27 Nov
2010
27 Nov
'10
11:24 a.m.
Hi,
When I set cpu scaling with sudo cpufreq-set -g performance, I have the
best proaudio performance on my thinkpad T61. But, that will also heat
things up very quickly with as consequence more sound from the fans.
The laptop is more silent when I use, sudo cpufreq-set -g ondemand, but
I get more xruns then.
Is it possible to adjust the ondemand settings to find more balance
between performance and noise? How?
Btw., I use thinkfan to control the temperature and fan speed
http://thinkfan.sourceforge.net/
My /etc/thinkfan.conf settings are:
sensor /proc/acpi/ibm/thermal (2, 7, 8, 0, 0, 0, 7, 0, 12, 5, 5)
(0, 0, 54)
(1, 50, 57)
(2, 52, 59)
(3, 54, 63)
(4, 56, 65)
(5, 59, 66)
(7, 63, 32767)
Thanks in advance,
/r
27 Nov
27 Nov
1:11 p.m.
On Saturday 27 November 2010 11:24:01 rosea.grammostola wrote:
When I set cpu scaling with sudo cpufreq-set -g
performance, I have the
best proaudio performance on my thinkpad T61. But, that will also heat
things up very quickly with as consequence more sound from the fans.
The laptop is more silent when I use, sudo cpufreq-set -g ondemand, but
I get more xruns then.
Is it possible to adjust the ondemand settings to find more balance
between performance and noise? How?
Set ther cpu-governor to performance (so it doesn't switch frequencies which
produces xruns) but reduce the maximum allowed frequency.
Have fun,
Arnold
11:11 p.m.
On 11/27/2010 01:11 PM, Arnold Krille wrote:
On Saturday 27 November 2010 11:24:01
rosea.grammostola wrote:
Thanks. Which numbers are reasonable?
It might be handy to be able to use the cpu-governor performance as it
is now, sometimes. Is it possible to make another custom governor type
which is a adjusted 'performance' setting... I read something about a
userspace governor(?)...
Regards,
/r
When I set cpu scaling with sudo cpufreq-set -g
performance, I have the
best proaudio performance on my thinkpad T61. But, that will also heat
things up very quickly with as consequence more sound from the fans.
The laptop is more silent when I use, sudo cpufreq-set -g ondemand, but
I get more xruns then.
Is it possible to adjust the ondemand settings to find more balance
between performance and noise? How?
Set ther cpu-governor to performance (so it doesn't switch frequencies
which
produces xruns) but reduce the maximum allowed frequency.
9 Dec
9 Dec
6:36 p.m.
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?
I do have realtime-kernels (2.6.31.12-rt21, 2.6.33.7-rt29) running on 4
PCs/Laptops (Intel core duo 32bit, Intel Core 2 Duo 64bit, Atom and an
Intel-i5) and on none of them frequency-scaling ever caused any x-runs.
The ondemand governor is ruling them all.
It /might/ be an issue with other CPUs or mainboards.. or non-RT
systems, or with extreme low latency. But I'm not seeing any issues with
freq scaling even at 32*2@48kHz.
ciao,
robin
6:39 p.m.
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 ?
/r
9:10 p.m.
On 12/09/2010 06:39 PM, rosea.grammostola wrote:
On 12/09/2010 06:35 PM, Robin Gareus wrote:
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.
Cheers!
robin
--
Robin Gareus
84bis Rue de Grenelle
75007 Paris, France
mobile: +33 612 738 346
mailto: robin(a)gareus.org
jabber: xmpp:rgareus@ik.nu
phone & fax: +33 95 222 567 2
Public Key at http://pgp.mit.edu/
Fingerprint : 7107 840B 4DC9 C948 076D 6359 7955 24F1 4F95 2B42
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 ? 
10 Dec
10 Dec
12:54 a.m.
On 12/09/2010 12:10 PM, Robin Gareus wrote:
On 12/09/2010 06:39 PM, rosea.grammostola wrote:
Turns out that there are two parameters in my laptop's BIOS (Lenovo
T510) that can disable both pci and pci express power management.
I'll be darned! Guess what? The xruns are gone! Like magic!! Just on a
preliminary test (I don't have all the hardware I would need here) but
running with the "Ondemand" governor.
This is a _golden_ tip. Thanks much!
-- Fernando
On 12/09/2010 06:35 PM, Robin Gareus wrote:
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. 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 ?
2:13 a.m.
