30 Jan
2007
30 Jan
'07
1:04 p.m.
Can anyone suggest ways to compare audio/midi performance between Linux
and Windows that (1) are relevant to non-technical musicians and (2)
make Linux compare favorably?
Not things like "I just don't like Windows" or software feature
comparisons or the politics of open vs. closed source, but rather things
like responsiveness to audio interrupts, RAM footprint of the OS and ...?
I work for a company that sells a Linux based piece of hardware that
plays windows VSTs. We spend alot of time on compatibility, especially
on getting the plugins to work with Wine. I often get asked about
switching to Windows and I don't have a good answer.
My sense is that the main benefit of Linux is that audio interrupts are
serviced faster and more predictably than in Windows because of
SCHED_FIFO and Linux's low overhead. And clearly musicians could feel
that, especially at lower buffer size settings so that's the kind of
thing that could matter.
But is it _really_ true? Is there a standard way to measure it? Or
published results about it?
Are there any other things to compare? Thanks for any input. I _want_ to
believe! %) ... mo
PS: apologies for any confusion that comes from posting to both
linux-dev and linux-user. I wasn't sure which list this is more
appropriate to...?
30 Jan
30 Jan
1:15 p.m.
Le 30 janv. 07 à 18:03, Michael Ost a écrit :
Can anyone suggest ways to compare audio/midi
performance between
Linux
and Windows that (1) are relevant to non-technical musicians and (2)
make Linux compare favorably?
Not things like "I just don't like Windows" or software feature
comparisons or the politics of open vs. closed source, but rather
things
like responsiveness to audio interrupts, RAM footprint of the OS
and ...?
I work for a company that sells a Linux based piece of hardware that
plays windows VSTs. We spend alot of time on compatibility,
especially
on getting the plugins to work with Wine. I often get asked about
switching to Windows and I don't have a good answer.
My sense is that the main benefit of Linux is that audio interrupts
are
serviced faster and more predictably than in Windows because of
SCHED_FIFO and Linux's low overhead. And clearly musicians could feel
that, especially at lower buffer size settings so that's the kind of
thing that could matter.
But is it _really_ true? Is there a standard way to measure it? Or
published results about it?
Are there any other things to compare? Thanks for any input. I
_want_ to
believe! %) ... mo
PS: apologies for any confusion that comes from posting to both
linux-dev and linux-user. I wasn't sure which list this is more
appropriate to...?
You'll probably first have to decide which Windows version you're
comparing since Vista is supposed to be better than XP:
See: http://createdigitalmusic.com/2007/01/19/vista-for-music-pro-
audio-exclusive-under-the-hood-with-cakewalks-cto/
Regards
Stephane Letz
1:26 p.m.
Stéphane Letz wrote:
You'll probably first have to decide which Windows
version you're
comparing since Vista is supposed to be better than XP:
See:
http://createdigitalmusic.com/2007/01/19/vista-for-music-pro-audio-exclusiv…
Thanks for the link. There sure were alot of "ifs" and qualifications in
that article. But I could see an advantage once you can get a proper
WaveRT driver for your hardware.
- mo
1:43 p.m.
On Tue, 2007-01-30 at 09:03 -0800, Michael Ost wrote:
Can anyone suggest ways to compare audio/midi
performance between Linux
and Windows that (1) are relevant to non-technical musicians and (2)
make Linux compare favorably?
Not things like "I just don't like Windows" or software feature
comparisons or the politics of open vs. closed source, but rather things
like responsiveness to audio interrupts, RAM footprint of the OS and ...?
Any comparison would depend on what Linux you're using. Vanilla code
from kernel.org, binary packages from some distribution, RT-patch or
not? There are hundreds (thousands?) of kernel configuration parameters
that affect the speed and code size.
--ll
2:17 p.m.
On 30 Jan 2007, at 17:03, Michael Ost wrote:
Can anyone suggest ways to compare audio/midi
performance between
Linux
and Windows that (1) are relevant to non-technical musicians and (2)
make Linux compare favorably?
