8 Feb
2013
8 Feb
'13
10:17 p.m.
I unfortunately had to download/code "new" versions of software recently due to
burglary in what used to be a studio. This is what I found:
1. I never realised just how much time I had spent writing freaking .desktop files,
designing icons, creating MIME types, associating programs, and mucking about with menus
in general. Of course, where does one send their bunch of xdg files for others to use?
It's not like packagers accept outsiders commenting on their sub-standard packaging. I
have binaries installed right into /usr/share - how did that happen? I told the packager
and was told basically where to go "NOOB". (30 years of CM production, 15 of
those years on Linux. I'm a noob, ok, sorry).
2. What happened since 2010? About half my VST and windows programs stopped working. New
wine sucks; lots of xruns, plenty of crashes, some programs won't even install or run.
New kernel blows hard, I'm getting Xruns on a machine that's over 4-times as
powerful as the last. Never, ever had that before. Why do we always have to have
"cutting edge"? Screw the 1.5 release, the last good one was the 1.1.13/28/32,
series. Use THAT tried and tested version for your install and allow upgrades later.
3. So far, NONE of my old multi-app projects have been immediately useable, due to the
above. An OS upgrade has killed some projects. It shouldn't happen.
So I went searching for Ubuntu Lucid/KXStudio, just like I ran without incident for 3
years...it ran everything I could throw at it. There seems to have been an all-out
scorched-earth policy towards Ubuntu Lucid. It no longer exists. KXStudio just wants to
"upgrade" me - the original repos have been razed as well - but I don't want
"upgrade", I want a wine that works and a kernel that can keep up on hardware
that is 4 times faster than before.
4. And now to the biggest gripe. On my laptop, I decided to create a login for my partner
so she could quickly check her mail without me having to unblock facebook etc. When I
finally got her to log in, the DEFAULT BACKGROUND WALLPAPER HAD A HENTAI CHICK IN A
SUBSERVIENCE POSE. WTF!? Partner not amused. Are all the distro people 15 year old losers?
5. Related to 4 and 1: can we just spend an hour checking the defaults on the desktop
before releasing stuff? KDE as presented by the distros is worse than XP off a Dell ghost
disk. Horrible, horrible, horrible defaults. Nepomuk/Strigi disaster turned on by default.
Why? We already knew it was a disaster. If you're gonna say "we're a music
distro", then set it up for MUSIC not for eye-candy.
So basically, there's a whole layer missing between distro and public. Q&A
- shane
9 Feb
9 Feb
5:49 a.m.
New subject: So what do you think sucks about Linux audio ?
On 02/08/2013 11:17 AM, shane richards wrote:
I unfortunately had to download/code "new"
versions of software
recently due to burglary in what used to be a studio. This is what I
found:
1. I never realised just how much time I had spent writing freaking
.desktop files, designing icons, creating MIME types, associating
programs, and mucking about with menus in general. Of course, where
does one send their bunch of xdg files for others to use? It's not
like packagers accept outsiders commenting on their sub-standard
packaging. I have binaries installed right into /usr/share - how did
that happen? I told the packager and was told basically where to go
"NOOB". (30 years of CM production, 15 of those years on Linux. I'm a
noob, ok, sorry).
Yah, I've gotten those kind of responses from developers. Not from
packagers, though.
2. What happened since 2010? About half my VST and
windows programs
stopped working. New wine sucks; lots of xruns, plenty of crashes,
some programs won't even install or run. New kernel blows hard, I'm
getting Xruns on a machine that's over 4-times as powerful as the
last. Never, ever had that before. Why do we always have to have
"cutting edge"? Screw the 1.5 release, the last good one was the
1.1.13/28/32, series. Use THAT tried and tested version for your
install and allow upgrades later.
Hardware changes seem to have big effects on xruns.
3. So far, NONE of my old multi-app projects have been
immediately
useable, due to the above. An OS upgrade has killed some projects. It
shouldn't happen. So I went searching for Ubuntu Lucid/KXStudio, just
like I ran without incident for 3 years...it ran everything I could
throw at it. There seems to have been an all-out scorched-earth
policy towards Ubuntu Lucid. It no longer exists. KXStudio just wants
to "upgrade" me - the original repos have been razed as well - but I
don't want "upgrade", I want a wine that works and a kernel that can
keep up on hardware that is 4 times faster than before.
I don't consider Ubuntu a good choice. Easy to use, but they're
definitely heading an unfriendly direction.
4. And now to the biggest gripe. On my laptop, I
decided to create a
login for my partner so she could quickly check her mail without me
having to unblock facebook etc. When I finally got her to log in, the
DEFAULT BACKGROUND WALLPAPER HAD A HENTAI CHICK IN A SUBSERVIENCE
POSE. WTF!? Partner not amused. Are all the distro people 15 year old
losers?
Yah, some seem to be. Which distro was that? Have never seen one that
had any background graphic that original.
I have a Live CD version of Aptosid. When you boot it, the first thing
you notice is that the trash can has a huge pile of stuff in it. Didn't
anyone think of simply emptying the trash before they made their live
cd?????
5. Related to 4 and 1: can we just spend an hour
checking the
defaults on the desktop before releasing stuff? KDE as presented by
the distros is worse than XP off a Dell ghost disk. Horrible,
horrible, horrible defaults. Nepomuk/Strigi disaster turned on by
default. Why? We already knew it was a disaster. If you're gonna say
"we're a music distro", then set it up for MUSIC not for eye-candy.
In my opinion, as a former longtime KDE user, KDE4 *is not usable as a
music workstation desktop*. Too many background daemons, too much
eye-candy wasting processor time and memory (note: the eye candy can all
be turned off), etc.
Start with an audio distro. Musix works as a music distro. ArtistX
worked for me, too. Aptosid works for me, cuz I can always boot from the
live CD of whatever version I had installed, and reinstall it. And
Aptosid doesn't automatically upgrade items like the kernel; you have to
do that manually.
Or start with a non-KDE/non-GNOME3 desktop like LXDE or XFCE and add the
music stuff to it. While you're doing it - document everything you did
and keep that offsite. Similar backups for /etc, all your customized
.desktop files, etc. After it's all configured and works the way you
want it to, boot from a live Linux CD and make an image of your system
partition. Back that up and save it offsite.
--
David
gnome(a)hawaii.rr.com
authenticity, honesty, community
http://clanjones.org/david/
http://dancing-treefrog.deviantart.com/
6:46 a.m.
New subject: So what do you think sucks about Linux audio ?
On Fri, February 8, 2013 1:17 pm, shane richards wrote:
I unfortunately had to download/code "new"
versions of software recently
due to burglary in what used to be a studio. This is what I found:
1. I never realised just how much time I had spent writing freaking
.desktop files, designing icons, creating MIME types, associating
programs, and mucking about with menus in general. Of course, where does
one send their bunch of xdg files for others to use? It's not like
packagers accept outsiders commenting on their sub-standard packaging. I
have binaries installed right into /usr/share - how did that happen? I
told the packager and was told basically where to go "NOOB". (30 years of
CM production, 15 of those years on Linux. I'm a noob, ok, sorry).
I have to admit I have found my share of missing... and just wrong
.desktop files. The XDG opendesktop standard is a great idea, but the
number of audio programs that just have a category of "multimedia" is very
high. On an audio machine with lots of audio apps, the multimedia sub-menu
becomes unusable pretty quick. Of coarse now with things like Unity we are
supposed to go menu-less...
2. What happened since 2010? About half my VST and
windows programs
stopped working. New wine sucks; lots of xruns, plenty of crashes, some
programs won't even install or run. New kernel blows hard, I'm getting
Xruns on a machine that's over 4-times as powerful as the last. Never,
ever had that before. Why do we always have to have "cutting edge"? Screw
the 1.5 release, the last good one was the 1.1.13/28/32, series. Use THAT
tried and tested version for your install and allow upgrades later.
As someone who has tested (sometimes new every day) ISO installs I have
seen things change from kernel to kernel. But that is not the whole thing.
