27 Jan
2021
27 Jan
'21
5:41 p.m.
Hi!
Yesterday, I read an article on golem.de
(https://www.golem.de/news/ryzen-mobile-5000-cezanne-ist-modular-und-sparsam…)
about the future of ryzen (mobile) processors from AMD. Inside the
article, it was told that energy saving is done by the processor itself
and that the OS will have nothing to do with it.
Using realtime apps, I wrote in the comments that I'd really prefer the
possibility for userspace interaction and criticized my Intel core i3
8100 not to allow the user to switch to firm frequencies. It was
answered that this seems to be the future and that in case of doubt I
could switch off frequency scaling in the BIOS.
That would mean a lack of comfort to me, as I can't certainly tell, if I
am using my machine for audio or another task at the moment of powering on.
As soon as frequency switching was introduced we LAU were told, not to
use it to save xruns. And as far as I can tell, the rule still valid. Is
there a chance in the future that we can stop thinking about it, because
it just won't matter? Are we forced blowing loads of energy or do we
spent too much time in sluggish UEFI menus?
I'd really like to hear your opinion about it (and possible solutions, too)!
Greets!
Mitsch
27 Jan
27 Jan
5:45 p.m.
You're just going to have to buy the same hardware that high frequency
machine stock traders are using. Their latency requirements these days are
even more demanding than audio, and they've got the money to spend.
But also, you're likely going to need to forget hardware targeting mobile
(or more generally, battery powered) use cases.
On Wed, Jan 27, 2021 at 8:42 AM Michael Jarosch <riotsound(a)riotmusic.de>
wrote:
Hi!
Yesterday, I read an article on golem.de
(
https://www.golem.de/news/ryzen-mobile-5000-cezanne-ist-modular-und-sparsam…)
about the future of ryzen (mobile) processors from AMD. Inside the
article, it was told that energy saving is done by the processor itself
and that the OS will have nothing to do with it.
Using realtime apps, I wrote in the comments that I'd really prefer the
possibility for userspace interaction and criticized my Intel core i3
8100 not to allow the user to switch to firm frequencies. It was
answered that this seems to be the future and that in case of doubt I
could switch off frequency scaling in the BIOS.
That would mean a lack of comfort to me, as I can't certainly tell, if I
am using my machine for audio or another task at the moment of powering on.
As soon as frequency switching was introduced we LAU were told, not to
use it to save xruns. And as far as I can tell, the rule still valid. Is
there a chance in the future that we can stop thinking about it, because
it just won't matter? Are we forced blowing loads of energy or do we
spent too much time in sluggish UEFI menus?
I'd really like to hear your opinion about it (and possible solutions,
too)!
Greets!
Mitsch
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
https://lists.linuxaudio.org/listinfo/linux-audio-user
8:03 p.m.
On Wed, 27 Jan 2021, Michael Jarosch wrote:
As soon as frequency switching was introduced we LAU
were told, not to
use it to save xruns. And as far as I can tell, the rule still valid. Is
there a chance in the future that we can stop thinking about it, because
it just won't matter? Are we forced blowing loads of energy or do we
spent too much time in sluggish UEFI menus?
It depends. Setting jack to frame size 64 or lower has in my experience
shown xruns with frequency switching. This includes Intel's "Boost"
setting which is not turned off by setting performance. There seems to be
no problem when the speed goes up but I often see an xrun when the speed
goes down (just at speed change).
However, Intel has been doing speed switching in the CPU for a while now
and we are still able to set a steady speed with that on the fly. So with
the AMD it may be similar. It may still be possible to set an upper and
lower speed limit. What they do not say, is that the advertized speed may
not be usable in steady state. With the Intel, the advertized speed can be
set for all cores and run that way at 100% core use and run forever
without over heating. AMD tends to advertize a cpu speed based on some
cores running slower and the cpu managing heat by slowing some cores down.
In this case one will have to experiment to find out what speed can be
safely run on all cores without over heating and use that speed for audio.
Hopefully this can be set on the fly.
Another comment of "blowing loads of energy" with performance mode. It was
actually found that the old "ondemand" governor actually used more power
than "performance" in many cases. Ondemand has to wake up every once in a
while to see what is happening, but in performance mode the core can go to
an idle state. The newer intel powersave mode does not have this problem
but AMD (although they started work on their own governor for linux) can
only use ondemand.
