From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S265863AbTL3RBc (ORCPT ); Tue, 30 Dec 2003 12:01:32 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S265864AbTL3RBc (ORCPT ); Tue, 30 Dec 2003 12:01:32 -0500 Received: from wsip-68-99-153-203.ri.ri.cox.net ([68.99.153.203]:19347 "EHLO blue-labs.org") by vger.kernel.org with ESMTP id S265863AbTL3RBa (ORCPT ); Tue, 30 Dec 2003 12:01:30 -0500 Message-ID: <3FF1AF65.4010107@blue-labs.org> Date: Tue, 30 Dec 2003 12:01:25 -0500 From: David Ford User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6b) Gecko/20031210 X-Accept-Language: en-us, en MIME-Version: 1.0 To: John M Flinchbaugh CC: linux-kernel@vger.kernel.org Subject: Re: 2.6.0: alsa, esd, mpg123 References: <20031230155358.GB23963@butterfly.hjsoft.com> In-Reply-To: <20031230155358.GB23963@butterfly.hjsoft.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Unless esd has ever gotten fixed, esd doesn't handle errors or write() return values. Some sound cards can't handle the flood of data and only accept so many bytes at a time. Esd does a write of 4K, only 2387 bytes (this is an example) are actually written. Esd doesn't pay attention to this and starts writing with the next 4K chunk. This leads to the skips. On slow machines or slow kernels you don't notice this (as much), on faster stuff, it's very apparent. For what it's worth, there is hardly any error checking at all in esd. David John M Flinchbaugh wrote: >on my debian (unstable) laptop newly running 2.6.0, i've noticed >an irritating tendency for music to not pause, but instead to >try to go too fast, skipping small parts of the song (fractions >of a second). this results in music with regular beats sounding >erratic. > >i'm using gqmpeg -> mpg123-esd -> esd -> oss -> alsa (maestro3). > >switching esd to use -tcp instead of -unix seems to alleviate >the trouble a bit. ogg123 playing through esd doesn't seem to >do it as much either. > >has anyone else noted this problem and tuned it away? thanks. > >