20 Sep
2011
20 Sep
'11
8:32 a.m.
Hi
A friend of mine who would like to use his new UA-25EX with ubuntu on
his (i think) x200 (at least it's a 2.4Ghz dualcore 64 bit thinkpad with
6 or so gigs of ram). A very nice machine, that should be great for
setting up a linux DAW. However I cannot make it run without x-runs (did
the usual limits.conf dance). Whenever a window is dragged, or there's
disk activity or something like that there's an x-run.
He has two ubuntus installed, a 64 bit and a 32 bit (one is natty one is
marverick, forgot which is which). Both systems (mis)behave the same. I
tried running in openbox with pulseaudio killed, I installed the
lowlatency kernel from Alessio Igor Bogani.
I've been testing with ardour (which he's probably gonna use) and
renoise (which I know best).
The soundcard runs great on my dualcore 1.4Ghz x61s with arch.
Any ideas would be highly appreciated.
NB: Exact details regarding software or hardware can be provided, but
the laptop is at his place...
--
Atte
http://atte.dk http://modlys.dk
20 Sep
20 Sep
8:55 a.m.
On 09/20/2011 08:32 AM, Atte André Jensen wrote:
Whenever a window is dragged, or there's disk
activity or something like
that there's an x-run.
Hello Atte,
Maybe the USB host controller shares its interrupt with the GPU? Could
you post the output of *cat /proc/interrupts* ? And what JACK settings
are you using? And did you try different USB ports, different USB cables?
Best,
Jeremy
9:46 a.m.
On 09/20/2011 09:32 AM, Renato wrote:
On Tue, 20 Sep 2011 08:55:04 +0200
Jeremy Jongepier <autostatic(a)gmail.com> wrote:
Hello Renato,
When the cable is of a very bad quality (especially those see-through
Mac style cheap cables:
http://usa.esaleschina.com/products/l-1-5m-usb-port-to-printer-usb-square-i…)
or has a cut somewhere it could affect the streams going through.
Best,
Jeremy
different USB cables?
even cables may affect latency?
renato 
9:40 a.m.
On 09/20/2011 08:55 AM, Jeremy Jongepier wrote:
On 09/20/2011 08:32 AM, Atte André Jensen wrote:
As I'm away from the computer ATM, I cannot provide the output, but I
ran cat /proc/interrupts, and AFAIR almost nothing was shared, I
remember being both impressed and puzzled...
Whenever a window is dragged, or there's disk
activity or something like
that there's an x-run.
Hello Atte,
Maybe the USB host controller shares its interrupt with the GPU? Could
you post the output of *cat /proc/interrupts* ? And what JACK settings
are you using?
Don't remember the exact settings, but priority round 65 or 85,
frames/period 128 (even very problematic at 256), periods/buffer at 3,
samplerate 48000.
And did you try different USB ports, different USB
cables?
I thought about that but didn't get around to trying that. I think it
has 4 usb ports + it sits in a docking station with additional 2-3
ports. I *should* make no difference according to my logic, but my logic
has failed before, so...
The cable was the same I used on my box, where performance was great...
--
Atte
http://atte.dk http://modlys.dk
9:55 a.m.
On 09/20/2011 09:40 AM, Atte André Jensen wrote:
On 09/20/2011 08:55 AM, Jeremy Jongepier wrote:
Thanks, well that bars out a lot of possible causes. Did you also run
*lsusb* to check what devices were connected to the USB controllers?
Maybe the keyboard and/or touchpad are connected to a USB controller too
and get in the way of the sound device.
Jeremy
On 09/20/2011 08:32 AM, Atte André Jensen wrote:
As I'm away from the computer ATM, I cannot provide the output, but I
ran cat /proc/interrupts, and AFAIR almost nothing was shared, I
remember being both impressed and puzzled...
Whenever a window is dragged, or there's disk
activity or something like
that there's an x-run.