The easiest way to see power use is to watch core temperature... all power
used ends up as heat.
--
Len Ovens
www.ovenwerks.net
8:17 p.m.
On Wed, 27 Jan 2021 10:02:48 -0800 (PST)
Len Ovens <len(a)ovenwerks.net> wrote:
On Wed, 27 Jan 2021, Michael Jarosch wrote:
FWIW I'm running an earlier Ryzen5 here at 32 frames, mostly with Rosegarden and
Yoshimi, using quite complex patches, and averaging 8 active channels at once. I
don't see any Xruns. I do see the occasional one if I do things like load a full
patch set over a running program.
--
Will J Godfrey
http://www.musically.me.uk
http://yoshimi.github.io
Say you have a poem and I have a tune.
Exchange them and we can both have a poem, a tune, and a song.
As soon as frequency switching was introduced we
LAU were told, not to
use it to save xruns. And as far as I can tell, the rule still valid. Is
there a chance in the future that we can stop thinking about it, because
it just won't matter? Are we forced blowing loads of energy or do we
spent too much time in sluggish UEFI menus?
It depends. Setting jack to frame size 64 or lower has in my experience
shown xruns with frequency switching. This includes Intel's "Boost"
setting which is not turned off by setting performance. There seems to be
no problem when the speed goes up but I often see an xrun when the speed
goes down (just at speed change).
However, Intel has been doing speed switching in the CPU for a while now
and we are still able to set a steady speed with that on the fly. So with
the AMD it may be similar. It may still be possible to set an upper and
lower speed limit. What they do not say, is that the advertized speed may
not be usable in steady state. With the Intel, the advertized speed can be
set for all cores and run that way at 100% core use and run forever
without over heating. AMD tends to advertize a cpu speed based on some
cores running slower and the cpu managing heat by slowing some cores down.
In this case one will have to experiment to find out what speed can be
safely run on all cores without over heating and use that speed for audio.
Hopefully this can be set on the fly.
Another comment of "blowing loads of energy" with performance mode. It was
actually found that the old "ondemand" governor actually used more power
than "performance" in many cases. Ondemand has to wake up every once in a
while to see what is happening, but in performance mode the core can go to
an idle state. The newer intel powersave mode does not have this problem
but AMD (although they started work on their own governor for linux) can
only use ondemand.
The easiest way to see power use is to watch core temperature... all power
used ends up as heat.
28 Jan
28 Jan
7:07 a.m.
On 1/27/21 8:17 AM, Will Godfrey wrote:
On Wed, 27 Jan 2021 10:02:48 -0800 (PST)
Len Ovens <len(a)ovenwerks.net> wrote:
Ages ago, I forced my old i7 laptop (with USB2) to performance mode (via
command lines on bootup, IIRC) and it stayed there, running at 2.4GHz
all the time. Occasional xruns but nothing that interfered with anything.
I've not touched the performance modes on my current i9 laptop. It
ranges from sub-1GHz to 4.5GHz clock, as needed. I think I've seen
fewer xruns than I had on the i7 with an RT kernel. I don't know if
that's related to the fact that the i9 laptop uses USB-C/USB3.2 and I
have almost everything (including the Ethernet network connection) going
through an external Thunderbolt/USB-C dock. USB soundcard and USB MIDI
connection also go through the dock.
I'm not running an RT kernel on the i9. I don't know how USB3.2/USB-C
handles interrupts vs how USB2 handled them, or if that would have an
impact on things.
Audio-wise, I don't do anything anywhere near as ambitious as Will does.
--
David W. Jones
gnome(a)hawaii.rr.com
authenticity, honesty, community
http://dancingtreefrog.com
"My password is the last 8 digits of π."
On Wed, 27 Jan 2021, Michael Jarosch wrote:
FWIW I'm running an earlier Ryzen5 here at 32 frames, mostly with Rosegarden
and
Yoshimi, using quite complex patches, and averaging 8 active channels at once. I
don't see any Xruns. I do see the occasional one if I do things like load a full
patch set over a running program.
As soon as frequency switching was introduced we
LAU were told, not to
use it to save xruns. And as far as I can tell, the rule still valid. Is
there a chance in the future that we can stop thinking about it, because
it just won't matter? Are we forced blowing loads of energy or do we
spent too much time in sluggish UEFI menus?