There is a pile of services that run before the login screen appears and
another bunch that start after login as the DE comes up. I don't know how
many people that do kernel development care about low latency (there are
some), but when it comes to system services I think there are less. There
have been tests done on the same system with the same kernel version where
there were good results that vanished a week later when a system upgrade
changed something else in the system. This is a straight manpower issue.
It takes time and effort just to catch that something has changed. Just
testing that all the apps even start is hours of work, let alone testing a
set of workflows. A someone who helps assemble packages into a distro, I
would welcome help.
3. So far, NONE of my old multi-app projects have been
immediately
useable, due to the above. An OS upgrade has killed some projects. It
shouldn't happen.
So I went searching for Ubuntu Lucid/KXStudio, just like I ran without
incident for 3 years...it ran everything I could throw at it. There seems
to have been an all-out scorched-earth policy towards Ubuntu Lucid. It no
longer exists. KXStudio just wants to "upgrade" me - the original repos
have been razed as well - but I don't want "upgrade", I want a wine that
works and a kernel that can keep up on hardware that is 4 times faster
than before.
I see ISOs in here:
http://cdimage.ubuntu.com/ubuntustudio/releases/lucid/release/
I do not know if the repos still support it or not for adding other
software to it. The general method of upgrade I follow (and I am not
alone... probably linux is not alone in this either) is to buy a new hard
drive and keep the old one loaded till I have made sure I can work on any
projects that require that set of software. I am keeping one now just for
gcdmaster (gnome2 and gtk2 libs are no longer shipped). I understand in
your case that was not possible.
4. And now to the biggest gripe. On my laptop, I
decided to create a login
for my partner so she could quickly check her mail without me having to
unblock facebook etc. When I finally got her to log in, the DEFAULT
BACKGROUND WALLPAPER HAD A HENTAI CHICK IN A SUBSERVIENCE POSE. WTF!?
Partner not amused. Are all the distro people 15 year old losers?
Ouch! I would also find that less than amusing. I work with UbuntuStudio
and we purposely choose backgrounds that look nice but do not distract the
user from their work. We do ship plain XFCE as well and I see we will have
to make sure to at least look at what a login to an XFCE session looks
like... though so far they have been pretty plain. Thanks for the heads
up. As the main DEs are moving more towards a consumer oriented use
profile, they are getting harder to use for development or audio
production... At least I find it that way. The new DEs also seem to be
more cpu/memory intensive. (there are lots of people that like them too)
5. Related to 4 and 1: can we just spend an hour
checking the defaults on
the desktop before releasing stuff? KDE as presented by the distros is
worse than XP off a Dell ghost disk. Horrible, horrible, horrible
defaults. Nepomuk/Strigi disaster turned on by default. Why? We already
knew it was a disaster. If you're gonna say "we're a music distro",
then
set it up for MUSIC not for eye-candy.
I'll check for those. We were without KDE libs till the release we are
working on now. We added kdenlive and I notice nepomuk gets pulled in with
it. I had not noticed if it is running be default. Again, thanks for the
heads up.
So basically, there's a whole layer missing
between distro and public. Q&A
I agree. The thing that I find frustrating is that QA is one place you do
not have know coding to help with. If each person just checked out one
workflow it would help... it may not get fixed (requires more people than
are around), but it may effect which packages are included... or get to
someone who does have time too. Things move slow, I have made changes over
a month ago where one package has been rebuilt, but the second package to
activate it has not. (I have waited longer for changes in the past)
It would be really nice if more users just documented their own workflow
for new users with comments of what additions might make it better.
--
Len Ovens
www.OvenWerks.net
7:47 a.m.
New subject: So what do you think sucks about Linux audio ?
On Fri, 2013-02-08 at 16:17 -0500, shane richards wrote:
4. And now to the biggest gripe. On my laptop, I
decided to create a
login for my partner so she could quickly check her mail without me
having to unblock facebook etc. When I finally got her to log in, the
DEFAULT BACKGROUND WALLPAPER HAD A HENTAI CHICK IN A SUBSERVIENCE
POSE. WTF!? Partner not amused. Are all the distro people 15 year old
losers?
Arts and sexuality are interconnected. A lot of those young (and old)
_girls_ and boys are very gifted in drawing and they are also far away
of being social losers. Such a default wallpaper IMO is more pleasant
than Michelangelo Buonarroti's "divine" nudity. Manga and porn are
trendy. Trends tend to become a pain in the neck after a while, but we
easily can change a wallpaper. That most popular drawn arts tend to
reduce human live to 24/7 SM indeed is weak, but OTOH "a few years ago"
they painted transfigured nudity for ridiculous biblical motives, that
was much more worse than a "chick in a subservience pose".
IMO a wallpaper doesn't have impact to Linux audio, but the trend
regarding to flashy DEs has got a bad influence. Assumed you kept your
hardware, newer kernels etc. are ok on your machine, the changes for the
DEs anyway will break the workflow. I suspect that most Linux audio
users today are using Xfce, Lxde, Fluxbox etc., but they are missing
some comfort provided by KDE and GNOME.
Regards,
Ralf
10 Feb
10 Feb
3:36 a.m.
New subject: So what do you think sucks about Linux audio ?
On Fri, Feb 08, 2013 at 04:17:38PM -0500, shane richards wrote:
4. And now to the biggest gripe. On my laptop, I
decided to create a login for my partner so she could quickly check her mail without me
having to unblock facebook etc. When I finally got her to log in, the DEFAULT BACKGROUND
WALLPAPER HAD A HENTAI CHICK IN A SUBSERVIENCE POSE. WTF!? Partner not amused. Are all the
distro people 15 year old losers?
Is this after an upgrade from Ubuntu Lucid? So what is the name of
Ubuntu that is installed on your machine now?
P.S. The person who did/condoned that should be relegated to the back
ranks and not have anything to do with public exposure/HR/PR etc, etc.
P.P.S Always keep backups of your configuration files! :)
--
"If you're not careful, the newspapers will have you hating the people
who are being oppressed, and loving the people who are doing the
oppressing." --- Malcolm X
3:58 a.m.
New subject: So what do you think sucks about Linux audio ?
On 02/09/2013 04:18 PM, Chris Bannister wrote:
On Fri, Feb 08, 2013 at 04:17:38PM -0500, shane
richards wrote:
P.P.P.S Ubuntu should be informed, this doesn't sound like something the
Ubuntu business would like.
--
David
gnome(a)hawaii.rr.com
authenticity, honesty, community
http://clanjones.org/david/
http://dancing-treefrog.deviantart.com/
4. And now to the biggest gripe. On my laptop, I
decided to create
a login for my partner so she could quickly check her mail without
me having to unblock facebook etc. When I finally got her to log
in, the DEFAULT BACKGROUND WALLPAPER HAD A HENTAI CHICK IN A
SUBSERVIENCE POSE. WTF!? Partner not amused. Are all the distro
people 15 year old losers?
Is this after an upgrade from Ubuntu Lucid? So what is the name of
Ubuntu that is installed on your machine now?
P.S. The person who did/condoned that should be relegated to the
back ranks and not have anything to do with public exposure/HR/PR
etc, etc.
P.P.S Always keep backups of your configuration files! :) 
6:37 a.m.
New subject: So what do you think sucks about Linux audio ?
Am 10.02.2013 03:58, schrieb david:
On 02/09/2013 04:18 PM, Chris Bannister wrote:
Before you go on with dissing and bashing, you may wane have a look at
the picture in question:
http://kxstudio.sourceforge.net/tmp/scr124thumb.png
Besides from that I didn't like it as well, it is clearly just a "ART
Picture".
On Fri, Feb 08, 2013 at 04:17:38PM -0500, shane
richards wrote:
P.P.P.S Ubuntu should be informed, this doesn't sound like something the
Ubuntu business would like.