Hello Atte,
Maybe the USB host controller shares its interrupt with the GPU? Could
you post the output of *cat /proc/interrupts* ? And what JACK settings
are you using?
Don't remember the exact settings, but priority round 65 or 85,
frames/period 128 (even very problematic at 256), periods/buffer at 3,
samplerate 48000.
And did you try different USB ports, different
USB cables?
I thought about that but didn't get around to trying that. I think it
has 4 usb ports + it sits in a docking station with additional 2-3
ports. I *should* make no difference according to my logic, but my logic
has failed before, so...
The cable was the same I used on my box, where performance was great...
10:15 a.m.
On 09/20/2011 09:55 AM, Jeremy Jongepier wrote:
Thanks, well that bars out a lot of possible causes.
Did you also run
*lsusb* to check what devices were connected to the USB controllers?
Maybe the keyboard and/or touchpad are connected to a USB controller too
and get in the way of the sound device.
No I didn't, will try that. He *is* using a usb keyboard and usb mouse.
I'm not sure what to look for though. Could we take my box as an
example? It's also a x-series thinkpad like his, although I realize they
could still be very different:
atte@blokhus:~$ lsusb
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 005 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 006 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 001 Device 002: ID 17ef:1000 Lenovo
Bus 002 Device 002: ID 05e3:0606 Genesys Logic, Inc. USB 2.0 Hub /
D-Link DUB-H4 USB 2.0 Hub
Bus 002 Device 003: ID 046d:c041 Logitech, Inc. G5 Laser Mouse
Bus 002 Device 004: ID 0d49:7410 Maxtor Mobile Hard Disk Drive (1TB)
Bus 004 Device 002: ID 0582:0018 Roland Corp. UA-1A
This is when I plug in my (normally unused) roland ua-1a usb soundcard,
that BTW seems to run without xruns here down to buffersize 64 (lowest
possible on that particular soundcard AFAICT) ~ 4ms latency.
How should I use this information to locate problems with the soundcard
like on my friends box?
--
Atte
http://atte.dk http://modlys.dk
10:20 a.m.
On 09/20/2011 10:15 AM, Atte André Jensen wrote:
No I didn't, will try that. He *is* using a usb
keyboard and usb mouse.
How should I use this information to locate problems
with the soundcard
like on my friends box?
If the USB keyboard and/or mouse are on the same bus as the soundcard
x-runs might occur whenever you move the mouse or type something (just
guessing here though). You might want to try to put the UA-25EX on a
different bus than the USB keyboard and mouse. In your case the sound
device sits on its own on bus 004, maybe the situation is different in
your friend's case.
Jeremy
10:22 a.m.
On 09/20/2011 10:20 AM, Jeremy Jongepier wrote:
If the USB keyboard and/or mouse are on the same bus
as the soundcard
x-runs might occur whenever you move the mouse or type something (just
guessing here though). You might want to try to put the UA-25EX on a
different bus than the USB keyboard and mouse. In your case the sound
device sits on its own on bus 004, maybe the situation is different in
your friend's case.
Ah, I see, thanks. That makes sense, will try that!
--
Atte
http://atte.dk http://modlys.dk
23 Sep
23 Sep
1:42 a.m.
On 09/20/2011 10:20 AM, Jeremy Jongepier wrote:
On 09/20/2011 10:15 AM, Atte André Jensen wrote:
Ok, been to his place tonight go dig deeper. We tried everything, and
eventually ended up with 4ms latency + ardour without xruns. For some
reason (maybe 64 bit ubuntu/natty) renoise didn't behave that well,
which threw me off, since it runs perfectly here.
What also confused me was that although PAM (limits.conf) seemed to be
configured correctly, renoise with realtime alsa kept on throwing the
"not able to use realtime threads" warning on startup. This never
happened to me before, for me it's always been a sure sign that
limits.conf was not configured correctly. Strange...
Anyways, he's gonna use ardour, so although puzzled I decided to leave
the renoise strangeness alone.