Not things like "I just don't like Windows" or software feature
comparisons or the politics of open vs. closed source, but rather
things
like responsiveness to audio interrupts, RAM footprint of the OS
and ...?
I work for a company that sells a Linux based piece of hardware that
plays windows VSTs. We spend alot of time on compatibility,
especially
on getting the plugins to work with Wine. I often get asked about
switching to Windows and I don't have a good answer.
My sense is that the main benefit of Linux is that audio interrupts
are
serviced faster and more predictably than in Windows because of
SCHED_FIFO and Linux's low overhead. And clearly musicians could feel
that, especially at lower buffer size settings so that's the kind of
thing that could matter.
I would have thought that the best way to measure scheduler
performance is to write a simple VST plugin that writes the precise
time at which it was called into a ram buffer, and writes the buffer
out to disk after a few tens of thousands of calls.
You can the measure the times between adjacent runs and work out if
there's any significant difference in jitter.
I would think that the RAM footprint is essentially impossible to
measure, as you can't tell what proportion of the kernel space is
buffers etc.
From a commercial point of view, your are at the very least saving
licence fees for each piece of hardware, that would surely eat into
your profit margin.
- Steve
2:35 p.m.
Hi
I can only give you a musicians 'real world' comment.
I had Sonar on my XP box and used it a lot for multitracking.
It was an a Athlon 3400 64 ( running 32 xp) 1 gig ram and an RME card for
sound.
It ran fine on the XP box but still had a few wobblies re dropout after
about 10 tracks.
Can I point out I use all instruments live and record them to disk, haven't
got Midi working yet.
Thing is as you changed the tracks redoing some live again I noticed a
slight compression on the tracks.
It seemed to build up as you used more stuff on the tracks IE: reverbs and
such.
Not long after I installed 64studio and used Ardour doing the same kind of
stuff, running 46 of course,
and I didn't get this effect even on more tracks and using effects.
To my ears the end result sounds cleaner and more like the original.
Just my 2 p's worth!
Cheers
Bob
wavesound
On 30/01/07, Steve Harris <steve(a)plugin.org.uk> wrote:
On 30 Jan 2007, at 17:03, Michael Ost wrote:
Can anyone suggest ways to compare audio/midi
performance between
Linux
and Windows that (1) are relevant to non-technical musicians and (2)
make Linux compare favorably?
Not things like "I just don't like Windows" or software feature
comparisons or the politics of open vs. closed source, but rather
things
like responsiveness to audio interrupts, RAM footprint of the OS
and ...?
I work for a company that sells a Linux based piece of hardware that
plays windows VSTs. We spend alot of time on compatibility,
especially
on getting the plugins to work with Wine. I often get asked about
switching to Windows and I don't have a good answer.
My sense is that the main benefit of Linux is that audio interrupts
are
serviced faster and more predictably than in Windows because of
SCHED_FIFO and Linux's low overhead. And clearly musicians could feel
that, especially at lower buffer size settings so that's the kind of
thing that could matter.
I would have thought that the best way to measure scheduler
performance is to write a simple VST plugin that writes the precise
time at which it was called into a ram buffer, and writes the buffer
out to disk after a few tens of thousands of calls.
You can the measure the times between adjacent runs and work out if
there's any significant difference in jitter.
I would think that the RAM footprint is essentially impossible to
measure, as you can't tell what proportion of the kernel space is
buffers etc.
From a commercial point of view, your are at the very least saving
licence fees for each piece of hardware, that would surely eat into
your profit margin.
- Steve
5:13 p.m.
On Tue, 2007-01-30 at 09:03 -0800, Michael Ost wrote:
Can anyone suggest ways to compare audio/midi
performance between Linux
and Windows that ... make Linux compare favorably?
I work for a company that sells a Linux based piece of
hardware that
plays windows VSTs.
The word "FUD" comes to mind. No idea why.
Bob
--
Bob Ham <rah(a)bash.sh>
5:24 p.m.
On Tue, 2007-01-30 at 21:05 +0000, Bob Ham wrote:
On Tue, 2007-01-30 at 09:03 -0800, Michael Ost wrote:
Further to that, something constructive: perhaps you could try telling
your customers why *you* chose linux, rather than trying to find reasons
to tell them *they* should.
