PulseAudio Ported To Android

[Alexandru Stoian]

Well as the title says , Pulse Audio has been ported to android. Maybe
this would mean less audio issues if used by android x-86. More info
here : http://www.phoronix.com/scan.php?page=news_item&px=MTA0MzY and
here : http://arunraghavan.net/2012/01/pulseaudio-vs-audioflinger-fight/
.

[Ikem Krueger]

> 2012/1/16 Alexandru Stoian <lgst...@gmail.com>:

> Well as the title says, Pulse Audio has been ported to android.

> Maybe this would mean less audio issues if used by android x86.

Pulse Audio means additional cpu overhead/audio latency.

I don't know how that will help with audio issues..

Nevertheless, interesting links.

Thanks for that.

--
My blog: http://linux-workers.posterous.com/

[Ikem Krueger]

> 2012/1/16 Alexandru Stoian <lgst...@gmail.com>:

> Well as the title says, Pulse Audio has been ported to android.

> Maybe this would mean less audio issues if used by android x86.

Pulse Audio means additional cpu overhead/audio latency.

I can't imagine how that will help with audio issues..

[Nathaniel Husted]

The original announcement  discussed Pulse as a replacement for AudioFlinger showing performance was equal, memory usage slightly higher, and battery life far greater. These, perhaps, have a far greater effect on the ARM/phone platform than they do x86. Food for thought and great work in general.

--
You received this message because you are subscribed to the Google Groups "Android-x86" group.
To post to this group, send email to andro...@googlegroups.com.
To unsubscribe from this group, send email to android-x86...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/android-x86?hl=en.

[Ikem Krueger]

> 2012/1/17 Nathaniel Husted <nhu...@gmail.com>:

> announcement discussed [..] AudioFlinger

> Pulse as a replacement

> performance was equal,

> memory usage slightly higher,

> and battery life far greater.

But you need to compare that with plain ALSA..

[Ikem Krueger]

> 2012/1/17 Nathaniel Husted <nhu...@gmail.com>:

> Pulse [Audio] as a replacement for AudioFlinger

> performance was equal,

> memory usage slightly higher,

> and battery life far greater.

I believe that has more todo with ALSA then with Pulse Audio.

[fuzzy7k]

I think ALSA is irrelevant here. If I'm not mistaken, stock android
uses some lite form of ALSA, with AudioFlinger on top. Regardless of
the stock case, A-x86 uses ALSA and AudioFlinger, just look at your
logcat.

On Jan 17, 11:38 am, Ikem Krueger <ikem.krue...@googlemail.com> wrote:
> > 2012/1/17 Nathaniel Husted <nhus...@gmail.com>:

[Chih-Wei Huang]

Yes, ALSA is the kernel driver layer, while
PulseAudio and AudioFlinger are user space program
that handle audio data in a higher layer.
They still depend on the underlying driver layer
(ALSA or something else) to drive the audio chip.

On arm platform, some android phones use ALSA,
but some use private api. On Android-x86 we always
use ASLA, the only standard kernel api for most platform.

2012/1/18 fuzzy7k <kva...@gmail.com>:

--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[Ikem Krueger]

> 2012/1/18 Chih-Wei Huang <cwh...@android-x86.org>:

> Yes, ALSA is the kernel driver layer,

> PulseAudio and AudioFlinger are user space programs

What I meant was, that it's probably performanter to bypass
AudioFlinger and PulseAudio at all.

--
My blog: http://linux-workers.posterous.com/

2012/1/18, Chih-Wei Huang <cwh...@android-x86.org>:

> --
> You received this message because you are subscribed to the Google Groups
> "Android-x86" group.
> To post to this group, send email to andro...@googlegroups.com.
> To unsubscribe from this group, send email to
> android-x86...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/android-x86?hl=en.
>
>

[Ikem Krueger]

> 2012/1/18, Chih-Wei Huang <cwh...@android-x86.org>:

> Yes, ALSA is the kernel driver layer,

> PulseAudio and AudioFlinger are user space programs

I think it's faster to bypass AudioFlinger and PulseAudio at all.

And use ALSA directly.