--
Atte
http://atte.dk http://modlys.dk
No I didn't, will try that. He *is* using a
usb keyboard and usb mouse.
How should I use this information to locate
problems with the soundcard
like on my friends box?
If the USB keyboard and/or mouse are on the same bus as the soundcard
x-runs might occur whenever you move the mouse or type something (just
guessing here though). You might want to try to put the UA-25EX on a
different bus than the USB keyboard and mouse. In your case the sound
device sits on its own on bus 004, maybe the situation is different in
your friend's case. 
20 Sep
20 Sep
12:09 p.m.
On 09/20/2011 03:40 AM, Atte André Jensen wrote:
[re: JACK settings] ...
Don't remember the exact settings, but priority round 65 or 85,
frames/period 128 (even very problematic at 256), periods/buffer at 3,
samplerate 48000.
Have you tried using a period of 2 ? That value is not supposed to work
well with my laptop's hda-intel chipset, but I get better performance at
2 than with 3. Just a thought.
Best,
dp
12:17 p.m.
On 09/20/2011 12:09 PM, Dave Phillips wrote:
On 09/20/2011 03:40 AM, Atte André Jensen wrote:
I *think* I tried that, but will definitely make sure to try it next
time, thanks!
--
Atte
http://atte.dk http://modlys.dk
[re: JACK settings] ...
Don't remember the exact settings, but priority round 65 or 85,
frames/period 128 (even very problematic at 256), periods/buffer at 3,
samplerate 48000.
Have you tried using a period of 2 ? That value is not supposed to work
well with my laptop's hda-intel chipset, but I get better performance at
2 than with 3. Just a thought. 
27 Sep
27 Sep
10:26 p.m.
On Tue, Sep 20, 2011 at 5:09 AM, Dave Phillips <dlphillips(a)woh.rr.com> wrote:
Have you tried using a period of 2 ? That value is not
supposed to work well
with my laptop's hda-intel chipset, but I get better performance at 2 than
with 3. Just a thought.
Hi Dave (list),
This brings up a question that has been on my mind lately - what tools
or process do you use to find out what the best settings are for an
internal soundcard?
--
Josh Lawrence
903-490-3700
10:42 p.m.
On 09/27/2011 04:26 PM, Josh Lawrence wrote:
On Tue, Sep 20, 2011 at 5:09 AM, Dave
Phillips<dlphillips(a)woh.rr.com> wrote:
Hi Josh,
Here's Dave's can't-lose process:
0) Experiment to exhaustion.
1) Search Google with high hopes.
2) Send message of desperation to Linux Audio Users list.
3) Profit!
More seriously, the available tools (lspci, lsusb, etc) can tell you
what's there - possibly with default values - but they don't tell you
what's optimal. I'm not sure they could, given the influence of other
factors e.g. shared IRQs, interface bus, other hardware/mobo considerations.
A system analysis tool would be nice, especially one targeted to
configuring an audio production system for a Linux box.
Best,
dp
Have you tried using a period of 2 ? That value
is not supposed to work well
with my laptop's hda-intel chipset, but I get better performance at 2 than
with 3. Just a thought.
Hi Dave (list),
This brings up a question that has been on my mind lately - what tools
or process do you use to find out what the best settings are for an
internal soundcard?
11:17 p.m.
On Tue, Sep 27, 2011 at 04:42:35PM -0400, Dave Phillips wrote:
Here's Dave's can't-lose process:
0) Experiment to exhaustion.
1) Search Google with high hopes.
2) Send message of desperation to Linux Audio Users list.
3) Profit!
Very much to the point, in particular (0).
More seriously, the available tools (lspci, lsusb,
etc) can tell you
what's there - possibly with default values - but they don't tell you
what's optimal. I'm not sure they could, given the influence of other
factors e.g. shared IRQs, interface bus, other hardware/mobo
considerations.
