16 Sep
2011
16 Sep
'11
7:33 p.m.
if you can run a USB 1 class compliant audio device on your linux
system with lower latency than the current minimum so far posted to
this thread, please post with the settings you use and the type of
device. if your best performance has already been posted, no need to
repeat it.
i'll get the ball rolling with 2048 frames. note that we are assuming
2 channels only, but duplex mode (simultaneous capture+playback).
thanks.
16 Sep
16 Sep
7:43 p.m.
On 09/16/2011 07:33 PM, Paul Davis wrote:
if you can run a USB 1 class compliant audio device
And how do I figure out if the device is usb *1* class compliant?
--
Atte
http://atte.dk http://modlys.dk
7:46 p.m.
On Fri, Sep 16, 2011 at 1:43 PM, Atte André Jensen <atte(a)email.dk> wrote:
On 09/16/2011 07:33 PM, Paul Davis wrote:
actually, it almost doesn't matter because almost all of the USB audio
devices that work on linux are USB1; even if you have a USB2 device,
i'd be interested in hearing about that too (assuming it works with
linux).
if you can run a USB 1 class compliant audio device
And how do I figure out if the device is usb *1* class compliant? 
19 Sep
19 Sep
12:02 p.m.
On 16/09/11 18:46, Paul Davis wrote:
On Fri, Sep 16, 2011 at 1:43 PM, Atte André
Jensen<atte(a)email.dk> wrote:
I have an Edirol UA-1000: it was one of the first USB2 audio devices
made, and has its own kernel driver (common with the Edirol UA-101,
which superceded it). AFAIK it's not class compliant with either USB1 or
USB2. It has 12 inputs and 10 outputs.
jackd runs very nicely with:
/usr/bin/jackd -P89 -t2000 -dalsa -r44100 -p256 -n2 -D -Chw:UA1000
-Phw:UA1000 -zs -I110 -O110
Hardware is a Dell Vostro 1520 laptop, Core2 Duo 2.4 GHz P8600.
Other software is Ubuntu 10.04, vanilla kernel from kernel.org
(currently 3.0.4), and jack1 from svn.
I get the occasional xrun if I forget to disable CPU frequency scaling
and the DSP load gets high, but otherwise it goes all day xrun free with
these settings.
Colin.
On 09/16/2011 07:33 PM, Paul Davis wrote:
actually, it almost doesn't matter because almost all of the USB audio
devices that work on linux are USB1; even if you have a USB2 device,
i'd be interested in hearing about that too (assuming it works with
linux).
if you can run a USB 1 class compliant audio device
And how do I figure out if the device is usb *1* class compliant? 
16 Sep
16 Sep
8:02 p.m.
On Fri, Sep 16, 2011 at 01:33:42PM -0400, Paul Davis wrote:
if you can run a USB 1 class compliant audio device on
your linux
system with lower latency than the current minimum so far posted to
this thread, please post with the settings you use and the type of
device. if your best performance has already been posted, no need to
repeat it.
i'll get the ball rolling with 2048 frames. note that we are assuming
2 channels only, but duplex mode (simultaneous capture+playback).
Pardon my
obtuseness, but... Well, are we talking frames per period, as
in the -p option in jack's audio driver?
Cheers,
S.M.
8:11 p.m.
On Fri, Sep 16, 2011 at 2:02 PM, S. Massy <lists(a)wolfdream.ca> wrote:
On Fri, Sep 16, 2011 at 01:33:42PM -0400, Paul Davis
wrote:
in JACK terms, the combination of:
* sample rate
* period size (frames per period)
* number of periods (periods per buffer)
if you can run a USB 1 class compliant audio
device on your linux
system with lower latency than the current minimum so far posted to
this thread, please post with the settings you use and the type of
device. if your best performance has already been posted, no need to
repeat it.
i'll get the ball rolling with 2048 frames. note that we are assuming
2 channels only, but duplex mode (simultaneous capture+playback).
Pardon my
obtuseness, but... Well, are we talking frames per period, as
in the -p option in jack's audio driver?
8:24 p.m.
On Fri, Sep 16, 2011 at 02:11:35PM -0400, Paul Davis wrote:
On Fri, Sep 16, 2011 at 2:02 PM, S. Massy
<lists(a)wolfdream.ca> wrote:
With my Roland UA25EX I can run JACK
with -r48000 -p64 -n3 without any
xruns or stability issue. Anything lower gets dicey.
Cheers,
S.M.
On Fri, Sep 16, 2011 at 01:33:42PM -0400, Paul
Davis wrote:
in JACK terms, the combination of:
* sample rate
* period size (frames per period)
* number of periods (periods per buffer) if you can run a USB 1 class compliant audio
device on your linux
system with lower latency than the current minimum so far posted to
this thread, please post with the settings you use and the type of
device. if your best performance has already been posted, no need to
repeat it.
i'll get the ball rolling with 2048 frames. note that we are assuming
2 channels only, but duplex mode (simultaneous capture+playback).