4. And now to the biggest gripe. On my laptop, I
decided to create
a login for my partner so she could quickly check her mail without
me having to unblock facebook etc. When I finally got her to log
in, the DEFAULT BACKGROUND WALLPAPER HAD A HENTAI CHICK IN A
SUBSERVIENCE POSE. WTF!? Partner not amused. Are all the distro
people 15 year old losers?
Is this after an upgrade from Ubuntu Lucid? So what is the name of
Ubuntu that is installed on your machine now?
P.S. The person who did/condoned that should be relegated to the
back ranks and not have anything to do with public exposure/HR/PR
etc, etc.
P.P.S Always keep backups of your configuration files! :) 
7:05 a.m.
New subject: So what do you think sucks about Linux audio ?
Am 10.02.2013 06:37, schrieb hermann meyer:
Am 10.02.2013 03:58, schrieb david:
and here is a entry in Wikipedia which explain what the meaning of this
picture is:
http://en.wikipedia.org/wiki/Hatsune_Miku
On 02/09/2013 04:18 PM, Chris Bannister wrote:
Before you go on with dissing and bashing, you may wane have a look at
the picture in question:
http://kxstudio.sourceforge.net/tmp/scr124thumb.png
Besides from that I didn't like it as well, it is clearly just a "ART
Picture".
On Fri, Feb 08, 2013 at 04:17:38PM -0500, shane
richards wrote:
P.P.P.S Ubuntu should be informed, this doesn't sound like something the
Ubuntu business would like.
4. And now to the biggest gripe. On my laptop, I
decided to create
a login for my partner so she could quickly check her mail without
me having to unblock facebook etc. When I finally got her to log
in, the DEFAULT BACKGROUND WALLPAPER HAD A HENTAI CHICK IN A
SUBSERVIENCE POSE. WTF!? Partner not amused. Are all the distro
people 15 year old losers?
Is this after an upgrade from Ubuntu Lucid? So what is the name of
Ubuntu that is installed on your machine now?
P.S. The person who did/condoned that should be relegated to the
back ranks and not have anything to do with public exposure/HR/PR
etc, etc.
P.P.S Always keep backups of your configuration files! :) 
10:32 a.m.
New subject: So what do you think sucks about Linux audio ?
On 02/10/2013 07:05 AM, hermann meyer wrote:
This is not art. This is soft male heterosexual pedo-pornography.
Before you go
on with dissing and bashing, you may wane have a look at
the picture in question:
http://kxstudio.sourceforge.net/tmp/scr124thumb.png
Besides from that I didn't like it as well, it is clearly just a "ART
Picture".
and here is a entry in Wikipedia which explain what the meaning of this
picture is:
http://en.wikipedia.org/wiki/Hatsune_Miku 
11:32 a.m.
New subject: So what do you think sucks about Linux audio ?
On Sun, Feb 10, 2013 at 1:34 PM, SxDx wrote:
On 02/10/2013 07:05 AM, hermann meyer wrote:
This is not art. This is soft male heterosexual pedo-pornography.
Oh hey, it looks like we got ourselves a Japanese culture professor here :)
Alexandre Prokoudine
http://libregraphicsworld.org
Before you go on with dissing and bashing, you may wane have a look at
the picture in question:
http://kxstudio.sourceforge.net/tmp/scr124thumb.png
Besides from that I didn't like it as well, it is clearly just a "ART
Picture".
and here is a entry in Wikipedia which explain what the meaning of this
picture is:
http://en.wikipedia.org/wiki/Hatsune_Miku 
11:57 a.m.
New subject: So what do you think sucks about Linux audio ?
On Sun, 10 Feb 2013 14:32:56 +0400
Alexandre Prokoudine <alexandre.prokoudine(a)gmail.com> wrote:
On Sun, Feb 10, 2013 at 1:34 PM, SxDx wrote:
Either that or a foot fetishist :)
Meanwhile, back on topic.
One thing I find confusing is the way the kernel performance changes so
dramatically from one version to the next.
In debian testing, 64bit V3.1 vanilla works beautifully on my current
machine, but V3.2 requires jack to be set to 4x the latency! That's for both
vanilla & RT.
Unfortunately, making a new install only offers V3.2 now, so I hope I never
have a major problem requiring a full re-install.
--
Will J Godfrey
http://www.musically.me.uk
Say you have a poem and I have a tune.
Exchange them and we can both have a poem, a tune, and a song.
On 02/10/2013 07:05 AM, hermann meyer wrote:
This is not art. This is soft male heterosexual pedo-pornography.
Oh hey, it looks like we got ourselves a Japanese culture professor here :)
Alexandre Prokoudine
http://libregraphicsworld.org
Before you go on with dissing and bashing, you may wane have a look at
the picture in question:
http://kxstudio.sourceforge.net/tmp/scr124thumb.png
Besides from that I didn't like it as well, it is clearly just a "ART
Picture".
and here is a entry in Wikipedia which explain what the meaning of this
picture is:
http://en.wikipedia.org/wiki/Hatsune_Miku 
4:48 p.m.
New subject: So what do you think sucks about Linux audio ?
2013/2/10 Folderol <folderol(a)ukfsn.org>
One thing I find confusing is the way the kernel performance changes so
dramatically from one version to the next.
In debian testing, 64bit V3.1 vanilla works beautifully on my current
machine, but V3.2 requires jack to be set to 4x the latency! That's for
both
vanilla & RT.
Unfortunately, making a new install only offers V3.2 now, so I hope I never
have a major problem requiring a full re-install.
strange indeed...
debian testing, 64bit here, running flawlessly debian "stock"
3.2.0-4-rt-amd64 #1 SMP PREEMPT RT on Athlon X2
64 frames/period
48Khz
latency 2.67msec
--
*L'unica speranza di catarsi, ammesso che ne esista una, resta affidata
all'istinto di ribellione, alla rivolta non isterilita in progetti, alla
protesta violenta e viscerale. (V. Evangelisti)
*
5:14 p.m.
New subject: So what do you think sucks about Linux audio ?
On Sun, February 10, 2013 2:57 am, Folderol wrote:
One thing I find confusing is the way the kernel
performance changes so
dramatically from one version to the next.
In debian testing, 64bit V3.1 vanilla works beautifully on my current
machine, but V3.2 requires jack to be set to 4x the latency! That's for
both
vanilla & RT.
This may or not be it, but try looking at the irq assignment from one to
the next. The kernel sets IRQs for at least some of the devices.
--
Len Ovens
www.OvenWerks.net
5:59 p.m.
New subject: So what do you think sucks about Linux audio ?
On Sun, 2013-02-10 at 08:14 -0800, Len Ovens wrote:
On Sun, February 10, 2013 2:57 am, Folderol wrote:
The only difference for the IRQ I noticed between older and newer kernel
versions are different IRQs for the graphics. The graphics on my machine
doesn't share it's IRQ with another device any more.
Regards,
Ralf
One thing I find confusing is the way the kernel
performance changes so
dramatically from one version to the next.
In debian testing, 64bit V3.1 vanilla works beautifully on my current
machine, but V3.2 requires jack to be set to 4x the latency! That's for
both
vanilla & RT.
This may or not be it, but try looking at the irq assignment from one to
the next. The kernel sets IRQs for at least some of the devices.
6:25 p.m.
New subject: kernel changes -was - So what do you think sucks about Linux audio ?
On Sun, February 10, 2013 8:59 am, Ralf Mardorf wrote:
On Sun, 2013-02-10 at 08:14 -0800, Len Ovens wrote:
Ok, your machine, but the reply was for someone else. No matter you do say
irqs did change which is what I was saying.
What I am says though is that I'm not sure these problems are kernel
problems. They may be kernel module problems or something else. I have
used all of these kernel versions and been able to run low latency on all
of them no problems (ICE1712 based card). Other people on this list have
had no problems too. Yet there is not just one person with this same
problem. We (on this list) might do well to explore the differences
between our hardware/software to see if we can find the problem.
--
Len Ovens
www.OvenWerks.net
On Sun, February 10, 2013 2:57 am, Folderol
wrote:
The only difference for the IRQ I noticed between older and newer kernel
versions are different IRQs for the graphics. The graphics on my machine
doesn't share it's IRQ with another device any more. One thing I find confusing is the way the kernel
performance changes
so
dramatically from one version to the next.