Bob
--
Bob Ham <rah(a)bash.sh>
Can anyone suggest ways to compare audio/midi
performance between Linux
and Windows that ... make Linux compare favorably?
I work for a company that sells a Linux based
piece of hardware that
plays windows VSTs.
The word "FUD" comes to mind. No idea why. 
6:25 p.m.
On Tue, Jan 30, 2007 at 09:18:06PM +0000, Bob Ham wrote:
On Tue, 2007-01-30 at 21:05 +0000, Bob Ham wrote:
the customers dont notice. they still use windows or no computers at
all.
it looks rather like a question from the management.
On Tue, 2007-01-30 at 09:03 -0800, Michael Ost
wrote:
Further to that, something constructive: perhaps you could try telling
your customers why *you* chose linux, rather than trying to find reasons
to tell them *they* should. Can anyone suggest ways to compare audio/midi
performance between Linux
and Windows that ... make Linux compare favorably?
I work for a company that sells a Linux based
piece of hardware that
plays windows VSTs.
The word "FUD" comes to mind. No idea why.
Bob
--
Bob Ham <rah(a)bash.sh>
--
torben Hohn
http://galan.sourceforge.net -- The graphical Audio language
31 Jan
31 Jan
5:44 a.m.
Hi,
torbenh(a)gmx.de skrev:
On Tue, Jan 30, 2007 at 09:18:06PM +0000, Bob Ham
wrote:
Whatever the reasons, it's a valid and interesting question. In truth
Linux is often touted (not the least with respect to audio) as a better
performer than Windows.
Though I can't say that I have personally experienced this. It is hard
work getting a Linux system "tuned", I have actually never succeeded
without some drawback that have forced me back to generic configurations.
Not that I complain, my current (k)ubuntu kernel performs "good
enough"tm, but I am certain it would be no problem getting equal
performance under Windows. My choice of using Linux has more to do with
the freedom of opensourceness (it's a word!..now atleast).
Steve's idea with a vst timing plugin sounds very interesting. One using
LADSPA would be equally interesting for comparing Linux to Linux. Are
there other performance measurements that would be nice? xruns under
load I suppose.
Having a test suite for system performance would be great!
I would not rule out that Linux is found to perform worse under some
circumstances. But that is ok. Adaptability is one of the strong points
of open source, once we know the problems we can start fixing them.
Regards,
Robert
On Tue, 2007-01-30 at 21:05 +0000, Bob Ham
wrote:
the customers dont notice. they still use windows or no computers at
all.
it looks rather like a question from the management.
On Tue, 2007-01-30 at 09:03 -0800, Michael Ost
wrote:
Further to that, something constructive: perhaps you could try telling
your customers why *you* chose linux, rather than trying to find reasons
to tell them *they* should.
Can anyone suggest ways to compare audio/midi
performance between Linux
and Windows that ... make Linux compare favorably?
I work for a company that sells a Linux based piece of hardware that
plays windows VSTs.
The word "FUD" comes to mind. No idea why.
6 a.m.
I would not rule out that Linux is found to perform worse under some
circumstances. But that is ok. Adaptability is one of the strong points
of open source, once we know the problems we can start fixing them.
Nice point and this is the strength of OS. the problems are addressed far
quicker than in Prop' software.
Cheers
Bob
On 31/01/07, Robert Jonsson <rj(a)spamatica.se> wrote:
Hi,
torbenh(a)gmx.de skrev:
Whatever the reasons, it's a valid and interesting question. In truth
Linux is often touted (not the least with respect to audio) as a better
performer than Windows.
Though I can't say that I have personally experienced this. It is hard
work getting a Linux system "tuned", I have actually never succeeded
without some drawback that have forced me back to generic configurations.
Not that I complain, my current (k)ubuntu kernel performs "good
enough"tm, but I am certain it would be no problem getting equal
performance under Windows. My choice of using Linux has more to do with
the freedom of opensourceness (it's a word!..now atleast).
