29 Jan
2011
29 Jan
'11
3:03 p.m.
On 01/29/2011 03:00 PM, Robin Gareus wrote:
On 01/29/2011 12:49 PM, rosea.grammostola wrote:
I was asking cause I did a short test with Jack1 svn this morning and I
didn't got xruns with yoshimi and the hammon patch. Yoshimi zombified 3
times though, probably because of asking to much from my cpu. The mp
support of Jack2 seems to be an advantage when it comes to cpu load.
\r
On 12/09/2010 09:10 PM, Robin Gareus wrote:
jackdmp 1.9.6~dfsg.1-2 from debian/testing
though I'm experimenting/testing with tschack (git) and jack1 (svn) from
time to time: it does not make a difference.
On 12/09/2010 06:39 PM, rosea.grammostola wrote:
@Robin, Which version of JACK do you use?
\r
On 12/09/2010 06:35 PM, Robin Gareus wrote:
> On 11/27/10 13:11, Arnold Krille wrote:
>
>
>
>> Set ther cpu-governor to performance (so it doesn't switch
>> frequencies which
>> produces xruns) but reduce the maximum allowed frequency.
>>
>>
>>
> what kernel are you running?
>
>
>
> 2.6.33.7-rt29
>
>
>
That one ^
You tested it with yoshimi, hammond or other heavy patches? Foo-yc20 ?
different toolset here, but I can go from naught to 16 jconvolvers,
csound, fluidsynth, ardour& jamin in a few seconds. causing the CPU
usage to jump up and change freq w/o x-runs.
..what does cause problems here is FSBus frequency scaling, I've
disabled that in the BIOS.
29 Jan
29 Jan
3:29 p.m.
New subject: Balance between performance and noise
On 01/29/2011 03:03 PM, rosea.grammostola wrote:
The mp support of Jack2 seems to be an advantage when
it comes to cpu load.
I just learned via #jack that smp support only applies when you use JACK
with parallel clients. Bit cryptic but better described here:
http://www.grame.fr/~letz/jackdmp.html
"Jackdmp current implementation allows explicit parallel clients in a
graph to be processed on several available CPU at the same time. A
typical case is:
* in == > A ==> C ==> out and in == > B ==> C ==> out
A and B depend of the same input (the "in" driver in this case) and can
be activated on 2 processors, C waits for A and B outputs. "
So afaik if you use Jack2 with a single client like Yoshimi it doesn't
take advantage of its smp support, hence it shouldn't make a difference
whether you use Jack1, Jack2 or tschack.
Hopefully I didn't get this wrong ;)
Btw, I'm working on a tschack package for Ubuntu 10.04, almost done :)
Best,
Jeremy
3:32 p.m.
New subject: Balance between performance and noise
On 01/29/2011 03:29 PM, Jeremy Jongepier wrote:
On 01/29/2011 03:03 PM, rosea.grammostola wrote:
nope, I was just going to point out the same.
what's different is the zombification behaviour. try using jack1 with "-Z"
The mp support of Jack2 seems to be an advantage
when it comes to cpu
load.
I just learned via #jack that smp support only applies when you use JACK
with parallel clients. Bit cryptic but better described here:
http://www.grame.fr/~letz/jackdmp.html
"Jackdmp current implementation allows explicit parallel clients in a
graph to be processed on several available CPU at the same time. A
typical case is:
* in == > A ==> C ==> out and in == > B ==> C ==> out
A and B depend of the same input (the "in" driver in this case) and can
be activated on 2 processors, C waits for A and B outputs. "
So afaik if you use Jack2 with a single client like Yoshimi it doesn't
take advantage of its smp support, hence it shouldn't make a difference
whether you use Jack1, Jack2 or tschack.
Hopefully I didn't get this wrong ;) Btw, I'm working on a tschack package for Ubuntu
10.04, almost done :)
Best,
Jeremy
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user
3:43 p.m.
New subject: Balance between performance and noise
On 01/29/2011 03:32 PM, Robin Gareus wrote:
On 01/29/2011 03:29 PM, Jeremy Jongepier wrote:
Hmm yeah, probably the bad thing on my side is the non systematically
approach I took... So I could be wrong...
Anyway, I also made a setup with oom2 > LS via Jack and it looked to me
that just changing jack2 to jack1 gives me a lot of extra trouble,
because of the order in which JACK ins and out gets displayed in
patchage and also the naming of the soundcard ins and outs looks
different to me. LS crashed pretty fast and that made me decide to run
back to jack2.
\r
On 01/29/2011 03:03 PM, rosea.grammostola wrote:
nope, I was just going to point out the same.
what's different is the zombification behaviour. try using jack1 with "-Z"
The mp support of Jack2 seems to be an advantage
when it comes to cpu
load.
I just learned via #jack that smp support only applies when you use JACK
with parallel clients. Bit cryptic but better described here:
http://www.grame.fr/~letz/jackdmp.html
"Jackdmp current implementation allows explicit parallel clients in a
graph to be processed on several available CPU at the same time. A
typical case is:
* in ==> A ==> C ==> out and in ==> B ==> C ==> out
A and B depend of the same input (the "in" driver in this case) and can
be activated on 2 processors, C waits for A and B outputs. "
So afaik if you use Jack2 with a single client like Yoshimi it doesn't
take advantage of its smp support, hence it shouldn't make a difference
whether you use Jack1, Jack2 or tschack.
Hopefully I didn't get this wrong ;)
5050
days inactive
5050
days old
linux-audio-user@lists.linuxaudio.org
3 comments
3 participants
participants (3)
-
Jeremy Jongepier -
Robin Gareus -
rosea.grammostola