On 12/09/2010 03:54 PM, Fernando Lopez-Lezcano wrote:
On 12/09/2010 12:10 PM, Robin Gareus wrote:
Ah Fernando, always jumping to conclusions before all the evidence is in :-)
A more thorough test (reproducing almost all the stuff I used last time
I performed live - the only thing missing are the usb controllers -
reveals that the problem is different (but strangely related to power
management).
Everything works fine in both cases - pci/pci express power management
On and Off, and using the "On Demand" governor - until I redo all Jack
connections using jmess. This is what triggers problems in jackd
(1.9.6). Tons of messages and xruns reported. Switching to the
"Performance" governor stops the error messages, but something is left
in a broken state inside jackd. For example, opening Firefox triggers
xruns (remember, I switched to "Performance" so now the cpu is running
at top speed). There were no xruns doing that before using jmess. Now to
the more confusing part. If I turn OFF the cpuspeed daemon
(/sbin/service cpuspeed stop) the xruns go away. What?? :-)
So now I stop everything (all jack clients, jack itself). I restart the
cpuspeed daemon. Then restart everything from scratch except for jmess
and I get no xruns.
Will keep investigating... it would seem that jmess is doing something
(or most probably not doing something) that is really upsetting jackd.
-- Fernando
On 12/09/2010 06:39 PM, rosea.grammostola wrote:
Turns out that there are two parameters in my laptop's BIOS (Lenovo
T510) that can disable both pci and pci express power management.
I'll be darned! Guess what? The xruns are gone! Like magic!! Just on a
preliminary test (I don't have all the hardware I would need here) but
running with the "Ondemand" governor.
This is a _golden_ tip. Thanks much! On 12/09/2010 06:35 PM, Robin Gareus wrote:
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. 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 ?
28 Dec
28 Dec
10:03 p.m.
On 12/09/2010 09:10 PM, Robin Gareus wrote:
On 12/09/2010 06:39 PM, rosea.grammostola wrote:
It's been a while, but I'm not sure if I have that setting on my T61
I have in my BIOS:
Intel R Speedstep technology <enabled>
Mode for AC <auto>
Battery <auto>
Adaptive Therman Management
AC <balanced>
Batt. <balanced>
CPU Power Management <auto>
PCI <auto>
PCI E <auto>
Regards,
\r
On 12/09/2010 06:35 PM, Robin Gareus wrote:
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.
Cheers!
robin On 11/27/10 13:11, Arnold Krille wrote:
That one ^
You tested it with yoshimi, hammond or other heavy patches? Foo-yc20 ? 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
7 Jan
7 Jan
2:12 p.m.
On 12/28/2010 10:03 PM, rosea.grammostola wrote:
On 12/09/2010 09:10 PM, Robin Gareus wrote:
I hazard a guess at the last two.
Here it is called EIST (Enhanced Intel SpeedStep Technology) which I
needed to disable in order to disable bus frequency scaling.
Speedstep is still enabled and CPU freq scaling does still work when I
set "C1E" and "C5/C6" (CPU Power management) states to "yes"
or "auto"
in the BIOS. (check with 'powertop').
HTH,
robin
..what does cause problems here is FSBus
frequency scaling, I've
disabled that in the BIOS.
Cheers!
robin
It's been a while, but I'm not sure if I have that setting on my T61
I have in my BIOS:
Intel R Speedstep technology <enabled>
Mode for AC <auto>
Battery <auto>
Adaptive Therman Management
AC <balanced>
Batt. <balanced>
CPU Power Management <auto>
PCI <auto>
PCI E <auto>
Regards,
\r
29 Jan
29 Jan
12:50 p.m.
On 12/09/2010 09:10 PM, Robin Gareus wrote:
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:
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.
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 ?
9 Dec
9 Dec
6:41 p.m.
I don't see any use for cpu-governor else for saving batery-power on laptops. I
don't even have it installed here
Em quinta-feira 09 dezembro 2010, às 15:35:53, Robin Gareus escreveu:
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?
I do have realtime-kernels (2.6.31.12-rt21, 2.6.33.7-rt29) running on 4
PCs/Laptops (Intel core duo 32bit, Intel Core 2 Duo 64bit, Atom and an
Intel-i5) and on none of them frequency-scaling ever caused any x-runs.
The ondemand governor is ruling them all.