It depends. Setting jack to frame size
64 or lower has in my experience
shown xruns with frequency switching. This includes Intel's "Boost"
setting which is not turned off by setting performance. There seems to be
no problem when the speed goes up but I often see an xrun when the speed
goes down (just at speed change).
However, Intel has been doing speed switching in the CPU for a while now
and we are still able to set a steady speed with that on the fly. So with
the AMD it may be similar. It may still be possible to set an upper and
lower speed limit. What they do not say, is that the advertized speed may
not be usable in steady state. With the Intel, the advertized speed can be
set for all cores and run that way at 100% core use and run forever
without over heating. AMD tends to advertize a cpu speed based on some
cores running slower and the cpu managing heat by slowing some cores down.
In this case one will have to experiment to find out what speed can be
safely run on all cores without over heating and use that speed for audio.
Hopefully this can be set on the fly.
Another comment of "blowing loads of energy" with performance mode. It was
actually found that the old "ondemand" governor actually used more power
than "performance" in many cases. Ondemand has to wake up every once in a
while to see what is happening, but in performance mode the core can go to
an idle state. The newer intel powersave mode does not have this problem
but AMD (although they started work on their own governor for linux) can
only use ondemand.
The easiest way to see power use is to watch core temperature... all power
used ends up as heat.
7:26 a.m.
On 1/27/21 8:02 AM, Len Ovens wrote:
On Wed, 27 Jan 2021, Michael Jarosch wrote:
Intel doesn't run at top speed "forever". Fastest clock I've got
from my
nominal 5GHz-max i9 in my Dell XPS 15 is 4.7GHz. Intel runs clock speed
until thermals say "Slow down." If you're talking about a laptop, the
specific laptop's ability to dissipate heat is the controlling factor.
I'm not sure that Intel thermal control is anywhere within reach of BIOS
and OS software.
As soon as frequency switching was introduced we
LAU were told, not
to use it to save xruns. And as far as I can tell, the rule still
valid. Is there a chance in the future that we can stop thinking
about it, because it just won't matter? Are we forced blowing loads
of energy or do we spent too much time in sluggish UEFI menus?
It depends. Setting jack to frame size 64 or lower has in my
experience shown xruns with frequency switching. This includes Intel's
"Boost" setting which is not turned off by setting performance. There
seems to be no problem when the speed goes up but I often see an xrun
when the speed goes down (just at speed change).
However, Intel has been doing speed switching in the CPU for a while
now and we are still able to set a steady speed with that on the fly.
So with the AMD it may be similar. It may still be possible to set an
upper and lower speed limit. What they do not say, is that the
advertized speed may not be usable in steady state. With the Intel,
the advertized speed can be set for all cores and run that way at 100%
core use and run forever without over heating. AMD tends to advertize
a cpu speed based on some cores running slower and the cpu managing
heat by slowing some cores down. In this case one will have to
experiment to find out what speed can be safely run on all cores
without over heating and use that speed for audio. Hopefully this can
be set on the fly.
Another comment of "blowing loads of energy"
with performance mode. It
was actually found that the old "ondemand" governor actually used more
power than "performance" in many cases. Ondemand has to wake up every
once in a while to see what is happening, but in performance mode the
core can go to an idle state. The newer intel powersave mode does not
have this problem but AMD (although they started work on their own
governor for linux) can only use ondemand.
The easiest way to see power use is to watch core temperature... all
power used ends up as heat.
It and clock are good indicators.
--
David W. Jones
gnome(a)hawaii.rr.com
authenticity, honesty, community
http://dancingtreefrog.com
"My password is the last 8 digits of π."
8:41 a.m.
On Wed, 27 Jan 2021, david wrote:
Intel doesn't run at top speed
"forever". Fastest clock I've got from my
nominal 5GHz-max i9 in my Dell XPS 15 is 4.7GHz. Intel runs clock speed
until thermals say "Slow down." If you're talking about a laptop, the
Yes heat is the limit. My (older now) i5 will run at rated speed forever
and in fact can run at .2Ghz over that without triggering heat throttling.
specific laptop's ability to dissipate heat is the
controlling factor.
I'm not sure that Intel thermal control is anywhere within reach of BIOS
and OS software.
