Re: PulseAudio (was Re: Sid Foibles)
On Mon, 2014-06-09 at 18:56 +0300, Andrei POPESCU wrote:
> On Lu, 09 iun 14, 09:30:01, Steve Litt wrote:
> >
> > I look forward to hearing how other people do or don't work with
> > PulseAudio (and ALSA) in this thread.
>
> I'll try to explain it simply, but I have a feeling this will turn out
> quite long:
>
> ALSA (Advanced Linux Sound Architecture) is the kernel driver part plus
> a library. In an ideal world it would be all you need to have sound,
> but:
>
> - by default ALSA would not do software mixing (and most consumer grade
> sound cards don't have a hardware mixer)
> - the dmix plugin had (still has?) various quirks and limitations and
> for a long time was not enabled by default
>
> Because of this in addition to ALSA on most systems one would also need
> a sound server to do software mixing: aRts for KDE, esd for Gnome, JACK
> for profesional audio (Ralf's domain).
>
> Unfortunately at this point http://xkcd.com/927/ happened, so instead of
> taking one (JACK would have been a good idea probably) and improving it
> to do whatever was missing some guy(s) thought it would be a good idea
> to create a new sound server (pulseaudio).
>
> The good:
> - aRts and esd died
> - most (all) user applications now have native pulseaudio support, so
> the fake ALSA sink that is used for programs without pulseaudio
> support will probably go away soon.
> - pulseaudio has interesting additional features, some of them even
> useful >:)
>
> The bad:
> - it adds complexity, especially now that ALSA + dmix mostly works for
> the common use (have two applications play sounds at the same time)
> - when it doesn't work it's difficult to find out why.
> - it's not performant enough to also replace JACK. Professional audio is
> probably not pulseaudio's goal, but if we've got to have a sound
> server I'd rather we had only one that works properly for all use
> cases.
>
> The ugly:
> - sound not working seems to be happening especially when pulseaudio is
> installed afterwards (e.g. as dependency of a new application) as
> opposed to new installations. Since removing it quite often magically
> makes sound work again not many people bother to understand what
> happened or even just file (useful) bug reports.
>
> Hope this explains,
IMO your "Alberto Grimaldi" explanation is ok ;). I'm biased ;).
Reply to: