[pulseaudio-discuss] Problems with mplayer in Ubuntu 10.04
gmane at colin.guthr.ie
Sun May 9 04:23:14 PDT 2010
Just for clarity to others reading, we discussed and tried to debug this
issue a lot on IRC already, so I'm familiar with the problem.
'Twas brillig, and Mihai Sucan at 09/05/10 12:06 did gyre and gimble:
> 4. somehow i think that mplayer -ao pulse does "pretty much the same" as
> it does mplayer -ao oss from the rvm Ubuntu PPA. The latter did take
> control of the sound card, via OSS. The former does somehow take control
> of the sound card, pulseaudio looses it, then mplayer happily goes on to
> use pulseaudio for subsequent audio stream output, but now ... pulseaudio
> is silenced.
Well mplayer -ao oss will bypass pulse completely and, as you say,
basically "hog" the sound h/w. You can theoretically use:
padsp mplayer -ao oss
and that should redirect oss output via PA too, but this may or may not
work overly reliably (aka YMMV).
The interesting thing about the report is that when PA stops outputting
audible sound, the vumeters are properly moving up and down which means
that the sinks are not suspended, nor are the sinks unloaded and
replaced with a "Dummy Output" (null-sink) because they've been
unloaded. In other words everything seems setup to work correctly but
the reasons are eluding us!
One thing you don't mention above, but is included in your debug output
is the fact that the alsa mixer output is not any different before vs.
after the problem. My initial hypothesis was that mplayer was somehow
flipping the digital output switch in alsa, thus causing it to seem like
it had simply stopped working. However analysis of the amixer output
seems to kill this idea.
So if anyone else has any bright ideas, please speak up!
Tribalogic Limited [http://www.tribalogic.net/]
Mandriva Linux Contributor [http://www.mandriva.com/]
PulseAudio Hacker [http://www.pulseaudio.org/]
Trac Hacker [http://trac.edgewall.org/]
More information about the pulseaudio-discuss