In debian testing, 64bit V3.1 vanilla works beautifully on my current
machine, but V3.2 requires jack to be set to 4x the latency! That's
for
both
vanilla & RT.
This may or not be it, but try looking at the irq assignment from one to
the next. The kernel sets IRQs for at least some of the devices. 
6:34 p.m.
New subject: kernel changes -was - So what do you think sucks about Linux audio ?
I've actually seen some dramatic changes myself. I had great luck with
an ICE1724 based card, but I've noticed that multiple USB audio
devices don't perform as well on the 3.x series kernels compared to my
old trusty 2.6.33, which I realize is ancient now :) I'm wondering if
anyone else has had the same experience. It's only really noticeable
if you're pushing the CPU hard-- it's not a "set to 4x latency" kind
of thing.
But I'd say make super sure you're running at realtime priority. I
can't emphasize how much of a difference that makes. If you're unsure,
do this at a terminal:
ps -eLo cmd,rtprio | grep APPLICATION_NAME_HERE.
Check for your app and for JACK's prio.
Also, grab rtirq and run that if you haven't. I'd assume if you're
using a music-oriented distro that this is all already done.
-Louis
On Sun, Feb 10, 2013 at 9:25 AM, Len Ovens <len(a)ovenwerks.net> wrote:
On Sun, February 10, 2013 8:59 am, Ralf Mardorf wrote:
On Sun, 2013-02-10 at 08:14 -0800, Len Ovens
wrote:
Ok, your machine, but the reply was for someone else. No matter you do say
irqs did change which is what I was saying.
What I am says though is that I'm not sure these problems are kernel
problems. They may be kernel module problems or something else. I have
used all of these kernel versions and been able to run low latency on all
of them no problems (ICE1712 based card). Other people on this list have
had no problems too. Yet there is not just one person with this same
problem. We (on this list) might do well to explore the differences
between our hardware/software to see if we can find the problem.
--
Len Ovens
www.OvenWerks.net
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user On Sun, February 10, 2013 2:57 am, Folderol
wrote:
The only difference for the IRQ I noticed between older and newer kernel
versions are different IRQs for the graphics. The graphics on my machine
doesn't share it's IRQ with another device any more. One thing I find confusing is the way the kernel
performance changes
so
dramatically from one version to the next.
In debian testing, 64bit V3.1 vanilla works beautifully on my current
machine, but V3.2 requires jack to be set to 4x the latency! That's
for
both
vanilla & RT.
This may or not be it, but try looking at the irq assignment from one to
the next. The kernel sets IRQs for at least some of the devices.
6:59 p.m.
New subject: kernel changes
On Sun, February 10, 2013 9:34 am, Louis Gorenfeld wrote:
Also, grab rtirq and run that if you haven't.
I'd assume if you're
using a music-oriented distro that this is all already done.
Change the priority in the rtirq conig file before using it. It comes set
up to prioritize internal/pci sound devices over either USB or FW. Choose
a USB port and always use the same physical port (assuming you have picked
one on it's own irq) and put that USB port first... by going USB3
something something USB for example. This puts USB3 first and all the
other USB ports at the end. (USB3 happens to be the good one on my
machine, yours may be different)
This would be a great place for some automated setup BTW. A dialog that
pops up if you plug a USB sound card in to a port that uses a shared IRQ
suggesting to try another port even.
--
Len Ovens
www.OvenWerks.net
7:30 p.m.
New subject: kernel changes
On Sun, 2013-02-10 at 09:59 -0800, Len Ovens wrote:
On Sun, February 10, 2013 9:34 am, Louis Gorenfeld
wrote:
Sometimes it's possible to unbind devices, but I never experienced a
serious difference between a simple set up machine and a special tuned
machine:
$ cat tuning
#!/bin/bash
# sudo bash tuning - Ubuntu Studio Quantal
# 2012/Nov/04
### http://www.mythtv.org/wiki/PCI_Latency
###
http://wiki.linuxmusicians.com/doku.php?id=system_configuration#pci_bus_lat…
### Bluetooth
service bluetooth stop
### Network
#service network-manager stop
#service networking stop # does cause serious issues
#modprobe -r r8169 # Ethernet NIC driver
### TerraTec EWX 24/96
modprobe -r snd_ice1712
### Others
modprobe -r firewire-ohci
modprobe -r firewire_core
service cups stop
modprobe -r ppdev # parallel port
modprobe -r lp # printer
### Unbinding devices
echo -n "0000:00:13.2" > /sys/bus/pci/drivers/ohci_hcd/unbind
echo -n "0000:00:13.4" > /sys/bus/pci/drivers/ohci_hcd/unbind
### Log file
l="log/tuning.log"
#echo "$ lspci -v " > $l
#lspci -v >> $l
#echo >> $l
echo "$ /etc/init.d/rtirq status " > $l
/etc/init.d/rtirq status >> $l
echo "$ grep 18: /proc/interrupts" >> $l
grep 18: /proc/interrupts >> $l
echo >> $l
echo -n $(date)" - "$(uname -r)" - " >> $l
cat /etc/issue >> $l
echo "##############################" ; cat $l
exit 0
Also, grab rtirq and run that if you haven't.
I'd assume if you're
using a music-oriented distro that this is all already done.
Change the priority in the rtirq conig file before using it. It comes set
up to prioritize internal/pci sound devices over either USB or FW. Choose
a USB port and always use the same physical port (assuming you have picked
one on it's own irq) and put that USB port first... by going USB3
something something USB for example. This puts USB3 first and all the
other USB ports at the end. (USB3 happens to be the good one on my
machine, yours may be different)
This would be a great place for some automated setup BTW. A dialog that
pops up if you plug a USB sound card in to a port that uses a shared IRQ
suggesting to try another port even.
10:09 p.m.
New subject: kernel changes
On Sun, February 10, 2013 10:30 am, Ralf Mardorf wrote:
Sometimes it's possible to unbind devices, but I
never experienced a
serious difference between a simple set up machine and a special tuned
machine:
There are some things that used to work that do not any more either
because the kernel has changed or the HW.
### http://www.mythtv.org/wiki/PCI_Latency
###
http://wiki.linuxmusicians.com/doku.php?id=system_configuration#pci_bus_lat…
Even though I have an older PC that only has PCI slots, I have found that
changing pci-bus latency has not helped. In fact it made my system a lot
less stable. I did not take the time to try smaller increments of this
setting as my machine has good performance as it is. Changing PCI_bus
latency has no effect on PCIe IFs at all as they are forced to zero in
hardware.
--
Len Ovens
www.OvenWerks.net
7:25 p.m.
New subject: kernel changes -was - So what do you think sucks about Linux audio ?
On Sun, Feb 10, 2013 at 5:34 PM, Louis Gorenfeld
<louis.gorenfeld(a)gmail.com>wrote;wrote:
devices don't perform as well on the 3.x series
kernels compared to my old
trusty 2.6.33,
I used to be of the same opinion. Currently running a 3.6.11-rt kernel, and
getting very low latencies from it, on a modest 5 year old laptop.
We're talking 0.666 ms, while compiling a kernel, 0 xruns.
Notes:
-logged in as root (yes yes, i know)
-rtirq running on startup to fix interrupts
-wifi is a killer, turn it off (Thanks to Jeremy's LAC workshop on RT-ing a
system, I didn't know about it till then)
-cpu-scaling can cause xruns, cpupower frequency-set -g performance to run
all cores at 100% all the time.
-i kicked out bluetooth to be sure
-as few daemons as possible
-there is no proprietary driver for the Ati HD2400XT integrated for the 3.x
series kernel. Hence "normal" graphics, no 3d acceleration
Details:
Acer travelmate laptop: Intel Core2Duo 2.0 Ghz
Arch linux based system: Linux audio 3.6.11-rt25-1-rt #1 SMP PREEMPT RT Tue
Dec 25 00:18:17 CET 2012 x86_64 GNU/Linux
Soundcards: RME Multiface II, running trough PCMCIA adapter. Also have an
Echo Indigo DJ card, PCMCIA too.
