4 Jul
2013
4 Jul
'13
9:01 p.m.
On Thu, July 4, 2013 9:40 pm, Jeremy Jongepier wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 07/01/2013 05:16 PM, Patrick Shirkey wrote:
Have you been using this for testing at low latency?
On Tue, July 2, 2013 12:44 am, Jeremy Jongepier wrote:
jeremy@rk3066:~$ cat /proc/asound/cards
0 [RK29RK1000 ]: RK29_RK1000 - RK29_RK1000
RK29_RK1000
1 [HDMI ]: ROCKCHIP_HDMI - ROCKCHIP HDMI
ROCKCHIP HDMI
2 [Device ]: USB-Audio - Generic USB Audio Device
Generic USB Audio Device at usb-usb20_host-1.1,
full speed
'Device' is a $2 USB audio interface with a C-Media chipset. On 07/01/2013 04:12 PM, Patrick Shirkey wrote:
Does this happen with or without wifi enabled?
IIUC the potential issues with period sizes with
non powers of
two are at the application level. JACK is agnostic about it.
Yes, guitarix for instance doesn't like period sizes that are not
a power of two.
Anyway it's a little spooky that you get
better performance
with -p64 than -p128 even in playback only mode.
Well I'm testing and haven't done any real stress-tests yet.
Maybe it was a one-off. At least I get undistorted sound out of
this device. Now if I could only figure out why JACK bails out
every once so often. And why full-duplex doesn't work, well, it
works but the audio
coming out is distorted.
System load causing distortion on the Audio device in duplex mode.
It suggests something is wrong with the bus and or interrupts but I
don't see why it would be ok in playback only mode. Not sure if it
is related but I have noticed that my old usb quattro often takes a
while to warm up. Sometimes I have to run it for a few hours and
start/stop jack several times to get undistorted output.
Another issue is the Rockchip devs probably haven't tested the
device at low latency with duplex mode so they haven't picked up
that issue. We can try flagging it with them.
I forgot what the audio chipset is. Can you post the output of cat
/proc/asound/cards?
But I got
one step further:
jeremy@rk3066:~$ uname -a
Linux rk3066 3.0.36-rt58 #1 SMP PREEMPT RT Thu Jul 4 13:18:23 CEST
2013 armv7l GNU/Linux
So I've got a RT kernel running on it. Moving on :)
Nicely done!
What was the blocker in the end?
--
Patrick Shirkey
Boost Hardware Ltd