Re: [Jack-Devel] How to get mplayer and firefox/flash to play

On 12/20/2011 09:12 PM, Fons Adriaensen wrote:
> On Tue, Dec 20, 2011 at 10:02:03AM -0800, Robert M. Riches Jr. wrote:
>  
>> BINGO!!!  The ALSA loopback page is a winner!  At only 1% CPU
>> each for alsa_in, alsa_out, and jackd, the CPU load is definitely
>> tolerable.  Some time, I'll have to study up on ALSA loopback to
>> see whether I can have multiple loopback cards, but a single one
>> will be good enough for now.

Hi Robert,

Yes, you can. Up to SNDRV_CARDS (=32 with CONFIG_SND_DYNAMIC_MINORS else
=8) defined in the alsa kernel source.
Use the snd-aloop module 'enable=1,1,[1|0],..' option.

http://www.alsa-project.org/main/index.php/Matrix:Module-aloop#The_module_options_for_snd-aloop

You may also want to set 'pcm_substreams=2[,...]' to limit the loopback
device to two channels, because e.g. /wine/ currently does not recognize
any sound-card with more than two channels..

> Which still leaves the question why things don't just work
> with the alsa->jack plugin. 

Hi Fons,
Do you really think that this is interesting and important?
Why don't you have a look?

I assume that it's due to the multitude of possible ways that clients
can interact with ALSA: some push, some pull, some request specific
buffer-sizes and/or bit-depth,... others use weird workarounds for
internal [timing] issues: In particular closed source apps like flash,
skype..

Frankly, I think that the alsa-jack plugin was a design mistake to begin
with (sorry Maarten). It was a certainly a cool hack in 2003, but
it has not been fixed in 8 years and I doubt it ever will.

> For one, it doesn't show up in aplay -L or -l. Which probably
> means some apps won't ever use it (anything based on juce for
> example).

Indeed, they're created dynamically.

The loopback device ports OTOH are persistent: The port connections
[with jack] remain, even when the alsa-app goes away. And similarly: the
loopback-device is still there even when jackd is down. Nicely
decoupled. No auto-connect either. ++

The only criticism is that there's re-sampling involved (aloop uses
jiffies as clock source) and that it adds a bit more latency.

I've been thinking to address these issues by patching snd-aloop. Yet
found that it is much easier and more convenient to jackify applications
instead. For skype, flash and f[r]iends those are non-issues anyway.

> OTOH, when I plug in my Edirol USB interface, aplay -L shows
> surround40,41,50,51 devices based on it which is plain nonsense
> as the interface hase just two channels.

I get those for every audio interface, independent of the number of
channels it has.

They're generic definitions that likely come in handy for the multitude
of people who switch between headphones at work and their consumer-grade
surround systems at home. You can change them in
/usr/share/alsa/alsa.conf (or where-ever your ALSA-lib resides).

> All of this just doesn't make any sense to me.

Just ask upstream for the motivation.. run `git blame` on
http://git.alsa-project.org/?p=alsa-lib.git;a=blob;f=src/conf/alsa.conf

ciao,
robin

1324418665.8895_0.ltw:2,a <4EF1064D.6090704 at gareus dot org>