Steve's idea with a vst timing plugin sounds very interesting. One using
LADSPA would be equally interesting for comparing Linux to Linux. Are
there other performance measurements that would be nice? xruns under
load I suppose.
Having a test suite for system performance would be great!
I would not rule out that Linux is found to perform worse under some
circumstances. But that is ok. Adaptability is one of the strong points
of open source, once we know the problems we can start fixing them.
Regards,
Robert
On Tue, Jan 30, 2007 at 09:18:06PM +0000, Bob Ham
wrote:
> On Tue, 2007-01-30 at 21:05 +0000, Bob Ham wrote:
>
>> On Tue, 2007-01-30 at 09:03 -0800, Michael Ost wrote:
>>
>>> Can anyone suggest ways to compare audio/midi performance between
Linux
>>> and Windows that ... make Linux
compare favorably?
>>>
>>> I work for a company that sells a Linux based piece of hardware that
>>> plays windows VSTs.
>>>
>> The word "FUD" comes to mind. No idea why.
>>
> Further to that, something constructive: perhaps you could try telling
> your customers why *you* chose linux, rather than trying to find
reasons
to tell
them *they* should.
the customers dont notice. they still use windows or no computers at
all.
it looks rather like a question from the management.
7:36 a.m.
On Tue, 2007-01-30 at 23:30 +0100, torbenh(a)gmx.de wrote:
On Tue, Jan 30, 2007 at 09:18:06PM +0000, Bob Ham
wrote:
You are correct there. From the modern business perspective, though,
management are Michael's personal customers.
I think the argument still applies. If your goal is to convince someone
that your choice of linux is correct, telling them why you chose linux
in the first place seems, to me at least, to be the best method, rather
than seeking out reasons which you yourself haven't been concerned with
previously.
Bob
--
Bob Ham <rah(a)bash.sh>
On Tue, 2007-01-30 at 21:05 +0000, Bob Ham
wrote:
the customers dont notice. they still use windows or no computers at
all.
it looks rather like a question from the management. On Tue, 2007-01-30 at 09:03 -0800, Michael Ost
wrote:
Further to that, something constructive: perhaps you could try telling
your customers why *you* chose linux, rather than trying to find reasons
to tell them *they* should. Can anyone suggest ways to compare audio/midi
performance between Linux
and Windows that ... make Linux compare favorably?
I work for a company that sells a Linux based
piece of hardware that
plays windows VSTs.
The word "FUD" comes to mind. No idea why.
7:41 a.m.
On 31 Jan 2007, at 11:27, Bob Ham wrote:
On Tue, 2007-01-30 at 23:30 +0100, torbenh(a)gmx.de
wrote:
I don't think that's necessarily the case, just because Linux had
better RT performance in 2000 doesn't mean it still does today, with
Vista and general improvements.
I think it's reasonable for management to question if it's still the
best choice.
- Steve
On Tue, Jan 30, 2007 at 09:18:06PM +0000, Bob Ham
wrote:
You are correct there. From the modern business perspective, though,
management are Michael's personal customers.
I think the argument still applies. If your goal is to convince
someone
that your choice of linux is correct, telling them why you chose linux
in the first place seems, to me at least, to be the best method,
rather
than seeking out reasons which you yourself haven't been concerned
with
previously. On Tue, 2007-01-30 at 21:05 +0000, Bob Ham
wrote:
the customers dont notice. they still use windows or no computers at
all.
it looks rather like a question from the management. On Tue, 2007-01-30 at 09:03 -0800, Michael Ost
wrote:
> Can anyone suggest ways to compare audio/midi performance
> between Linux
> and Windows that ... make Linux compare favorably?
> I work for a company that sells a Linux based piece of hardware
> that
> plays windows VSTs.
The word "FUD" comes to mind. No idea why.
Further to that, something constructive: perhaps you could try
telling
your customers why *you* chose linux, rather than trying to find
reasons
to tell them *they* should. 
8:54 a.m.
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Steve Harris wrote:
On 31 Jan 2007, at 11:27, Bob Ham wrote:
pick two out of three: cheap, quick, good. - Who's in the management? :)
IMO most modern PC/PPC hardware and OS are equally good enough for
non-hi-end audio engineering. -> tweak your personal flavor!