Pardon my
obtuseness, but... Well, are we talking frames per period, as
in the -p option in jack's audio driver? 
9:01 p.m.
S. Massy wrote:
On Fri, Sep 16, 2011 at 02:11:35PM -0400, Paul Davis
wrote:
64 msec latency is best I can do with my UCA202, -r48000 -p1024 -n3, but
not 2-way (that's playback only), on non-RT kernel. Recording doesn't
work through JACK at the moment, error message says something about
impossible sample size.
--
David
gnome(a)hawaii.rr.com
authenticity, honesty, community
On Fri, Sep 16, 2011 at 2:02 PM, S. Massy
<lists(a)wolfdream.ca> wrote:
With my Roland UA25EX I can run JACK
with -r48000 -p64 -n3 without any
xruns or stability issue. Anything lower gets dicey. On Fri, Sep 16, 2011 at 01:33:42PM -0400, Paul
Davis wrote:
in JACK terms, the combination of:
* sample rate
* period size (frames per period)
* number of periods (periods per buffer) if you can run a USB 1 class compliant audio
device on your linux
system with lower latency than the current minimum so far posted to
this thread, please post with the settings you use and the type of
device. if your best performance has already been posted, no need to
repeat it.
i'll get the ball rolling with 2048 frames. note that we are assuming
2 channels only, but duplex mode (simultaneous capture+playback).
Pardon my
obtuseness, but... Well, are we talking frames per period, as
in the -p option in jack's audio driver? 
8:15 p.m.
On 09/16/2011 06:33 PM, Paul Davis wrote:
if you can run a USB 1 class compliant audio device on
your linux
system with lower latency than the current minimum so far posted to
this thread, please post with the settings you use and the type of
device. if your best performance has already been posted, no need to
repeat it.
i'll get the ball rolling with 2048 frames. note that we are assuming
2 channels only, but duplex mode (simultaneous capture+playback).
with my, now old but "semper fi", tascam us-224 i've been happy with
-r48000 -p128 -n2, for ages (5y+). thanks to Karsten Wiese and me self
on this old and rusty us-x2y alsa "rawusb" jackd backend ;) alas, it is
not the world class compliant snd-usb-audio you're asking about but a
formerly derived one, snd-usb-usx2y.
but then, years have turned, and some usb 1.1 crappy devices (read very
cheap, like the ones fitting like a usb mini-dongle, with only one
line-out and other in) it's been quite usable with -r48000 -p128 -n3.
this time using that classy snd-usb-audio ;)
aha, it helps a lot having a full preempt_rt (or threadirqs enabled)
kernel and then fine tuned irq service threads priorities (think of,
yours truly rtirq:)
cheers
--
rncbc aka Rui Nuno Capela
rncbc(a)rncbc.org
8:22 p.m.
On Fri, Sep 16, 2011 at 01:33:42PM -0400, Paul Davis wrote:
if you can run a USB 1 class compliant audio device on
your linux
system with lower latency than the current minimum so far posted to
this thread, please post with the settings you use and the type of
device. if your best performance has already been posted, no need to
repeat it.
i'll get the ball rolling with 2048 frames. note that we are assuming
2 channels only, but duplex mode (simultaneous capture+playback).
Don't have it here ATM so I can't test but my old Edirol UA2 is
always used at 48 kHz, -p 256 -n 3 and I can't remember any
problems with it.
Ciao,
--
FA
9:09 p.m.
On Fri, Sep 16, 2011 at 01:33:42PM -0400, Paul Davis wrote:
if you can run a USB 1 class compliant audio device on
your linux
system with lower latency than the current minimum so far posted to
this thread, please post with the settings you use and the type of
device. if your best performance has already been posted, no need to
repeat it.
i'll get the ball rolling with 2048 frames. note that we are assuming
2 channels only, but duplex mode (simultaneous capture+playback).
M-Audio FastTrack Pro:
-R -P70 -dalsa -X seq -r44100 -p128 -n3 -Phw:1,0 -Chw:1,1
Works flawlessly.
-ken
17 Sep
17 Sep
8:14 a.m.
New subject: Re : lowest usable latency from a USB 1 device?
----- Mail original -----
De : Paul Davis <paul(a)linuxaudiosystems.com>
À : linux-audio-user <linux-audio-user(a)lists.linuxaudio.org>
Cc :
Envoyé le : Vendredi 16 Septembre 2011 19h33
Objet : [LAU] lowest usable latency from a USB 1 device?
if you can run a USB 1 class compliant audio device on your linux
system with lower latency than the current minimum so far posted to
this thread, please post with the settings you use and the type of
device. if your best performance has already been posted, no need to
repeat it.
i'll get the ball rolling with 2048 frames. note that we are assuming
2 channels only, but duplex mode (simultaneous capture+playback).
thanks.
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user
As some others here:
Edirol UA-25 with
nperiods 3, frame size 128, 2x2 (duplex) 48kHz
without any problems on a 2.6.33.6-1.rt26 kernel and some sporadic xruns on a 3.1.0-rc5
11:36 a.m.
