[LAU] perhaps why some of us have more trouble w/ pulseaudio than others (ICE1712/M-audio delta problem w/ pulseaudio)

Paul Davis paul at linuxaudiosystems.com
Wed May 12 13:14:53 UTC 2010


On Wed, May 12, 2010 at 8:58 AM, Philipp Überbacher
<hollunder at lavabit.com>wrote:

>
> I simply don't think that it can do the mobile thing and the desktop
> dance at the same time, since it is still very different.
> Mobile can be a lot simpler in many ways. You pretty much only ever need
> 2 channels, it doesn't need to be flexible, latency doesn't matter at
> all, it just needs to be there and not get in the way CPU, memory and
> power wise.


this is a misleading description. Mobile audio is really quite complex
indeed. You have a continual need to be aware of jack-sensing status, and be
willing to switch the routing of audio from one output to another at
arbitrary times and for arbitrary reasons. You have to consider pausing
streams midway to allow for other streams (e.g. a call in the middle of
music playback). Depending on the device, you may want to be doing so-called
"gapless" playback of music on the device, which implies a moderately
complex disk/memory buffering scheme, which only gets more complex if you
add crossfading. You may also choose to mix together a steady state stream
(eg. music) with alerts (e.g. new message). You may want to switch from high
latency during music playback to low latency during a voice call.

Note that all of these issues are much, much close to desktop audio, which
is why Nokia, Intel and others seem reasonably happy to use PA for this.


> PA tries to do all of this at once, and at least on the desktop it fails
> often enough. Sure, many user complaints are old and the problems might
> be solved, but many users are fed up with it already and don't ever want
> to try it again.


The users that I have seen who are fed up with are generally people who are
trying to get JACK and PA to work on the same system with only a single
audio interface. The impression that I have (and it may be wrong) is that
users who don't try to do this are reasonably happy with PA to the extent
that ALSA works correctly for their Intel HDA chipset. They get features
that people have wanted for a long, long time (device sharing, per-app sound
control, switching output based on jack-sensing status, on-the-fly device
switching and more) and most of this stuff works really well. The headaches
seem to come mostly from the same place that a lot of JACK user complaints
come from these days: poor/incomplete/hard to use HDA driver support.

There's a fun German expression that fits PA quite well:
> "Eierlegende Wollmilchsau"
>

yes, this was my favorite german expression that i learned while living in
berlin. i don't think it applies to PA though. Its happy to be a
wollmilchsau and forego the eirerlegende attributes :)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxaudio.org/pipermail/linux-audio-user/attachments/20100512/8c4398ef/attachment.htm>


More information about the Linux-audio-user mailing list