No thanks. The UA-25 is bus powered, has two XLR+jack inputs, phantom power, rca/jack outputs, direct monitor, individual channel gain control, master gain control, and in a road tuff metal case. I have two of them, so it makes sense to dump the PI if I can't get it to work with proven linux friendly hardware especially when the pi costs less.

Too bad. Back to iPad now ... :D
 
On May 4, 2013, at 12:42 PM, Antoine Villeret <antoine.villeret@gmail.com> wrote:

ok bad news...

what about using another sound card ?
UGM6 has 2 hi-Z input (but without no phantom) and 2 outputs
it may fit your needs

+
a

--
do it yourself                      
http://antoine.villeret.free.fr


2013/5/4 Dan Wilcox <danomatika@gmail.com>
Yes. Ethernet turbo disabled and Ethernet disabled. Disabled cpu freq scaling. Pd set with realtime permissions:

sudo chmod 4755 /usr/bin/pd-extended
sudo chmod 4755 /usr/bin/pd

Setting dwc_otg.speed=1 dosen't work for this card and yields an unstable system.

This is pretty disappointing as my old setup was on an embedded 500Mhz Celeron with half the RAM and almost no video memory, yet (so far) was running Pd + alsa better than the superiorly specced PI. The main reason I'm not using the wearable anymore is it's just worn out and trying to fix flexxed pcbs and the resultant random glitches was getting old years ago.

On May 4, 2013, at 12:31 PM, Antoine Villeret <antoine.villeret@gmail.com> wrote:

Hi dan, 

Did you try to disable Ethernet turbo or to disable Ethernet at all as describe on linuxaudio.org [1] ?
it gives me good result with ESI sound card (UGM-6, see my post on the list from Thursday).

Cheers

a

--
do it yourself                      
http://antoine.villeret.free.fr


2013/5/4 Dan Wilcox <danomatika@gmail.com>
For what it's worth, output did work using: 

pd-extended -nogui -noadc -audiooutdev 3 test.pd

with 3 being the direct alsa interface as opposed to the alsa plug interface.

It's when I enable the audioindev that I get crackles. I know there are people out there using the UA-25 + jack, but I never needed jack before with an embedded setup.

Part of the issue might be the RPI's real-time kernel. My old wearable setup was far more unstable with a realtime kernel and using a regular kernel with realtime permissions for PD worked far better.

On May 4, 2013, at 11:52 AM, Miller Puckette <msp@ucsd.edu> wrote:

HI Dan -

I tried and got even less far than you apparently did, and gave up.  It's too
bad - the UA 25 is the best USB-powered interface I've found so far.  For the
Pi I now use small, cheap, recent-vintage ones like the Griffin iMic. (but
there are even cheaper ones :)  If you have low-impedance mics, my best
guess would be to use an impedance matching transformer (lo to hi) and then
a "PC mic" (high impedance) interface.

cheers
M

On Sat, May 04, 2013 at 10:57:35AM -0400, Dan Wilcox wrote:
Howdy everyone,

Anyone had luck with a Roland/Edirol UA-25 with PD on a RaspberryPi? I've gotten it working fine with output, but enabling input gives me crackles and bad cpu usage. Ideally, I want 2 in / 2 out and this audio interface has worked wonderfully with Linux on older/slower machines in the past. The RPI should have no problem spec-wise to run this with PD using only alsa ...

--------
Dan Wilcox
@danomatika
danomatika.com
robotcowboy.com






_______________________________________________
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list


--------
Dan Wilcox
@danomatika






_______________________________________________
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list



--------
Dan Wilcox
@danomatika







--------
Dan Wilcox
@danomatika