9:59 p.m.
New subject: kernel changes -was - So what do you think sucks about Linux audio ?
On Sun, February 10, 2013 10:25 am, Harry van Haaren wrote:
On Sun, Feb 10, 2013 at 5:34 PM, Louis Gorenfeld
<louis.gorenfeld(a)gmail.com>wrote;wrote:
Same thing on my acer wireless net IF. In my case turning it off is not
enough, I have to unload the kernel module for it.
devices don't perform as well on the 3.x
series kernels compared to my
old
trusty 2.6.33,
I used to be of the same opinion. Currently running a 3.6.11-rt kernel,
and
getting very low latencies from it, on a modest 5 year old laptop.
We're talking 0.666 ms, while compiling a kernel, 0 xruns.
Notes:
-logged in as root (yes yes, i know)
-rtirq running on startup to fix interrupts
-wifi is a killer, turn it off (Thanks to Jeremy's LAC workshop on RT-ing
a
system, I didn't know about it till then) -as few daemons as possible
When I suggest this, I get "this is 2013 now, any recent HW can handle
whatever is in there". :P
Start with cron and buddies that may want check if there is new software
available or scan the whole disc for new files to make search faster.
any of the sql services unless your software has to have them.
One of my personal goals is to find out what all the stuff running on my
system is there for. Not just the system stuff at start up but also the
session stuff that starts when the session does. For example, there is a
copy of thunar running that allows files in my Desktop directory to show
on my desktop. I don't use this feature... I don't need it. (I have all
icons turned off on my desktop) Anything that is on my desktop is normally
covered by apps anyway...
-there is no proprietary driver for the Ati HD2400XT
integrated for the
3.x
series kernel. Hence "normal" graphics, no 3d acceleration
Graphic sparkle is another thing I could use less of. There are some audio
apps that use libs that use graphic acceleration. Having drivers that
support that means the GPU is doing things your CPU doesn't have to. But
the open drivers seem to do that pretty good these days.
Details:
Acer travelmate laptop: Intel Core2Duo 2.0 Ghz
Arch linux based system: Linux audio 3.6.11-rt25-1-rt #1 SMP PREEMPT RT
Tue
Dec 25 00:18:17 CET 2012 x86_64 GNU/Linux
Soundcards: RME Multiface II, running trough PCMCIA adapter. Also have an
Echo Indigo DJ card, PCMCIA too.
PCMCIA is, as far as I know, connected as if it was PCI.
Yes I am looking for a pattern. So far I am seeing PCI cards work well.
--
Len Ovens
www.OvenWerks.net
11 Feb
11 Feb
12:39 a.m.
New subject: kernel changes -was - So what do you think sucks about Linux audio ?
On Sun, Feb 10, 2013 at 12:25 PM, Harry van Haaren <harryhaaren(a)gmail.com>wrote;wrote:
On Sun, Feb 10, 2013 at 5:34 PM, Louis Gorenfeld <
louis.gorenfeld(a)gmail.com> wrote:
Yeah, but you're a ninja :)
I've been wanting to upgrade my kernel, but I'm a bit skeered of doing so
right now. I started working on my current project (PdCUDA) a few years
ago and am a few driver and kernel versions behind. I couldn't get rt
kernels to work with the nvidia proprietary driver (which is necessary for
me). Now, I understand that's been resolved in recent versions, but I'm
leery of making any big jumps in the kernel tree. Sounds dumb... but I'll
just wait for debian wheezy, which I presume will be a stable platform for
the next 2 years and update my system and try to tune for rt then.
devices don't perform as well on the 3.x
series kernels compared to my
old trusty 2.6.33,
I used to be of the same opinion. Currently running a 3.6.11-rt kernel,
and getting very low latencies from it, on a modest 5 year old laptop.
We're talking 0.666 ms, while compiling a kernel, 0 xruns.
Notes:
-logged in as root (yes yes, i know)
-rtirq running on startup to fix interrupts
-wifi is a killer, turn it off (Thanks to Jeremy's LAC workshop on RT-ing
a system, I didn't know about it till then)
-cpu-scaling can cause xruns, cpupower frequency-set -g performance to run
all cores at 100% all the time.
-i kicked out bluetooth to be sure
-as few daemons as possible
-there is no proprietary driver for the Ati HD2400XT integrated for the
3.x series kernel. Hence "normal" graphics, no 3d acceleration
Details:
Acer travelmate laptop: Intel Core2Duo 2.0 Ghz
Arch linux based system: Linux audio 3.6.11-rt25-1-rt #1 SMP PREEMPT RT
Tue Dec 25 00:18:17 CET 2012 x86_64 GNU/Linux
Soundcards: RME Multiface II, running trough PCMCIA adapter. Also have an
Echo Indigo DJ card, PCMCIA too.
1:12 a.m.
New subject: kernel changes -was - So what do you think sucks about Linux audio ?
On Sun, Feb 10, 2013 at 11:39 PM, Charles Z Henry <czhenry(a)gmail.com> wrote:
Cool! I've rendered some stuff using Blender
& its CUDA... awesome. Really
impressive
stuff there, hope it makes pdCuda equally awesome!
Cheers, -Harry
Yeah, but you're a ninja :)
Still in training though... ;)
I've just set up a repo to contain my scripts for general admin of an Arch
linux systems,
but it will apply to most other distros too. I'm intending on adding more
for RT tuning
and whatever I see fit as time goes on..
github.com/harryhaaren/scripts
PdCUDA
10 Feb
10 Feb
6:37 p.m.
New subject: kernel changes -was - So what do you think sucks about Linux audio ?
On Sun, 2013-02-10 at 09:25 -0800, Len Ovens wrote:
What I am says though is that I'm not sure these
problems are kernel
problems. They may be kernel module problems or something else. I have
used all of these kernel versions and been able to run low latency on all
of them no problems (ICE1712 based card). Other people on this list have
had no problems too. Yet there is not just one person with this same
problem. We (on this list) might do well to explore the differences
between our hardware/software to see if we can find the problem.
Hi Len :)
regarding to latency and xruns _on my machine_ two cheap ICE1712 PCI
cards are working better on Linux, than my RME PCI express card. Tested
with kernels from 2.6.29 full preempt on Debian to
$ uname -a
Linux q 3.6.5-rt14 #1 SMP PREEMPT RT Fri Nov 2 21:36:37 CET 2012 x86_64
x86_64 x86_64 GNU/Linux
$ cat /etc/issue
Ubuntu 12.10 \n \l
Regards,
Ralf
7:09 p.m.
New subject: kernel changes
On Sun, February 10, 2013 9:37 am, Ralf Mardorf wrote:
On Sun, 2013-02-10 at 09:25 -0800, Len Ovens wrote:
I am wondering if PCIe or it's drivers may be a problem for low latency.
Having read a little about the standard, it would seem that the earliest
standards (v1) are better for low latency than later (v4) standards. The
improvements have been in throughput at the expense of latency. I wonder
if it is possible to have a PCIe slot set to a particular version of the
PCIe standard. Or to do so at the card end. As a distro assembly
volunteer, I can only test things with the HW I have, which is older (no
PCIe slots at all). What I have read about the RME PCIe products is that
they are native PCIe rather than bridged PCI to PCIe.
I wish I had the funding to test full time and buy a better representation
of the HW out there, but I don't.
--
Len Ovens
www.OvenWerks.net
What I am says though is that I'm not sure
these problems are kernel
problems. They may be kernel module problems or something else. I have
used all of these kernel versions and been able to run low latency on
all
of them no problems (ICE1712 based card). Other people on this list have
had no problems too. Yet there is not just one person with this same
problem. We (on this list) might do well to explore the differences
between our hardware/software to see if we can find the problem.