It /might/ be an issue with other CPUs or mainboards.. or non-RT
systems, or with extreme low latency. But I'm not seeing any issues with
freq scaling even at 32*2@48kHz.
ciao,
robin
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user
6:43 p.m.
ohhh, exuse me...it's about noise...people shouldn't even record near the cpu
anyways. there are also passive coolers.
Em quinta-feira 09 dezembro 2010, às 15:41:03, Fabio escreveu:
I don't see any use for cpu-governor else for
saving batery-power on laptops. I don't even have it installed here
Em quinta-feira 09 dezembro 2010, às 15:35:53, Robin Gareus escreveu:
On 11/27/10 13:11, Arnold Krille wrote:
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user
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?
I do have realtime-kernels (2.6.31.12-rt21, 2.6.33.7-rt29) running on 4
PCs/Laptops (Intel core duo 32bit, Intel Core 2 Duo 64bit, Atom and an
Intel-i5) and on none of them frequency-scaling ever caused any x-runs.
The ondemand governor is ruling them all.
It /might/ be an issue with other CPUs or mainboards.. or non-RT
systems, or with extreme low latency. But I'm not seeing any issues with
freq scaling even at 32*2@48kHz.
ciao,
robin
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user
6:41 p.m.
On 12/09/2010 06:35 PM, Robin Gareus wrote:
On 11/27/10 13:11, Arnold Krille wrote:
I think there might be other variables, like ffado usage (audiofire 2),
firewire chipset (ricoh)...
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?
I do have realtime-kernels (2.6.31.12-rt21, 2.6.33.7-rt29) running on 4
PCs/Laptops (Intel core duo 32bit, Intel Core 2 Duo 64bit, Atom and an
Intel-i5) and on none of them frequency-scaling ever caused any x-runs.
The ondemand governor is ruling them all.
It /might/ be an issue with other CPUs or mainboards.. or non-RT
systems, or with extreme low latency. But I'm not seeing any issues with
freq scaling even at 32*2@48kHz.
9:24 p.m.
On 12/09/2010 09:35 AM, Robin Gareus wrote:
On 11/27/10 13:11, Arnold Krille wrote:
Lucky you! :-) I have problems here running rt29 on a Lenovo T510
laptop. Frequency scaling causes xruns. And I found (after wasting a lot
of time on other possible causes) that setting the governor to
"Performance" was actually not enough. You had to actually stop the
cpuspeed daemon and then the xruns would just go away.
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?
I do have realtime-kernels (2.6.31.12-rt21, 2.6.33.7-rt29) running on 4
PCs/Laptops (Intel core duo 32bit, Intel Core 2 Duo 64bit, Atom and an
Intel-i5) and on none of them frequency-scaling ever caused any x-runs.
The ondemand governor is ruling them all. ..what does cause problems here is FSBus frequency
scaling, I've
disabled that in the BIOS.
Good to know! I suspect my laptop will not have that degree of configurability in the BIOS
but I'll try that with out desktops.
-- Fernando
It /might/ be an issue with other CPUs or mainboards..
or non-RT
systems, or with extreme low latency. But I'm not seeing any issues with
freq scaling even at 32*2@48kHz.
ciao,
robin
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user
27 Nov
27 Nov
1:15 p.m.
On Sat, 2010-11-27 at 11:24 +0100, rosea.grammostola wrote:
Hi,
When I set cpu scaling with sudo cpufreq-set -g performance, I have the
best proaudio performance on my thinkpad T61. But, that will also heat
things up very quickly with as consequence more sound from the fans.
The laptop is more silent when I use, sudo cpufreq-set -g ondemand, but
I get more xruns then.
Is it possible to adjust the ondemand settings to find more balance
between performance and noise? How?
You'll almost certainly be getting xruns not because the CPU is running
slower sometimes in "ondemand" mode, but because the CPU is changing
between fast and slow speeds. This speed change causes the CPU to pause
momentarily, and can very easily cause xruns; I've found that this is a
really common cause of JACK xrun issues with laptops.
The good thing is that, unless you're doing something quite CPU
intensive, you might be able to run your CPU at its lowest speed -- as
long as it's not changing speeds, you still shouldn't see xruns. Setting
the CPU governor to "conservative" will lock you in at your CPU's lowest
speed.
If that's not enough, you might be able to set your CPU to a specific
speed that's still lower than its maximum -- I know my 2.4Ghz laptop has
both 800Mhz and 1.6Ghz settings, for instance.
