From alsa-devel-owner@alsa.jcu.cz  Fri Nov 27 01:27:57 1998
Received: from entry.jcu.cz (IDENT:perex@entry.jcu.cz [160.217.1.111])
	by marvin.jcu.cz (8.9.1a/8.9.1) with SMTP id BAA31626;
	Fri, 27 Nov 1998 01:27:29 +0100
Date: Fri, 27 Nov 1998 01:27:28 +0100 (CET)
From: Jaroslav Kysela <perex@jcu.cz>
To: Benjamin GOLINVAUX <golinvaux@benjamin.net>
cc: alsa-devel@alsa.jcu.cz
Subject: Re: Linux sound cards latency
In-Reply-To: <002c01be197f$5163f980$34934ac1@w98bg>
Message-ID: <Pine.LNX.3.96.981127011353.4393I-100000@entry.jcu.cz>
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

On Thu, 26 Nov 1998, Benjamin GOLINVAUX wrote:

> Hi All...
> 
> While trying to write a paper on my website to prove Linux can be an option
> for HDR, I wish to know the latencies ALSA architecture achieves on various
> soundcards to compare them with Windows drivers for the same cards. If some
> of you have measured such latency, please let me know (thru this mailing
> list is ok).
> 
> I pardon you to steal some of your time, but it is for the sake of free
> software vs buggy cute unprofessionnal sequencers/HDR.
> 
> By latency, I mean the time needed to echo a signal on a card's input to the
> output, without any further treatment. Please let me know if there are
> additionnal parameters....

OK. I created simply program which measure latency for ALSA driver. This
program is available on cvs in alsa-lib module in test directory -
latency.c.

Code should be better, but it could show some results. I tried this code
with Turtle Beach Malibu and Ensoniq AudioPCI ES1370. Latency is measured
between time in which driver starts record and time when playback is
started (ALSA driver allows get this time over status structure)..

For CS4237B chip (Turtle Beach Malibu) I get various times between 1-7ms
(Round Robin schedule policy, maximum priority 99). For Ensoniq AudioPCI
is time always about 3ms, but seems that this time is stable if scheduling
policy isn't set, too.. Seems that top limit should be under 10ms..

Maybe someone other should do another tests and/or extend my example...

							Jaroslav

-----
Jaroslav Kysela <perex@jcu.cz>
Academic Computer Centre, University of South Bohemia
Branisovska 31, C. Budejovice, CZ-370 05 Czech Republic


