8 Nov
2002
8 Nov
'02
6:49 p.m.
Hi all,
I am currently writing an article talks (among other things) about Linux
as one of the currently viable pro-quality-audio platforms. I am trying
to get a hold of some specs in relation to JACK, so I'd appreciate it if
you can give me a quick rundown of the following information:
All of these pertain to ALSA or JACK:
1.) What is the lowest stable achievable latency with the current stable
(2.4) branch of the kernel (with ll and preempt patches) using Alsa on a
modern system (i.e. at least 1GHz machine)?
2.) What is the lowest stable achievable latency utilizing
app<--->jack<--->audio hw
kind of setup on a modern system (i.e. at least 1GHz machine) with the
current stable (2.4) branch of the kernel (with ll and preempt patches)
using Alsa?
I would appreciate it if you restrict your responses to simple numbers
(i.e.
1) 1.6ms
2) 2ms
)
since I do not have the time to have a discussion on this matter, but am
rather looking for a quick list of testimonials in regards to the two
given questions.
Thank you for your help! Sincerely,
Ivica Ico Bukvic, composer, multimedia sculptor,
programmer, webmaster & computer consultant
http://meowing.ccm.uc.edu/~ico/
============================
"To be is to do" - Socrates
"To do is to be" - Sartre
"Do be do be do" - Sinatra
"I am" - God
9 Nov
9 Nov
10:22 a.m.
1.) What is the lowest stable achievable latency with
the current stable
(2.4) branch of the kernel (with ll and preempt patches) using Alsa on a
modern system (i.e. at least 1GHz machine)?
total latency: 2.3ms on a dual PII-450MHz, 2.4.19+ll, SCSI, matrox
g400 video adapter, SCHED_FIFO+mlockall required.
2.) What is the lowest stable achievable latency
utilizing
app<--->jack<--->audio hw
kind of setup on a modern system (i.e. at least 1GHz machine) with the
current stable (2.4) branch of the kernel (with ll and preempt patches)
using Alsa?
i can't reliably do better than about 10ms right now on the same setup
as listed above. i am investigating the reasons for this.
--p
12:39 p.m.
Thanks for the info!
Just two more questions:
Was the Jack latency always been this way or was it in the past
comparable to the Alsa-only latency?
Also, would you presume that having for newer machines with faster FSB
would improve latency even further?
Once more thanks for your help!
Ivica Ico Bukvic, composer, multimedia sculptor,
programmer, webmaster & computer consultant
http://meowing.ccm.uc.edu/~ico/
============================
"To be is to do" - Socrates
"To do is to be" - Sartre
"Do be do be do" - Sinatra
"I am" - God
2:40 p.m.
Was the Jack latency always been this way or was it in
the past
comparable to the Alsa-only latency?
Also, would you presume that having for newer machines with faster FSB
would improve latency even further?
the JACK data i provided covers an out-of-process client. the all-in-process
client case has the same latency as ALSA only.
the difference between the two is that although 99.9% of the time, the
kernel operates as desired and JACK can provide ALSA-only/in-process
performance in the out-of-process case, every once in a few tens of
thousands of process() cycles, the kernel messes up our scheduling and
this causes an xrun. tracking this down is an important but very, very
difficult case. its possible that it may never be solved.
--p
10 Nov
10 Nov
1:20 a.m.
Could this be a possible solution to the in-process xrun issue?
http://www.onlamp.com/pub/a/onlamp/2002/11/07/linux_threads.html
Ivica Ico Bukvic, composer, multimedia sculptor,
programmer, webmaster & computer consultant
http://meowing.ccm.uc.edu/~ico/
============================
"To be is to do" - Socrates
"To do is to be" - Sartre
"Do be do be do" - Sinatra
"I am" - God
-----Original Message-----
From: linux-audio-dev-admin(a)music.columbia.edu
[mailto:linux-audio-dev-
admin(a)music.columbia.edu] On Behalf Of Paul Davis
Sent: Saturday, November 09, 2002 2:40 PM
To: linux-audio-dev(a)music.columbia.edu
Subject: Re: [linux-audio-dev] Looking for some PR stuff related to
Jack
Was the Jack latency always been this way or was
it in the past
comparable to the Alsa-only latency?
BTW, the best case should have been 2.6ms, not 2.3ms.
--p
9:56 a.m.
Could this be a possible solution to the in-process
xrun issue?
http://www.onlamp.com/pub/a/onlamp/2002/11/07/linux_threads.html
there is no "in-process" xrun issue. the problem is for
"out-of-process" clients. the problem has nothing whatsoever to do
with the pthreads implementation, and everything to do with the kernel
scheduler and worst-case kernel scheduling latencies.
the problem in 2.4.19 has been seen by other people in other
contexts. it may not may not be present in 2.5/2.6 - we don't know
yet. even if its gone, there might be other issues, perhaps new ones,
that surface.
the stuff mentioned in the article about the O(1) scheduler and some
other changes that ingo has made are of some relevance to the problem,
but its not clear how much since nobody has tried JACK on 2.5 AFAIK.
--p
11 Nov
11 Nov
1:23 p.m.
Could this be a
possible solution to the in-process xrun issue?
http://www.onlamp.com/pub/a/onlamp/2002/11/07/linux_threads.html
there is no "in-process" xrun issue. the problem is for
"out-of-process" clients. the problem has nothing whatsoever to do
with the pthreads implementation, and everything to do with the kernel
scheduler and worst-case kernel scheduling latencies.
5:47 p.m.
I think Linus mentioned somewhere in the changelog that the resolution
is now by default 1000. Please correct me if I am wrong...
Ico
-----Original Message-----
From: linux-audio-dev-admin(a)music.columbia.edu
[mailto:linux-audio-dev-admin@music.columbia.edu] On Behalf Of Fernando
Pablo Lopez-Lezcano
Sent: Monday, November 11, 2002 1:19 PM
To: linux-audio-dev(a)music.columbia.edu
Subject: Re: [linux-audio-dev] Looking for some PR stuff related to Jack
Would compiling the kernel with HERTZ > 100 (say, 500) help?
I imagine it would because the scheduler would run more often.
-- Fernando
Could this be a
possible solution to the in-process xrun issue?
http://www.onlamp.com/pub/a/onlamp/2002/11/07/linux_threads.html
there is no "in-process" xrun issue. the problem is for
"out-of-process" clients. the problem has nothing whatsoever to do
with the pthreads implementation, and everything to do with the kernel
scheduler and worst-case kernel scheduling latencies. 
2:26 p.m.
Ivica Bukvic wrote:
1.) What is the lowest stable achievable latency with the current stable
(2.4) branch of the kernel (with ll and preempt patches) using Alsa on a
modern system (i.e. at least 1GHz machine)?
1.5ms
2.) What is the lowest stable achievable latency utilizing
app<--->jack<--->audio hw
2.9ms
44.1kHz
ardour (~12 channels) and freqtweak at the same time
2.4.18 + low latency
1333MHz Athlon, VIA motherboard, RME HDSP soundcard
Marcus
8085
days inactive
8088
days old
linux-audio-dev@lists.linuxaudio.org
9 comments
4 participants
participants (4)
-
Fernando Pablo Lopez-Lezcano -
Ivica Bukvic -
Marcus Andersson -
Paul Davis