Thanks
Leigh
Btw., I use thinkfan to control the temperature and fan speed
http://thinkfan.sourceforge.net/
My /etc/thinkfan.conf settings are:
sensor /proc/acpi/ibm/thermal (2, 7, 8, 0, 0, 0, 7, 0, 12, 5, 5)
(0, 0, 54)
(1, 50, 57)
(2, 52, 59)
(3, 54, 63)
(4, 56, 65)
(5, 59, 66)
(7, 63, 32767)
Thanks in advance,
/r
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user
11:29 p.m.
On 11/27/2010 01:15 PM, Leigh Dyer wrote:
On Sat, 2010-11-27 at 11:24 +0100, rosea.grammostola
wrote:
Hmm when I read about the conservative mode, I wouldn't expect it to be
useful for audio work (playing pianoteq in this case), but I must say,
no xruns...
And it looks like my thinkfan configuration does more harm then good atm...
I will explore a bit more.
Thanks!
\r
Hi,
When I set cpu scaling with sudo cpufreq-set -g performance, I have the
best proaudio performance on my thinkpad T61. But, that will also heat
things up very quickly with as consequence more sound from the fans.
The laptop is more silent when I use, sudo cpufreq-set -g ondemand, but
I get more xruns then.
Is it possible to adjust the ondemand settings to find more balance
between performance and noise? How?
You'll almost certainly be getting xruns not because the CPU is running
slower sometimes in "ondemand" mode, but because the CPU is changing
between fast and slow speeds. This speed change causes the CPU to pause
momentarily, and can very easily cause xruns; I've found that this is a
really common cause of JACK xrun issues with laptops.
The good thing is that, unless you're doing something quite CPU
intensive, you might be able to run your CPU at its lowest speed -- as
long as it's not changing speeds, you still shouldn't see xruns. Setting
the CPU governor to "conservative" will lock you in at your CPU's lowest
speed.
If that's not enough, you might be able to set your CPU to a specific
speed that's still lower than its maximum -- I know my 2.4Ghz laptop has
both 800Mhz and 1.6Ghz settings, for instance.
9 Dec
9 Dec
5:53 p.m.
On 11/27/2010 01:15 PM, Leigh Dyer wrote:
On Sat, 2010-11-27 at 11:24 +0100, rosea.grammostola
wrote:
Has a choice between Jack1 and Jack2 any influence on 'the balance
between performance and noise' (on a dualcore)? Does one or the other
have higher cpu load etc. ?
Regards,
\r
Hi,
When I set cpu scaling with sudo cpufreq-set -g performance, I have the
best proaudio performance on my thinkpad T61. But, that will also heat
things up very quickly with as consequence more sound from the fans.
The laptop is more silent when I use, sudo cpufreq-set -g ondemand, but
I get more xruns then.
Is it possible to adjust the ondemand settings to find more balance
between performance and noise? How?
You'll almost certainly be getting xruns not because the CPU is running
slower sometimes in "ondemand" mode, but because the CPU is changing
between fast and slow speeds. This speed change causes the CPU to pause
momentarily, and can very easily cause xruns; I've found that this is a
really common cause of JACK xrun issues with laptops.
The good thing is that, unless you're doing something quite CPU
intensive, you might be able to run your CPU at its lowest speed -- as
long as it's not changing speeds, you still shouldn't see xruns. Setting
the CPU governor to "conservative" will lock you in at your CPU's lowest
speed.
If that's not enough, you might be able to set your CPU to a specific
speed that's still lower than its maximum -- I know my 2.4Ghz laptop has
both 800Mhz and 1.6Ghz settings, for instance.
Thanks
Leigh
>
> Btw., I use thinkfan to control the temperature and fan speed
> http://thinkfan.sourceforge.net/
>
> My /etc/thinkfan.conf settings are:
>
> sensor /proc/acpi/ibm/thermal (2, 7, 8, 0, 0, 0, 7, 0, 12, 5, 5)
>
>
> (0, 0, 54)
> (1, 50, 57)
> (2, 52, 59)
> (3, 54, 63)
> (4, 56, 65)
> (5, 59, 66)
> (7, 63, 32767)
>
4830
days inactive
4893
days old
linux-audio-user@lists.linuxaudio.org
17 comments
6 participants
participants (6)
-
Arnold Krille -
Fabio -
Fernando Lopez-Lezcano -
Leigh Dyer -
Robin Gareus -
rosea.grammostola