Hi Len :)
regarding to latency and xruns _on my machine_ two cheap ICE1712 PCI
cards are working better on Linux, than my RME PCI express card. Tested
with kernels from 2.6.29 full preempt on Debian to
7:44 p.m.
New subject: kernel changes
On Sun, 2013-02-10 at 10:09 -0800, Len Ovens wrote:
On Sun, February 10, 2013 9:37 am, Ralf Mardorf wrote:
"HDSPe AIO is the newly developed PCI Express version of the HDSP 9632.
A newly developed genuine PCI Express core consequently takes full
advantage of the new format, achieving significant performance gains in
multitrack audio and lowest latency. Thanks to RME's secure flash update
technology, firmware improvements, adjustments, and bugfixes can be
installed easily at any time." -
http://www.rme-audio.de/en_products_hdspe_aio.php
For this card the Linux driver is very limited and I never updated the
firmware, so also possible is, that
- perhaps newer firmware is needed
- perhaps a better Linux driver is needed
- perhaps PCI on my mobo is good, but PCI express on the same mobo is
bad
*?*
Regards,
Ralf
On Sun, 2013-02-10 at 09:25 -0800, Len Ovens
wrote:
I am wondering if PCIe or it's drivers may be a problem for low latency.
Having read a little about the standard, it would seem that the earliest
standards (v1) are better for low latency than later (v4) standards. The
improvements have been in throughput at the expense of latency. I wonder
if it is possible to have a PCIe slot set to a particular version of the
PCIe standard. Or to do so at the card end. As a distro assembly
volunteer, I can only test things with the HW I have, which is older (no
PCIe slots at all). What I have read about the RME PCIe products is that
they are native PCIe rather than bridged PCI to PCIe.
I wish I had the funding to test full time and buy a better representation
of the HW out there, but I don't. What I am says though is that I'm not sure
these problems are kernel
problems. They may be kernel module problems or something else. I have
used all of these kernel versions and been able to run low latency on
all
of them no problems (ICE1712 based card). Other people on this list have
had no problems too. Yet there is not just one person with this same
problem. We (on this list) might do well to explore the differences
between our hardware/software to see if we can find the problem.
Hi Len :)
regarding to latency and xruns _on my machine_ two cheap ICE1712 PCI
cards are working better on Linux, than my RME PCI express card. Tested
with kernels from 2.6.29 full preempt on Debian to
11:58 p.m.
New subject: kernel changes
On Sun, February 10, 2013 10:44 am, Ralf Mardorf wrote:
"HDSPe AIO is the newly developed PCI Express
version of the HDSP 9632.
A newly developed genuine PCI Express core consequently takes full
advantage of the new format, achieving significant performance gains in
multitrack audio and lowest latency. Thanks to RME's secure flash update
technology, firmware improvements, adjustments, and bugfixes can be
installed easily at any time." -
http://www.rme-audio.de/en_products_hdspe_aio.php
Ya, I've read that bit before. That is advertizing copy. I would like to
know what the experience of users who installed this in a win/OSX machine
is.... picky users :)
For this card the Linux driver is very limited and I never updated the
firmware, so also possible is, that
Two linux drivers to consider. The PCIe stuff in the kernel is probably
optimized for throughput to make the best use of Video cards, fast ether
net, etc. Means larger chunks of data for less overhead. Maybe higher
latency too. It shouldn't be as it is faster than PCI which handled things
just fine. The extra throughput should actually help for higher channel
counts. (from reading though some of the docs on netjack)
- perhaps newer firmware is needed
Both ends of the PCIe IF have firmware.
- perhaps a better Linux driver is needed
Testing with windows is the only way to tell.
- perhaps PCI on my mobo is good, but PCI express on
the same mobo is
bad
I don't know how to judge that. I have lots of ideas :) but my knowledge
is limited. The IRQ logic on the new MB chipsets is interesting (to say
the least) as it has to handle a number of standards (from dos to now,
five or six)... I have not read the whole thing, just about the internal
audio stuff which may be useful as outputs in some situations because most
of them have 8... might make good headphone mixes. But while the outputs
are often 24 bit, the inputs are most often 16 and there are only two...
and the preamps are less than useful. Anyway, the MB glue logic may be of
some importance in all this too. All of our (linux based) audio with MB
research seems to be from the PCI era. We need something new.
--
Len Ovens
www.OvenWerks.net
11 Feb
11 Feb
12:03 a.m.
New subject: kernel changes
On Sun, Feb 10, 2013 at 5:58 PM, Len Ovens <len(a)ovenwerks.net> wrote:
RME devices do not load "firmware". What gets loaded is really just some
configuration data for the I/O box.
Two linux drivers to consider. The PCIe stuff in the kernel is probably
optimized for throughput to make the best use of Video cards, fast ether
net, etc. Means larger chunks of data for less overhead. Maybe higher
latency too. It shouldn't be as it is faster than PCI which handled things
just fine. The extra throughput should actually help for higher channel
counts. (from reading though some of the docs on netjack)
not really. the drivers for RME devices are PCI-species independent. the
PCI bus doesn't really exist for them, which is why the same driver can
interact with the (old) cardbus version as if it is directly on the bus.
drivers see address spaces and registers, not the bus.
- perhaps newer firmware is needed
Both ends of the PCIe IF have firmware.
1:38 a.m.
New subject: kernel changes
On Sun, February 10, 2013 3:03 pm, Paul Davis wrote:
On Sun, Feb 10, 2013 at 5:58 PM, Len Ovens
<len(a)ovenwerks.net> wrote:
Ok, so what I get from this is that the CPU sees a PCIe device in the same
way it sees a PCI device. Yet I know from reading the specs on both the
PCI to PCIe bridge and the MB chip sets, that there is some firmware
involved at both ends. This may not be something that is run by the CPU
itself (though it could be) but something that could be run by another
processor in the glue logic itself. It may rather be part of the bios. I
don't know. Most of the PCI(e) sound cards have a DSP with some sort of
firmware as well. How much all of this interacts I don't know. What I do
see is a lot of brand new MBs that handle sound badly.
--
Len Ovens
www.OvenWerks.net
Two linux drivers to consider. The PCIe stuff in the kernel is probably
optimized for throughput to make the best use of Video cards, fast ether
net, etc. Means larger chunks of data for less overhead. Maybe higher
latency too. It shouldn't be as it is faster than PCI which handled
things
just fine. The extra throughput should actually help for higher channel
counts. (from reading though some of the docs on netjack)
not really. the drivers for RME devices are PCI-species independent. the
PCI bus doesn't really exist for them, which is why the same driver can
interact with the (old) cardbus version as if it is directly on the bus.
drivers see address spaces and registers, not the bus.
2:05 a.m.
New subject: kernel changes
On Sun, Feb 10, 2013 at 7:38 PM, Len Ovens <len(a)ovenwerks.net> wrote:
On Sun, February 10, 2013 3:03 pm, Paul Davis wrote:
that would likely be in the chipset that interacts with the bus. device
drivers for particular PCI(.) devices do not interact with this level of
hardware (in general).
On Sun, Feb 10, 2013 at 5:58 PM, Len Ovens
<len(a)ovenwerks.net> wrote:
Ok, so what I get from this is that the CPU sees a PCIe device in the same
way it sees a PCI device. Yet I know from reading the specs on both the
PCI to PCIe bridge and the MB chip sets, that there is some firmware
involved at both ends.
Two linux drivers to consider. The PCIe stuff in the kernel is probably
optimized for throughput to make the best use of Video cards, fast ether
net, etc. Means larger chunks of data for less overhead. Maybe higher
latency too. It shouldn't be as it is faster than PCI which handled
things
just fine. The extra throughput should actually help for higher channel
counts. (from reading though some of the docs on netjack)
not really. the drivers for RME devices are PCI-species independent. the
PCI bus doesn't really exist for them, which is why the same driver can
interact with the (old) cardbus version as if it is directly on the bus.
drivers see address spaces and registers, not the bus. This may not be something that is run by the CPU
itself (though it could be) but something that could be run by another
processor in the glue logic itself. It may rather be part of the bios. I
don't know. Most of the PCI(e) sound cards have a DSP with some sort of
firmware as well.