Learning to use tools such as lspci, lsusb, many others and
in particular understanding what udev is doing (a lot on
modern systems) and how to control it will help.
That's IMO one advantage of a distro such as Arch: whenever
you have some problem, the solution offered on the e.g. the
Arch wiki or suggestions from other users usually gently
force you to use the available tools and understand what is
happenng rather than giving a 'shrink wrapped' solution.
And after some time (and sweat and lapses of panic and
desperation) that pays off.
Ciao,
--
FA
29 Sep
29 Sep
5:05 p.m.
Just my two cents about UA25EX:
I have a UA25EX as well and I'm quite happy with it. I use it with Musix 3.0
(Debian Squeeze) and it works out of the box. Even I use it on an UMPC (not
as powerful as a PC) for realtime FX with jack configured at 256 buffer and
no xruns.
Hope it helps.
2011/9/27 Fons Adriaensen <fons(a)linuxaudio.org>
On Tue, Sep 27, 2011 at 04:42:35PM -0400, Dave
Phillips wrote:
--
Carlos sanchiavedraz
* Musix GNU+Linux
http://www.musix.es
Here's Dave's can't-lose process:
0) Experiment to exhaustion.
1) Search Google with high hopes.
2) Send message of desperation to Linux Audio Users list.
3) Profit!
Very much to the point, in particular (0).
More seriously, the available tools (lspci,
lsusb, etc) can tell you
what's there - possibly with default values - but they don't tell you
what's optimal. I'm not sure they could, given the influence of other
factors e.g. shared IRQs, interface bus, other hardware/mobo
considerations.
Learning to use tools such as lspci, lsusb, many others and
in particular understanding what udev is doing (a lot on
modern systems) and how to control it will help.
That's IMO one advantage of a distro such as Arch: whenever
you have some problem, the solution offered on the e.g. the
Arch wiki or suggestions from other users usually gently
force you to use the available tools and understand what is
happenng rather than giving a 'shrink wrapped' solution.
And after some time (and sweat and lapses of panic and
desperation) that pays off.
Ciao,
--
FA
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user
7:05 p.m.
On Thu, Sep 29, 2011 at 10:05 AM, Carlos sanchiavedraz
<csanchezgs(a)gmail.com> wrote:
I have a UA25EX as well and I'm quite happy with
it. I use it with Musix 3.0
(Debian Squeeze) and it works out of the box. Even I use it on an UMPC (not
as powerful as a PC) for realtime FX with jack configured at 256 buffer and
no xruns.
are you running 3 periods or 2 on your UMPC?
--
Josh Lawrence
903-490-3700
30 Sep
30 Sep
10:03 a.m.
2011/9/29 Josh Lawrence <josh(a)joshlawrence.me>
On Thu, Sep 29, 2011 at 10:05 AM, Carlos
sanchiavedraz
<csanchezgs(a)gmail.com> wrote:
2 periods, IIRC. When I'm doing RT playing I use Openbox desktop with just 2
bars, this way I save resources for FX. If I'm just messing around I use
KDE4 or Icewm instead.
If you need something more please tell.
Regards
--
Carlos sanchiavedraz
* Musix GNU+Linux
http://www.musix.es
I have a UA25EX as well and I'm quite happy
with it. I use it with Musix
3.0
(Debian Squeeze) and it works out of the box.
Even I use it on an UMPC
(not
as powerful as a PC) for realtime FX with jack
configured at 256 buffer
and
no xruns.
are you running 3 periods or 2 on your UMPC?
--
Josh Lawrence
903-490-3700
20 Sep
20 Sep
12:32 p.m.
