JACK, aRts, esd -> WHAT are those _exactly_ good for?

[Shadows]

Hi folks

Even after reading the FAQs and descriptions from the related project pages, I'm still not sure if I really understand what those software packages are good for?

Well, ALSA is an Hardware Abstraction Layer for audio devices, so, if one would like to output sound to a sound device, one should write an ALSA driver for his application.

Ok.

And JACK, aRts, esd etc. are some kind of "sound servers", making it possible to direct sound to them and eventually do some crazy stuff with the audio streams, like add an echo or simply mix them together, right?

So, without an sound server, it would be like:
app -> ALSA -> sound card

and with sound server:
app -> sound server -> ALSA -> sound card

Right?

Well, basically, I have disabled all those sound servers in my make.profile, since I'm thinking "What do I need those for if my favoured apps have an ALSA output plugin?".
Even all of my KDE apps work without aRts.

So, from a pure user point of view, my questions:
What benefits could I get from one of those sound servers, e.g. if switching my bmp ALSA output to one of those?
Where's the difference between them?

Is it possible to solve the following problem with one of those sound servers, and if so, which one would be the best to choose?
Basic problem is, that I would like to adjust the volume of my multimedia apps independently from each other, for example:
- mute xine while still being able to listen to my skype buddy I'm currently talking to
- only increase the skype volume or my favourite media player
etc.

If it's not possible to do that stuff with one of the sound servers currently available, how else could I achieve those goals? Or is it already possible to do that with ALSA? Oh, yes, and it should be easily configurable via a GUI of course :)


Thx for your help and wisdom :)

Greetz

Shad
_________________
# 1:
PIII 450, Intel 440BX, 448 MB SD-RAM ATI Rage IIC AGP 8 MB, SB AWE 64 PnP

# 2:
Athlon XP M 2800+ Desktop edition (running on 1.666 and 100 FSB) [was: Duron 800 (Spitfire)], Via KT266, 512 MB SD-RAM, GeForce 3 64 MB, Sound onBoard

[Ohnodoctor]

I would also like to know. The only time I ever use a sound server is when I'm streaming audio across the network with esd, which is nice.

[ppurka]

In the programs that you run, just set the audio mixer control setting to "software".

[mnxAlpha]

If you have a sound card with no hardware mixing, and don't have ALSA set up with dmix correctly, sound servers will let you use two programs that generate sound at once. As long as they both use the same sound server for output, of course. Some of them also add some network capability.

[Archangel1]

Software mixing is the big one as far as I can see.
However my desktop has a vaguely decent (ie. not built-in) soundcard, so it doesn't need it - ALSA works just fine.
My notebook could use software mixing, but I don't bother because I don't have one audio stream on it that often, let alone two

I won't touch aRts since it got installed when I first installed Linux; it was a total disaster.
_________________
What are you, stupid?

[Shadows]

[TactX]

JACK is designed for low-latency (professional) audio applications and allows applications to share (audio) data easily. I plan to use JACK for my synthesizer. It also has an easier API than Alsa imho.
_________________
"As your attorney, I advise you to drive at top speed. It'll be a goddamn miracle if we can get there before you turn into a wild animal."

[ph03n1x]