RME devices do not. they use Xilinx FPGA chips. the firmware (the actual
firmware) is on an eprom that is surface mounted.
personally, i consider audio interfaces with onboard DSP to be a mostly-bad
design. if i wanted DSP, i'd buy some kind of processor for that. i don't
want a decade old DSP chip cluttering up my audio interface - my hammerfall
cards are now 13 years old and still working perfectly, unaffected by the
fact that DSP technology has moved on.
How much all of this interacts I don't know. What
I do
see is a lot of brand new MBs that handle sound badly.
i don't know of any MBs that in and of themselves handle sound badly (these
days, anyway). the problems come from a different set of levels (a few
notes on this (to be expanded):
http://manual.ardour.org/setting-up-your-system/the-right-computer-system-f…
--
Len Ovens
www.OvenWerks.net
3:16 a.m.
New subject: kernel changes
On Sun, 2013-02-10 at 20:05 -0500, Paul Davis wrote:
RME devices do not. they use Xilinx FPGA chips. the
firmware (the
actual firmware) is on an eprom that is surface mounted.
"Thanks to RME's secure flash update technology, firmware improvements,
adjustments, and bugfixes can be installed easily at any time." -
http://www.rme-audio.de/en_products_hdspe_aio.php
12 Feb
12 Feb
3:42 p.m.
New subject: kernel changes
On Sun, February 10, 2013 5:05 pm, Paul Davis wrote:
Ya, that was what I was thinking.
Nice page. Simple, but a good base for study. (just google everything in
there) From reading the HP proliant RT Docs I am guessing most audio work
is not as picky about lowlatency as the stock exchange. They suggest
things that will help but void your warranty.
I found out USB keyboards/mice are bad. Hyperthreading is bad (how do I
turn that off?) How many new MB come with USB as the only keyboard/rodent
input?
It appears that CPU speed control needs to be turned off both in bios as
well as in the OS.
Stay away from allinone CPU/audio/video chips which are SMI hell.
I have to try some of these things...
Anyway, it seems that the big thing with a computer hw upgrade is there
needs to be a lowlatency audit every time. (OS does not matter, this would
be the same for win/OSx)
--
Len Ovens
www.OvenWerks.net
On Sun, Feb 10, 2013 at 7:38 PM, Len Ovens
<len(a)ovenwerks.net> wrote:
>
> On Sun, February 10, 2013 3:03 pm, Paul Davis wrote:
> > On Sun, Feb 10, 2013 at 5:58 PM, Len Ovens <len(a)ovenwerks.net> wrote:
Ok, so what I
get from this is that the CPU sees a PCIe device in the
same
way it sees a PCI device. Yet I know from reading the specs on both the
PCI to PCIe bridge and the MB chip sets, that there is some firmware
involved at both ends.
that would likely be in the chipset that interacts with the bus. device
drivers for particular PCI(.) devices do not interact with this level of
hardware (in general). How much all
of this interacts I don't know. What I do
see is a lot of brand new MBs that handle sound badly.
i don't know of any MBs that in and of themselves handle sound badly
(these
days, anyway). the problems come from a different set of levels (a few
notes on this (to be expanded):
http://manual.ardour.org/setting-up-your-system/the-right-computer-system-f…
10 Feb
10 Feb
9:52 p.m.
New subject: So what do you think sucks about Linux audio ?
On Sun, 10 Feb 2013 08:14:32 -0800
"Len Ovens" <len(a)ovenwerks.net> wrote:
On Sun, February 10, 2013 2:57 am, Folderol wrote:
My DAW is 'air-gapped' from the Internet except when I'm doing upgrades, but
I'll connect it up sometime during the week and give this a try. FWIW my
soundcard is a venerable 2496
--
Will J Godfrey
http://www.musically.me.uk
Say you have a poem and I have a tune.
Exchange them and we can both have a poem, a tune, and a song.
One thing I find confusing is the way the kernel
performance changes so
dramatically from one version to the next.
In debian testing, 64bit V3.1 vanilla works beautifully on my current
machine, but V3.2 requires jack to be set to 4x the latency! That's for
both
vanilla & RT.
This may or not be it, but try looking at the irq assignment from one to
the next. The kernel sets IRQs for at least some of the devices.
11:31 p.m.
New subject: kernel changes
On Sun, February 10, 2013 12:52 pm, Folderol wrote:
On Sun, 10 Feb 2013 08:14:32 -0800
"Len Ovens" <len(a)ovenwerks.net> wrote:
>
> On Sun, February 10, 2013 2:57 am, Folderol wrote:
>
> > One thing I find confusing is the way the kernel performance changes
> so
> > dramatically from one version to the next.
> >
> > In debian testing, 64bit V3.1 vanilla works beautifully on my current
> > machine, but V3.2 requires jack to be set to 4x the latency! That's
> for
> > both
> > vanilla & RT.
3.2 was shipped with UbuntuStudio 12.04 (Lowlatency) and I have done at
least one recording project (more than 12 tracks) with jack set -p 64 and
no xruns at all. This is on my old P4 which has had the least tweaks for
audio of my two machines (my netbook needs lots to work at all). I had
sometime before in the 2.6 kernel era moved my audio IF to a cpi slot that
did not share irqs, but that is about it. I tried lots of things, but that
was what made the most difference. 12.10 went to kernel 3.5 (I think) and
now 13.04 is using 3.8. All my machines are 32bit, don't know if that
makes a difference, and while the netbook has more problems... it is atom
based at 1.6Ghz and only 1g ram. Internally it has a PCIe bus, but I don't
know that has anything to do with it's audio problems considering all it's
other limitations. It can record audio through a midi IF with jack set
-p64 as well, so it is not that bad. (forget internal sound for low
latency)
I'll connect it up sometime during the week and
give this a try. FWIW my
soundcard is a venerable 2496
PCI then, VIA/ICE17*. Is that the 1712 or the 172* (1720 to 1724 look the
same to the PCI IF)? 32bit system? or 64 (if 64 is the kernel 32bit)?
Newer (has PCIe HW as well as PCI)?
I wonder if any of the new USB2.0 IFs use the VIA1730, AudioBox 1818VSL
maybe? Seems to be the right number of i/o.
--
Len Ovens
www.OvenWerks.net
12 Feb
12 Feb
3:57 p.m.
New subject: kernel changes
On Sun, 10 Feb 2013 14:31:47 -0800
"Len Ovens" <len(a)ovenwerks.net> wrote:
On Sun, February 10, 2013 12:52 pm, Folderol wrote:
> On Sun, 10 Feb 2013 08:14:32 -0800
> "Len Ovens" <len(a)ovenwerks.net> wrote:
>
>>
>> On Sun, February 10, 2013 2:57 am, Folderol wrote:
>>
>> > One thing I find confusing is the way the kernel performance changes
>> so
>> > dramatically from one version to the next.
>> >
>> > In debian testing, 64bit V3.1 vanilla works beautifully on my current
>> > machine, but V3.2 requires jack to be set to 4x the latency! That's
>> for
>> > both
>> > vanilla & RT.
> I'll connect it up sometime during the week and give this a try. FWIW my
> soundcard is a venerable 2496
Having been laid low by a cold, I eventually got bored and had a look at this
again.
I'm running only 64bit OSs on a quad core AMD. The MOBO is an Asus one with a
uefi bios. I can't find anything on that bios that is relevant to irq or
tweakable in any way that might benefit sound.
I only have one PCI slot, and the soundcard comes out at irq20 which is below
the usb irqs (interestingly the on-board sound is above usb).
As a side issue I'm wondering if this fact is causing very occasional Xruns
from ALSA - When I say occasional, I mean maybe once or twice in an entire
afternoon session.
Double-checking again. The 3.2 kernel behaviour is appalling, regardless of
whether I use the vanilla one of the RT one, 3.1 is smooth as silk.