I would tend to agree with the engineering, but it is possible to set the
top speed below rated if heat dissapation is not up to full speed. My
steady state device temperature while building Ardour lets me know when I
need to clean the dust out... after about 10min of all cores at 100% the
temperature reaches a top temperature and sits there with no speed drop.
If I have boost turned on the speed seems to sit .2 Ghz higher.
From helping people trouble shoot newer HW than mine,
it seems newer HW
(firmware?) does a better job handing out irqs with none of them
being
doubled up. However, it seems harder to prioritize a USB device above
other USB devices... maybe it matters less for some systems.
I got an older firewire device this past year (Audiofire 12) and while use
under the new ALSA fw modules has been poor (256 frame size minimum), Used
with the ffado modules... it seems to shine better than anything I have
seen... it is true that at 16/2 (I thought it needed /3 but /2 works) the
DSP is up at 25% with just jack running (with pulse bridging), there are
no xruns unless runninig for days when there may be one xrun. This
is better than my old ice1712 PCI card which can also run at 16/2 but not
without the odd xrun. USB? I don't have a "good" USB device... but I have
not heard great things about even the best of them. (my $1 USB 1.1 audio
devices will not run below 64/2 without locking jack up)
The audio community lost out with the "everything is USB" systems of
today. I can't afford to try the PCIe cards from AudioScience or the like
to compare. I hope the next gen is better.
--
Len Ovens
www.ovenwerks.net
29 Jan
29 Jan
7:41 a.m.
On 1/27/21 8:41 PM, Len Ovens wrote:
The only way my i9 will hit 5GHz, according to Intel specs, is if it's
only using 1 core.
On Wed, 27 Jan 2021, david wrote:
Heat *dissipation* is the limit. Laptops are very poor at heat
dissipation. An over-packed compact desk case can have problems. One
benefit to the old-school full height case. ;)
Intel doesn't run at top speed
"forever". Fastest clock I've got from
my nominal 5GHz-max i9 in my Dell XPS 15 is 4.7GHz. Intel runs clock
speed until thermals say "Slow down." If you're talking about a
laptop, the
Yes heat is the limit. My (older now) i5 will run at rated speed
forever and in fact can run at .2Ghz over that without triggering heat
throttling.
specific
laptop's ability to dissipate heat is the controlling
factor. I'm not sure that Intel thermal control is anywhere within
reach of BIOS and OS software.
I would tend to agree with the engineering, but it is possible to set
the top speed below rated if heat dissapation is not up to full speed.
My steady state device temperature while building Ardour lets me know
when I need to clean the dust out... after about 10min of all cores at
100% the temperature reaches a top temperature and sits there with no
speed drop. If I have boost turned on the speed seems to sit .2 Ghz
higher.
From helping people trouble shoot newer HW than mine,
it seems newer
HW (firmware?) does a better job handing out irqs with none of them
being doubled up. However, it seems harder to prioritize a USB device
above other USB devices... maybe it matters less for some systems.
Could be, I don't know. Particularly when you start mixing USB1, USB2,
USB3, USB3.2/USB-C into things.
I got an older firewire device this past year
(Audiofire 12) and while
use under the new ALSA fw modules has been poor (256 frame size
minimum), Used with the ffado modules... it seems to shine better than
anything I have seen... it is true that at 16/2 (I thought it needed
/3 but /2 works) the DSP is up at 25% with just jack running (with
pulse bridging), there are no xruns unless runninig for days when
there may be one xrun. This is better than my old ice1712 PCI card
which can also run at 16/2 but not without the odd xrun. USB? I don't
have a "good" USB device... but I have not heard great things about
even the best of them. (my $1 USB 1.1 audio devices will not run below
64/2 without locking jack up)
My USB device runs at 64/3. I didn't have any JACK lockups at 64/2, but
got lots of xruns.
The audio community lost out with the "everything
is USB" systems of
today. I can't afford to try the PCIe cards from AudioScience or the
like to compare. I hope the next gen is better.
I agree with that. I only hope the
USB-C/3.2 doesn't mean a barrage of
proprietary drivers.
--
David W. Jones
gnome(a)hawaii.rr.com
authenticity, honesty, community
http://dancingtreefrog.com
"My password is the last 8 digits of π."
1526
days inactive
1528
days old
linux-audio-user@lists.linuxaudio.org
7 comments
5 participants
participants (5)
-
david -
Len Ovens -
Michael Jarosch -
Paul Davis -
Will Godfrey