sound subsystem

Havoc Pennington hp at redhat.com
Mon May 31 20:31:43 UTC 2004


On Mon, 2004-05-31 at 15:36, Mariano Draghi wrote:
> I'm using software mixing w/ALSA here. Haven't tested it that much, but 
> so far so good.
> And it takes only a ~/.asoundrc file.

Ah geez, there's a .asoundrc... and the file format looks like some
made-up custom hack... grumble. ;-) Hopefully not an indication that
someone is punting things that should just work onto the end user.

> Well, someone posted a few days ago a tip on the fedora-list on how to 
> make the software input & output devices the default.
> That's the one I'm using, and as long the applications are configured to 
> use ALSA, they end using the sw mixing. I tested it w/ XMMS, Xine and 
> the system sounds in Gnome, it works and I didn't have to tweak anything 
> but just the ~/.asoundrc.

We really have to make the default whatever _works_, end user
configuration requirements in this area are pretty much unacceptable...
if software mixing is the only way to get mixing 100% of the time, and
we want to avoid the software mixing when hardware mixing is possible,
alsa will just have to be smart enough to auto-determine when hardware
mixing is possible and automatically fall back to software otherwise. Or
some other sane automatic policy.

In the meantime can we just pop the systemwide default over to software
mixing?

Havoc






More information about the Fedora-desktop-list mailing list