Linux USB audio soundcard question/rant.

[cdev]

How would an external USB sound card (Via vt1620a based) that is advertised in the chipset data sheet as supporting 192 KHz sampling, but which only enumerates as 96 kHz capable, be told, under Linux to use the full 192 KHz?

Basically that would enable a doubling in SDR bandwidth with this very cheap but good performing external sound card - that would likely please a lot of people.


Currently its working great, though at 24 bits@96 kHz.

[cdev]

I thought I should post the quick and easy way I am now using to route audio out of one program into another here. This is the least finicky and also least complicated way Ive found to date. (knock on wood) For more see the z_ALSA.txt file in the linrad package.

"The Linux sound system ALSA has a mechanism by which the output of one program can be sent to the input of another program. It is similar to VAC (virtual audio cable) under Windows. By installing snd-aloop one will get virtual soundcards that can be opened as the output of one program and as the input of another program. Unfortunately this feature is not (yet?) enabled in the standard distributions."

(Note: it is standard on many now)


Make sure the snd-aloop.ko module is loaded at boot time.

Edit or create a file /etc/modprobe.d/alsa-base.conf At the very end of the file add something like the following lines

options snd_xxxxxxx index=0   #(the name of your primary sound card's kernel module)
options snd_yyyyyyy index=1   #(the name of your secondary sound card's kernel module, if you have one)
options snd_aloop index=2         #(the loopback module which is effectively a digital patch cable)



Edit the file /etc/rc.local and add a line:

modprobe snd-aloop       

then reboot your system. You should then be able to select the loopback module.

[HAL-42b]

I'm no expert but there might be some config file you need to edit before Linux starts sampling at 192KHz. For example is your user member of the 'sound' group?

There are fiddly details like that. Read here https://wiki.archlinux.org/index.php/ALSA

Btw, the 'pro' audio system on Linux is JACK. It should offer better realtime performance at expense of processing power.

[cdev]

Thank you for the link. Arch's documentation is so much better organized than many others. It makes me want to be using it.

All told I would rather have the Linux sound ecosystem, chaotic as it is, than to have the system be a black box that wasn't accessible to the user.

[HAL-42b]

Arch's documentation is so much better organized than many others. It makes me want to be using it.

That's how I started. Arch is the perfect learner distribution.

[MrSlack]

Until it breaks. Which it does regularly.

Source: juggler of about 200 Linux machines daily amongst other hats.

Regarding the original issue, it depends if the driver implements that sample rate or knows how to configure it. Typically some cheap as chips codecs are issued with different capabilities with the same PCI vendor and device IDs. The commerical windows driver will know how to read some undocumented magic flag from a register. The Linux ones tend to take the safe option and cap it at the lowest capability. Then there's the particular audio subsystem you're using and the various quirks. TBH I wish the whole Linux audio architecture would die in a fire. NT since Vista is fsr superior!

Also, VIA. Ick. Notorious turd emitters.