asoundrc configuration file documentation

[Akkara]

At the risk of sounding like a total idiot, where is this thing documented?

I see configurations posted around, like here or here, among many others. And it all looks like black magic to me.

Excuse me while I rant a little: "ttable" - what's that do? Sounds to me like "translation table" i.e., some sort of routing. But then how's that different from input.bindings.0 "Input", which also looks like a routing sort of directive. When would I use one vs the other? Why is slave.channels 3 but yet there are four things in the ttable?

Or better yet, "id 1672 # 4 Pole Low-Pass Filter with Resonance". Oh, but of course! Plug-in number 1672. Why didn't I think of that. Seriously - I have never seen a more arcane, obtuse way of identifying and configuring stuff. Numeric module IDs? I thought that went out of fashion 'round the time of Fortran-77. If it weren't for the comment, there's not the tiniest smidgen of a hint as to what's going on. Does 1672 imply there's a plug-in id#1671? What's that one do? Or #1670? How high do the numbers go? Is there a list somewhere? Is my Google-foo such total fail when it comes to this?

And what's this "policy" thing mean? "Duplicate" I can imagine, but none? That like /dev/null? Is policy required? When? All modules? Just some? What about "type ladspa" - what does this signify? What other types are there? Is there a list? How does one choose? How does one begin to structure a configuration? It all looks like pulling magic words and numbers out of thin air.

...
Thanks for letting me vent. I'd like to learn more. I don't have a specific immediate need so no worries about specific replies. Reading lists will suffice. I've gotten sound mostly to my liking by using jack and avoiding alsa. This means I can't use many programs to output audio but that's a secondary issue for me. Accurate auditioning and mastering are the primary concerns.

The soundcard is a M-audio ice1712 24/96. It shows up as a 10-channel card even though it only has 4 I/Os: a stereo output, a stereo input, and a spdif input and output (and a midi port). And most programs throw their virtual hands up and say "I have no idea how to output to 10 channels" and quit with an error. (Why not default to using the first two?) The card has no "hardware mixing" a-la Creative style that alsa seems to understand, even tho it does have a DSP on board running a digital matrix mixer. I've tried dmix on and off but have never been satisfied with the quality nor the latency of that configuration.

Also: there are times where I _want_ the ability to lock out other programs. E.g., when recording audio, I don't want a web-page to start playing stuff at me.

I generally understand signal processing and I ought to be able to figure it out, if I knew the incantations. But the configuration language is like ancient elder scrolls full of dark magic to me and I'm having the darndest of time figuring out how to get ideas in my head written in a way that alsa will parse and act on.

[aCOSwt]

[PaulBredbury]

[_itsme_]

I was just looking at this yesterday trying to figure out why I couldn't have multiple simultaneous playback streams if ALSA supports software mixing via dmix. The asoundrc documentation is bad and people should feel bad about it.

I found some useful info at these places:
http://www.volkerschatz.com/noise/alsa.html
http://www.alsa-project.org/main/index.php/Asoundrc
http://www.alsa-project.org/alsa-doc/alsa-lib/pcm_plugins.html

Good luck with it. It seems to be mainly copy-paste and hope for the best unless you're a developer on the project who can actually understand this mess.