 99 13:38 +0100
Date: Thu, 4 Feb 1999 13:38:03 +0100 (MET)
From: Andy Lo A Foe <arloafoe@cs.vu.nl>
To: alsa-devel@alsa.jcu.cz
Subject: Re: Writing Article Re: Linux audio;  also have new hardware support
In-Reply-To: <Pine.LNX.3.96.990204103122.8890D-100000@entry.jcu.cz>
Message-ID: <Pine.GSO.4.05.9902041333060.1917-100000@keg.cs.vu.nl>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: alsa-devel@alsa.jcu.cz
Sender: alsa-devel-owner@alsa.jcu.cz
Precedence: list

hmm, something went wrong with the previous mail...

On Thu, 4 Feb 1999, Jaroslav Kysela wrote:

> I don't expect very much improvement. We can currently with good hardware
> (PCI soundcards) reach 1.5ms latency.

I just did some tests on kernel 2.2.1 and ALSA CVS. Somethings very wrong,
at least on my box. The latency program reports on average an optimal
fragment size of 1216! Configuring AlsAplayer to use 512 byte fragments
and 4 frags total makes it almost unusable (every 2 seconds underrun). I
traced these underruns back to disk access! I.e. if I dd the file to
/dev/null beforehand there are no more underruns. Using a fragment size
less than 512 constantly produces underruns even with the feeding process
at SCHED_FIFO and max priority. This could be a problem with my system only,
so can someone else please run the latency program (in alsa-lib/test/)
on kernel 2.2.1 and the latest ALSA? Thanks..

Andy
--
Andy Lo A Foe, arloafoe@cs.vu.nl |   Linux v2.2.1   | PII 266 
http://www.cs.vu.nl/~arloafoe    | GSM: 0622-440695 | Bebox 133
http://orbital.xs4all.nl/andy    |   ICQ: 7472949   | IRCNet: Adnans