On Tue, Sep 20, 2011 at 2:32 AM, Atte André Jensen <atte(a)email.dk> wrote:
Hi
A friend of mine who would like to use his new UA-25EX with ubuntu on his (i
think) x200 (at least it's a 2.4Ghz dualcore 64 bit thinkpad with 6 or so
gigs of ram). A very nice machine, that should be great for setting up a
linux DAW. However I cannot make it run without x-runs (did the usual
limits.conf dance). Whenever a window is dragged, or there's disk activity
or something like that there's an x-run.
you will likely need to check into the video driver that is in use.
although my 6 core system isn't as reliable as i'd like it to be,
things got a LOT better on the xrun front after i switched from the
nouveau driver to the proprietary nvidia one.
i used to be able to induce xruns by just moving the mouse pointer
from one monitor to another. now it takes a lot more than that. mostly
:(
8:36 p.m.
On Tue, Sep 20, 2011 at 06:32:30AM -0400, Paul Davis wrote:
On Tue, Sep 20, 2011 at 2:32 AM, Atte André Jensen
<atte(a)email.dk> wrote:
A more general question: are there any modern video drivers which DO NOT SUCK?
i.e. are there any RT-safe video drivers anymore?
I've stuck with old hardware using Intel Xorg open-source drivers, and always had good
results. But if I ever needed new hardware I'm not sure what options would exist.
-ken
Hi
A friend of mine who would like to use his new UA-25EX with ubuntu on his (i
think) x200 (at least it's a 2.4Ghz dualcore 64 bit thinkpad with 6 or so
gigs of ram). A very nice machine, that should be great for setting up a
linux DAW. However I cannot make it run without x-runs (did the usual
limits.conf dance). Whenever a window is dragged, or there's disk activity
or something like that there's an x-run.
you will likely need to check into the video driver that is in use.
although my 6 core system isn't as reliable as i'd like it to be,
things got a LOT better on the xrun front after i switched from the
nouveau driver to the proprietary nvidia one.
i used to be able to induce xruns by just moving the mouse pointer
from one monitor to another. now it takes a lot more than that. mostly
:( 
21 Sep
21 Sep
10:51 a.m.
Excerpts from Ken Restivo's message of 2011-09-20 20:36:10 +0200:
On Tue, Sep 20, 2011 at 06:32:30AM -0400, Paul Davis
wrote:
My hardware isn't exactly new either, and low-end, but for me with very
recent linux the intel drivers work just fine. lspci says:
00:02.0 VGA compatible controller: Intel Corporation Mobile GM965/GL960
Integrated Graphics Controller (primary) (rev 03)
Newer intel chips likely use completely different drivers, but the
hardware from a couple of years ago works fine.
On Tue, Sep 20, 2011 at 2:32 AM, Atte André
Jensen <atte(a)email.dk> wrote:
A more general question: are there any modern video drivers which DO NOT SUCK?
i.e. are there any RT-safe video drivers anymore?
I've stuck with old hardware using Intel Xorg open-source drivers, and always had
good results. But if I ever needed new hardware I'm not sure what options would
exist.
-ken Hi
A friend of mine who would like to use his new UA-25EX with ubuntu on his (i
think) x200 (at least it's a 2.4Ghz dualcore 64 bit thinkpad with 6 or so
gigs of ram). A very nice machine, that should be great for setting up a
linux DAW. However I cannot make it run without x-runs (did the usual
limits.conf dance). Whenever a window is dragged, or there's disk activity
or something like that there's an x-run.
you will likely need to check into the video driver that is in use.
although my 6 core system isn't as reliable as i'd like it to be,
things got a LOT better on the xrun front after i switched from the
nouveau driver to the proprietary nvidia one.
i used to be able to induce xruns by just moving the mouse pointer
from one monitor to another. now it takes a lot more than that. mostly
:(
4804
days inactive
4814
days old
linux-audio-user@lists.linuxaudio.org
20 comments
11 participants
participants (11)
-
Atte André Jensen -
Carlos sanchiavedraz -
Dave Phillips -
Fons Adriaensen -
Jeremy Jongepier -
Jeremy Jongepier -
Josh Lawrence -
Ken Restivo -
Paul Davis -
Philipp Überbacher -
Renato