[hardware-]end-users are concerned about low-latency, scalability and
reliability. (audio-processing-quality, easy-use, etc. are irrelevant
here)
Pro's for gnu/Linux that come to my mind:
- flexibility (all kinds of... GNU)
- reliability (you have the option to see what's the OS/HW is doing)
- existing "free" Live-cds
- active and open developer community ;-)
- ...
Cons;
If "windows wants" it can perform better than a fully fledged
rt-unix-kernel. - but that remains to be proven for Vista!
It would be nice to have a software-suite to compare
linux-audio-realtime performance .. - For linux/linux tests there seem
to be various code-snippets out there... does anyone care to share
pointers?
robin
[...]
I don't think that's necessarily the case, just because Linux had better
RT performance in 2000 doesn't mean it still does today, with Vista and
general improvements.
I think it's reasonable for management to question if it's still the
best choice.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
iD8DBQFFwJGUeVUk8U+VK0IRAlBcAJ92OtWRExkXGwzf1OFxdCTlXW2KLQCgpX+9
AbZjWHeYCQ47Zy5OnRebsnY=
=incw
-----END PGP SIGNATURE-----
10:06 a.m.
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Lars Luthman wrote:
On Wed, 2007-01-31 at 13:54 +0100, Robin Gareus
wrote:
maybe not for audio-apps.
I once read some books about "subverting the windows kernel" - one can
do marvelous stuff and get great performance out of it!! - only half
kiddingly. - luckily I have not booted into XP for years and this "Con"
might just be crap...
however some instinct tells me that a UNIX - no matter how pre-emptive
or tweaked - will always have more overhead than some DirectX-app. - but
IMO this is irrelevant for "normal" audio apps on modern machines.. -
memory management and -locking are more crucial..
robin
Cons;
If "windows wants" it can perform better than a fully fledged
rt-unix-kernel. - but that remains to be proven for Vista!
Are you saying that this is true for XP? Are there any references for
that? -----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
iD8DBQFFwKJJeVUk8U+VK0IRAguvAKC5WkK6enqRl5YXpuqoY6hpdcjiCwCeKxSG
Mqz4u/A+Eg8g/LD21Jvrlbc=
=+Sl0
-----END PGP SIGNATURE-----
12:59 p.m.
On 31 Jan 2007, at 14:06, Robin Gareus wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Lars Luthman wrote:
I don't know if it's still the case, but a few years ago the NT
kernel was very heavyweight compared to the UNIX kernel, and too a
long time to do things like context switches.
- Steve
On Wed, 2007-01-31 at 13:54 +0100, Robin Gareus
wrote:
maybe not for audio-apps.
I once read some books about "subverting the windows kernel" - one can
do marvelous stuff and get great performance out of it!! - only half
kiddingly. - luckily I have not booted into XP for years and this
"Con"
might just be crap...
however some instinct tells me that a UNIX - no matter how pre-emptive
or tweaked - will always have more overhead than some DirectX-app.
- but
IMO this is irrelevant for "normal" audio apps on modern machines.. -
memory management and -locking are more crucial.. Cons;
If "windows wants" it can perform better than a fully fledged
rt-unix-kernel. - but that remains to be proven for Vista!
Are you saying that this is true for XP? Are there any references for
that? 
1:32 p.m.
Greetings:
Michael Gogins recently tested Csound5 with Windows XP Media Center
Edition and with Ubuntu 6.10. His tests indicated that Linux was the
slightly better performer, you can check out his post and commentary on
the Csound mail list archive :
http://sourceforge.net/mailarchive/forum.php?thread_id=31517593&forum_i…
Best,
dp
6456
days inactive
6457
days old
linux-audio-dev@lists.linuxaudio.org
17 comments
10 participants
participants (10)
-
Bob Ham -
Dave Phillips -
Frank Smith -
Lars Luthman -
Michael Ost -
Robert Jonsson -
Robin Gareus -
Steve Harris -
Stéphane Letz -
torbenh＠gmx.de