damien.thebault at laposte.net
Tue Jun 10 18:07:52 CEST 2008
> Hi guys,
> we have been shipping pulseaudio (which is quite a CPU hog on embedded
> systems) on our rootfs for quite a while now. The main reason not to use alsa
> directly was because of mixing, since alsa dmix absolutely does not cut it.
I saw threads about this on the pulseaudio-discuss mailing-list .
I don't know if changes mentioned there are available now, we can read on the XO wiki page about pulseaudio  (linked in ) that:
"Pulseaudio requires FP for resampling calculations. [...]
This has been changed in the 0.9.7 release. Pulse is using the speex resampler by default which allows to configure pulse's resampler to do either float or fixed."
The "Audio Class Policy" thing from  seems great too and it might be nice to have something like that available on OpenMoko.
Currently I don't have any hardware and I didn't start looking at the code, so I don't have any opinion about dropping pulseaudio or not. Maybe when I will be more involved (with a FreeRunner in my hands, for example), I will really look at it ;).
 "[pulseaudio-discuss] PulseAudio for Embedded System"
 "[pulseaudio-discuss] Pulseaudio on the XO"
Créez votre adresse électronique prénom.nom at laposte.net
1 Go d'espace de stockage, anti-spam et anti-virus intégrés.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the community