Hi,
> Here at home I have a Creative Audigy which one I guess
> have hardwaremixing capability
I don't know if it is able to do hardware mixing.
> but at my job I have a
> nForce2 based onboard soundcard which one I'm not sure if
> have hardwaremixing.
I guess it is a device compliant to the AC '97 standard, and
if so it isn't.
> I did a test killing all process of
> esd and arts and trying to play more than one instance of
> music playing and that's works very well with Audigy but
> don't with nForce2 onboard audio.
This confirms my thought above.
> An other thing that I've noticed is that skype just work
> without arts or esd, so I've to run "skype oss" to be able
> to use it (at my home machine, since I don't use skype at
> job yet).
Hm, this confuses me a bit. What does the option OSS change in
the Skype behaviour?
I just found out that Skype 1.0 can work on arts, esd and ALSA
respectively OSS. So running arts or esd will cause Skype to
automatically connect to it. If you run arts, then please
ensure to enable full duplex for it via the KDE control
center. Per default, it works on playback only.
> I think that at home I do not need any of this daemons but
> at my job I will give dmix a try, so I have to ask you if
> even with dmix I'll need to use arts and/or esd?
This depends. Why? Difficult to understand.
An audio developer has to choose at least one audio output
system, may it be OSS direct device access, ALSA direct
device access, an soundserver like esound, arts, gstreamer or
JACK. And if the developer has lots of time and patience, he
makes the application's audio output based on plugins so it
can output sound to various audio subsystems via the plugins.
Have a look in the xmms preferences; it can output audio via
JACK, esound, arts, ALSA or OSS.
As soon as you configure DMIX on a card which is not able to
do hardware mixing, then DMIX does software mixing for any
application which wants to output sound to the ALSA device
directly.
OK, so you configure DMIX and stop all sound deamons. Multiple
apps which want to access the only available ALSA device will
now be able to output sound.
But what about applications which need arts, esd or gstreamer?
These will fail. So, if you want to be failsave you can now -
thanks to DMIX - run all soundservers simultaneously: arts,
esd, gstreamer and JACK. Additionally, any application which
cannot output sound via a soundserver but requires ALSA
direct access can play through the DMIX plugin, regardless
that there are already 4 soundservers running on the same
device. But I guess this costs some CPU power.
So if I only wanted to run xmms and Skype at the same time,
I'd run arts or esound depending on the desktop environment I
run and then I'd make xmms outputting audio via the running
soundserver. As soon as you run Skype, it will autodetect the
soundserver running and connect to it. Remember to enable
your soundserver to work in full duplex mode.
Phew, a long mail. Concerning audio on Linux, it's still a
difficult thing, but if you know what to do, you can get
almost everything to work ;-) .
Best regards
ce
--
gentoo-user@gentoo.org mailing list