New subject: Re : lowest usable latency from a USB 1 device?
On 09/17/2011 08:14 AM, Frank Kober wrote:
----- Mail original -----
Same device here, Edirol UA-25. Lowest possible latency I can achieve is
2ms (48 frames/period, 48 Khz sample rate and 2 periods/buffer) but that
isn't very usable. It runs in a pretty stable manner at 64/48000*3 (4ms)
but just to be sure I mostly use 128 frames/period.
Best,
Jeremy
De : Paul
Davis<paul(a)linuxaudiosystems.com>
À : linux-audio-user<linux-audio-user(a)lists.linuxaudio.org>
Cc :
Envoyé le : Vendredi 16 Septembre 2011 19h33
Objet : [LAU] lowest usable latency from a USB 1 device?
if you can run a USB 1 class compliant audio device on your linux
system with lower latency than the current minimum so far posted to
this thread, please post with the settings you use and the type of
device. if your best performance has already been posted, no need to
repeat it.
i'll get the ball rolling with 2048 frames. note that we are assuming
2 channels only, but duplex mode (simultaneous capture+playback).
thanks.
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user
As some others here:
Edirol UA-25 with
nperiods 3, frame size 128, 2x2 (duplex) 48kHz
without any problems on a 2.6.33.6-1.rt26 kernel and some sporadic xruns on a 3.1.0-rc5
5:35 p.m.
Am 16.09.2011 19:33, schrieb Paul Davis:
if you can run a USB 1 class compliant audio device on
your linux
system with lower latency than the current minimum so far posted to
this thread, please post with the settings you use and the type of
device. if your best performance has already been posted, no need to
repeat it.
i'll get the ball rolling with 2048 frames. note that we are assuming
2 channels only, but duplex mode (simultaneous capture+playback).
MAudio Mobile Pre, T60, Suse 11.3 with Standard-Kernel Desktop.
64/2 48Khz
running Guitarix works OK.
Usually I run this setup with 128/3 48Khz to avoid xruns in large
projects. And I find it hard to notice the difference, even with 256/3 I
can record guitars processed by amp simulation without getting a
"latency-assertion" for less precise recordings ;-)
thanks.
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user
19 Sep
19 Sep
11:03 p.m.
Paul Davis wrote:
if you can run a USB 1 class compliant audio device on
your linux
system with lower latency than the current minimum so far posted to
this thread, please post with the settings you use and the type of
device.
Class compliance doesn't matter, because it's not possible to have audio
transfers with a timing algorithm that isn't already covered by the USB
audio spec.
What matters are
1) the synchronization type (adaptive, asynchronous, implicit feedback);
2) how many packets per second are transferred.
1) In implicit feedback mode, the playback and capture streams always
run in perfect lockstep (i.e., when the streams are started at the
same time, period interrupts are guaranteed to happen at the same
time). In adaptive or asynchronous mode, there is no such
synchronization, so the corresponding playback and capture periods
might be completed one USB frame earlier or later.
The only devices to use implicit feedback are the Edirol UA-1000/101
and IIRC the Tascam devices handled by the snd-usx2y driver; some
others need it but have not yet proper driver support.
2) USB 1.x devices always use 1000 packets/s; USB 2.0/3.0 devices can
use up to 8000 packets/s, but may use less. The more packets are
used, the more precise the period completion interrupts are.
For devices handled by snd-usb-audio, these can be read from the file
/proc/asound/cardX/streamY, which contains entries like this:
| Interface 1
| Altset 18
| Format: S24_3LE
| Channels: 8
| Endpoint: 1 OUT (ASYNC) <-- sync. type
| Rates: 96000
| Data packet interval: 125 us <-- time per packet
The SB Audigy 2 NX is a USB 2.0 device in asynchronous mode; it uses
8K packets/s for playback and 1K packets/s for capture. Without duplex,
I can get down to 2 x 0.4 ms periods, but this is not quite stable
(jackd -d alsa -P hw:8 -r 96000 -p 37 -n 2). For any device with duplex,
I need at least a 4 ms buffer to start streaming, and that blows up when
an xrun happens.
Regards,
Clemens
20 Sep
20 Sep
5:48 p.m.
Just out of curiosity, has anybody tried running a USB audio device
over a CAT-5 extender?
http://www.blackbox.com/Store/Detail.aspx/USB-2-0-CATx-Extender-4-Port/IC25…
-- Niels
http://nielsmayer.com
4814
days inactive
4818
days old
linux-audio-user@lists.linuxaudio.org
15 comments
13 participants
participants (13)
-
Atte André Jensen -
Clemens Ladisch -
Colin Fletcher -
david -
Fons Adriaensen -
Frank Kober -
Hartmut Noack -
Jeremy Jongepier -
Ken Restivo -
Niels Mayer -
Paul Davis -
Rui Nuno Capela -
S. Massy