wolftune wrote:Alesis io2 / io4 are totally fine 24-bit and absolutely certain to be Linux compatible because it is USB 1.1. class compliant, no drivers needed. So to put it simply, there is absolutely no issue of Alesis working with or not working with Linux, the devices are just standardized and require no special drivers at all.
Wolf, that can't be so. Usb-audio 1 class compliance does not allow 24-bit audio. It's technically not possible. Only usb-audio 2 compliance allows that. If the Alesis does 24-bit audio under Linux, then one of two things
must be true:
1) The alesis is usb-audio 2 compliant, not usb-audio 1 compliant. This would be the first time I've heard that. I long ago read the io2 and io4 user manuals, and I definitely did not note any verification of 2 compliance. I would have definitely noticed that (like I did when I read RME's literature on the UCX).
2) Someone did in fact add alesis proprietary support to ALSA's usb_audio.ko driver. I just peeked at the source for ALSA's kernel/usb/quirks*.* drivers, which is where you'd find any evidence of proprietary usb support. For example, that's where you'll see the device specific support for various Edirol devices such as the UA-25ex, and the M-audio Fast Track pro, and other such proprietary "Special" non-standard support for devices which we have verified do indeed go beyond usb-audio 1 compliance under linux. There is no entry for the alesis io2 nor io4.
So I have to ask -- Can you verify if one of the above is true? Do you yourself own an io2 or io4, and can do something like grab a 24-bit wave file and fire up alsaplay to test that it does in fact support this? Are you suuuuuuuure we're not passing on some info we "heard" somewhere, but is actually based upon false assumptions (maybe from feel-good FAQs that merely say "unless someone tells me it doesn't work, I'll just assume otherwise") rather than bonafide testing? Are we sure we're
not setting up someone to get burned by Linux?
Because I gotta say, right now I'm really uneasy about letting him believe the alesis will do 24-bit under linux today. Incidentally, there are no entries for any lexicon device in ALSA's quirks either. And I long ago read that lexicon's online manual, and it definitely didn't mention usb-audio 2 compliance. (I'm not kidding around when I tell people I've done extensive, long-term research into linux audio support, with an eye toward everything I could uncover). And so far I'm the only guy to warn him exactly why his lexicon assumption was probably incorrect (and which I feel fairly confident is indeed incorrect). But unfortunately that bit of info has moved to another thread, and now he's left with only "Don't worry, be happy" responses that seem to be based on... well, no actual verified testing.
I'm worried this poor guy is about to get burned by yet more "linux audio misinformation". I'm not implying that anyone is doing that deliberately or maliciously. But I'm getting worried it may be happening.
So... anyone got any hard evidence to alleviate my apprehension?