Hey Jaime,
On 24/01/14 01:48, Jaime E Oliver wrote:
there was a thread on this just a while back and someone reported multiple inputs, I cannot find it right now, but it;s there!
I spent a fair bit of time reading back through the archives and various other sources yesterday and I did see mention of someone running four inputs but a later post seemed to contradict what they'd said, and I couldn't find anything on pd-list about that in particular.
Would love to be wrong about that of course!
In any case, I will report back here if successful.
Cheers,
Chris.
I don't want to discourage you, but from my experience there's definitely some things to be wary of with the Raspberry Pi. For example, a lot of people have reported needing to use USB 2.0 devices in USB 1.1 mode. I don't know if this would necessarily hinder the functionality of the device, but it seems like it could be a nuisance.
Personally, I've been using an interface with a single stereo input and output (Behringer UCG102) for my guitar. Performance for that device has been fairly satisfactory, but there are some problems I can't seem to resolve...
I get buffer xruns every ten seconds or so when running PD, even in -nogui mode with realtime priorities. This sounds like a little "pop," a one sample impulse, and while it's not unplayable, it is annoying. I am running at a pretty low latency (~8ms, which is awesome), but even raising the latency a little and messing around with the signal vector size doesn't seem to help stop the popping.
Also, while it's not super important for my guitar processing, I've been unable to run duplex audio at a samplerate higher than 32kHz. It seems that this is related to an issue in newer versions of the kernel (there's an issue posted here: https://github.com/raspberrypi/firmware/issues/197). I don't know anything about Operating System code, so I have no idea if this could be addressed somehow.
So while I can't say that multi-channel input and output won't work, I will say that even stereo I/O is a little broken.
I wonder if the BeagleBone xM would be better for this project? It is a bit pricier, but it seems like Stanford CCRMA students have had some success with it (more than the Pi, anyway) Check out this link for examples of that: https://ccrma.stanford.edu/~eberdahl/satellite/
Good luck,
Brian
On Thu, Jan 23, 2014 at 7:31 PM, Chris McCormick chris@mccormick.cx wrote:
Hey Jaime,
On 24/01/14 01:48, Jaime E Oliver wrote:
there was a thread on this just a while back and someone reported
multiple inputs, I cannot find it right now, but it;s there!
I spent a fair bit of time reading back through the archives and various other sources yesterday and I did see mention of someone running four inputs but a later post seemed to contradict what they'd said, and I couldn't find anything on pd-list about that in particular.
Would love to be wrong about that of course!
In any case, I will report back here if successful.
Cheers,
Chris.
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list