Also, there are more options available via Firewire with FFADO, including many companies whose hardware is absolutely not compatible any other way. PCI cards from RME are still the most prefered way though. I don't know about USB -- up until lately, most swore it was a great way to invite problems, but it seems not quite as discouraged lately, so maybe it's gotten better. I'm using an RME Multiface II with PCI-E card here, and it's fantastic. I just wish I had another converter with an ADAT output that I could hook up to it to get even more i/o. -- + Brent A. Busby + "We've all heard that a million monkeys + Sr. UNIX Systems Admin + banging on a million typewriters will + University of Chicago + eventually reproduce the entire works of + James Franck Institute + Shakespeare. Now, thanks to the Internet, + Materials Research Ctr + we know this is not true." -Robert Wilensky

Hi, Quoting Dan MacDonald <allcoms(a)gmail.com>om>: I was testing an RME fireface ucx in class-compliant mode on linux. I had to modify the file mixer.c and compile "my own" ALSA-part of the kernel. As far as I know, the alsa developers have been working on this issue but the patch is not yet in the official ALSA kernel. However, I was able to run 8 in- and 8 out-channels through JACK with a sample rate of up to 96 kHz. I think that this is actually the best you can get with usb-audio 2.0, full support of all features and in particular 18 in- and 18 out-channels would require a different driver. I tested the quality of signals with SuperCollider, I basically got the quality which is also obtained using RME's driver on a mac. Best, Florian

Hi Ralf, you are right, I should have been a bit more precise. First of all, it is not my ff ucx, I was lucky that I could borrow one from a friend for two weeks. During that time, I managed to get help from the ALSA community and fix mixer.c and at the end of that two week period, I could do some testing. Right now, I do not have access to a ucx and can not do further testing but I think I will get my own device after xmas. I hope that I can do proper testing then and send a "report". My tests so far included the following: As I already mentioned, I only used the class-compliance mode. Up to my knowledge, usb-audio 2.0 is limited to 8/8 in/out-channels and sample rates up to 96 kHz. ALSA driver recognizes these 8/8 channels. The device moreover appears in the JACK-setup and I was to start JACK-server running 8/8 channels with sample rates up to 96 kHz. I played back soundfiles through the ucx without any problems. I moreover send sine-signals with different frequencies from supercollider to all the 8 channels and these correctly appeared at the 8 jack-outputs (which are on the backside of the device). Finally, I made loopback recordings from pulses, sine-waves and triangular waves also using supercollider and different jack blocksizes / sample rates. The latency I read from the pulse-loopback records was 7 to 8 ms (blocksize: 64, samplerate: 42100/s). The sine- and triangular-loopback-recordings show clean signals, only a very close analysis of the triangular signals shows a slight distortion but this also appeared when the card was connected to a mac computer using RME's proprietary driver. Maybe the cause was actually the capacitance of the low quality cables used for the loopback, I don't know. I certainly did not test all the features of the ucx, it is not possible to access all of the channels and I also do not know how to access TotalMix and the build-in DSP-features, most likely it is impossible in the class-compliant mode. I hope this description is somehow clear, Best, Florian Quoting Ralf Mardorf <ralf.mardorf(a)alice-dsl.net>et>: