On 11/27/2008 05:14 PM, Anne Wilson wrote:
On Thursday 27 November 2008 14:16:25 Mary Ellen Foster wrote:2008/11/27 Tom Horsley<tom horsley att net>:On Thu, 27 Nov 2008 13:01:09 +0000 Anne Wilson<cannewilson googlemail com> wrote:as would noting any known work-arounds for problem.Heck, this works around every problem I have every time: yum erase pulseaudio :-).: :-) indeed, but I think that's actually the problem that Anne is referring to: generally if anyone has any problems with pulseaudio, the first advice they'll get is almost always to uninstall/disable it. Sure, that may possibly still be necessary with particularly weird hardware or audio programs, but if pulseaudio truly is the way of the future, it would probably be worth having a page somewhere (maybe it exists?) with simple-to-use troubleshooting techniques that could be used before the above blunt hammer.Exactly. I'll gladly help all I can. If someone feels capable of writing the necessary but is unsure of the English, I'll help with that. I also know a few kde folk that would help with a translation if it is easier to write in one's native language.
What needs to happen is for both projects (ALSA and PulseAudio) to properly recognize that there are interaction issues that they need to work on together. Filing bugs against PA results in a "this is not ours but ALSAs problem" but users cannot file proper bugs for ALSA because they have no idea how PA is actually wired into ALSA. So they switch to plain ALSA output and sound starts working again.
On useful piece of documentation would be a PA developer explaining how exactly PulseAudios usage of ALSA differs from the way applications interface with ALSA directly. That way people could actually file bugs against the ALSA drivers and point the developers toward that document.
Take a look at this bug: https://bugzilla.redhat.com/show_bug.cgi?id=462026This got filed agains alsa-lib, then kicked over to the driver folks who then immediately kicked it back to the alsa-lib folks. The problem seem to occur for at least three different drivers and was attached to the F10Target tracker but clearly didn't make it for the F10 release.
For users there is little more they can do here and none of the development stakeholders really want to "own up" to the problem so it just lingers. Since the involved users clearly want their sound to work they are left with no other choice than turning off PulseAudio.
How is addressing the FUD out there going to fix these real problems? Regards, Dennis