--
Will J Godfrey
http://www.musically.me.uk
Say you have a poem and I have a tune.
Exchange them and we can both have a poem, a tune, and a song.
10 Feb
10 Feb
1:24 p.m.
New subject: So what do you think sucks about Linux audio ?
On Sun, 2013-02-10 at 10:34 +0100, SxDx wrote:
This is soft male heterosexual pedo-pornography.
It includes an erotic component, that by the way has a lot of
heterosexual female fans too. The schema of childlike characteristics
isn't equal to pedophilia!
Pedophilic pornography and "normal" pornography are something completely
different, than what is shown by this picture.
You'll find Botticelli paintings that include a comparable kind of soft
erotic too, especially if you compare Botticelli with the gifted comic
drawers, e.g. compare paintings with Kosuke Fujishima's "Oh My
Goddess!".
Closer to the style of the old artists is Richard Corben. I wonder that
comic art often still is garbage, pornography etc. for some people.
Don't google for Kosuke Fujishima or Richard Corben, but take a look at
an art print of Kosuke Fujishima or take a look at the anatomy book of
Richard Corben.
Once people get aware of the talent those drawers have got, they
understand that this comic styles have less to do with a shortage of
niveau.
Regards,
Ralf
3:45 p.m.
New subject: So what do you think sucks about Linux audio ?
Haha, good thing you didn't see the Vocaloid booth at NAMM! Oh man.
But I do agree it's unprofessional for a Linux distribution to default
to that wallpaper (and are there copyright issues?).
On Sun, Feb 10, 2013 at 1:34 AM, SxDx <sed(a)free.fr> wrote:
On 02/10/2013 07:05 AM, hermann meyer wrote:
This is not art. This is soft male heterosexual pedo-pornography.
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user
Before you go on with dissing and bashing, you may wane have a look at
the picture in question:
http://kxstudio.sourceforge.net/tmp/scr124thumb.png
Besides from that I didn't like it as well, it is clearly just a "ART
Picture".
and here is a entry in Wikipedia which explain what the meaning of this
picture is:
http://en.wikipedia.org/wiki/Hatsune_Miku 
3:50 p.m.
New subject: So what do you think sucks about Linux audio ?
>>> Before you go on with dissing and
bashing, you may wane have a look at
>>> the picture in question:
>>> http://kxstudio.sourceforge.net/tmp/scr124thumb.png
> This is not art. This is soft male heterosexual
pedo-pornography.
This is just lol. What are you, religious? There are really backwards culture pockets on
this earth...
3:53 p.m.
New subject: So what do you think sucks about Linux audio ?
On 02/10/2013 04:34 AM, SxDx wrote:
This is not art. This is soft male heterosexual
pedo-pornography.
It may not be art, in the sense that the Mona Lisa is "art." But it
certainly isn't "soft male heterosexual pedo-pornography," either. Don't
be ridiculous.
4:53 p.m.
New subject: So what do you think sucks about Linux audio ?
2013/2/10 Ricardus Vincente <wizardofgosz(a)gmail.com>
On 02/10/2013 04:34 AM, SxDx wrote:
+1
..and does anybody remeber of Fujiko Mine? B-)
This is not art. This is soft male heterosexual
pedo-pornography.
It may not be art, in the sense that the Mona Lisa is "art." But it
certainly isn't "soft male heterosexual pedo-pornography," either.
Don't
be ridiculous.
5:50 p.m.
New subject: So what do you think sucks about Linux audio ?
On Sun, 2013-02-10 at 16:53 +0100, Raffaele Morelli wrote:
2013/2/10 Ricardus Vincente
<wizardofgosz(a)gmail.com>
On 02/10/2013 04:34 AM, SxDx wrote:
The skilled comic drawers do make art such as the Mona Lisa. I'm not a
Corben fan myself, but I've seen comics where each painting of a plot
was as sophisticated as a Mona Lisa, Sistine Chapel, Die Toteninsel,
etc.. As mentioned before
Regarding to mangas, there are also gifted drawers, who draw mangas, but
nobody does like to call them manga artists (since there are more
prejudices about mangas, than about other comics), I guess Katsuhiro
Otomo is the most known.
The component most people overinterpret is the sexual component. Reading
some manags from a female drawer might teach that manags are not only
about sexuality. I suspect Rumiko Takahashi is the most successful
female drawer on this planet.
Take a look at art prints from the "Oh My Godess!" drawer. I don't think
that Richard Corben and Kosuke Fujishima are less talented, than
Leonardo da Vinci. I've seen Corben's anatomy book, I don't know any old
master who was that good.
Out of context, just watching the pictures on a screen doesn't show the
high quality.
http://www.google.de/search?q=kosuke
+fujishima&hl=de&sa=X&tbo=u&tbm=isch&source=univ&ei=ps4XUa_5KMrMswb5h4CgDQ&ved=0CD8QsAQ&biw=1152&bih=729
Regards,
Ralf
This is not art. This is soft male heterosexual
pedo-pornography.
It may not be art, in the sense that the Mona Lisa is "art."
But it
certainly isn't "soft male heterosexual pedo-pornography,"
either. Don't
be ridiculous.
+1
..and does anybody remeber of Fujiko Mine? B-) 
10:33 a.m.
New subject: So what do you think sucks about Linux audio ?
As previously hinted at, the controversial wallpaper ships as default with
the latest KXstudio release.
falktx isn't on this list to explain for himself so this is how he explains
the wallpaper on the KX site:
"This release is dedicated to Hatsuke Miku, a [virtual] Vocaloid singer
that recently has seen original artwork released under a Creative Commons
license.
We believe that actions like this one (and others in 2012) indicate that
the world embraces open-source culture more and more, every day."
see http://kxstudio.sourceforge.net/News
I can see how some can see this wallpaper as slightly childish but
offensive? Really? Thats a sure sign of a low 'culture' tolerance threshold
if ever I heard one! If that upsets you,pretty much all modern media - ads
and music vids especially - are going to have you in tears of despair. I
can't realistically classify that picture as soft porn like most music vids
and many ads these days are, not that that'd be an improvement but its
below tame.
On Sun, Feb 10, 2013 at 6:05 AM, hermann meyer <brummer-(a)web.de> wrote:
Am 10.02.2013 06:37, schrieb hermann meyer:
Am 10.02.2013 03:58, schrieb david:
picture
is:
http://en.wikipedia.org/wiki/**Hatsune_Miku<http://en.wikipedia.org/wiki…
______________________________**_________________
Linux-audio-user mailing list
Linux-audio-user@lists.**linuxaudio.org<Linux-audio-user@lists.linuxaudio.org>
http://lists.linuxaudio.org/**listinfo/linux-audio-user<http://lists.lin…
On 02/09/2013 04:18 PM, Chris Bannister wrote:
Before you go on with dissing and bashing, you may wane have a look at
the picture in question:
http://kxstudio.sourceforge.**net/tmp/scr124thumb.png<http://kxstudio.so…
Besides from that I didn't like it as well, it is clearly just a "ART
Picture".
and here is a entry in Wikipedia which explain what the meaning of this On Fri, Feb 08, 2013 at 04:17:38PM -0500, shane
richards wrote:
> 4. And now to the biggest gripe. On my laptop, I decided to create
> a login for my partner so she could quickly check her mail without
> me having to unblock facebook etc. When I finally got her to log
> in, the DEFAULT BACKGROUND WALLPAPER HAD A HENTAI CHICK IN A
> SUBSERVIENCE POSE. WTF!? Partner not amused. Are all the distro
> people 15 year old losers?
>
Is this after an upgrade from Ubuntu Lucid? So what is the name of
Ubuntu that is installed on your machine now?
P.S. The person who did/condoned that should be relegated to the
back ranks and not have anything to do with public exposure/HR/PR
etc, etc.
P.P.S Always keep backups of your configuration files! :)
P.P.P.S Ubuntu should be informed, this doesn't sound like something the
Ubuntu business would like.
12:19 p.m.
New subject: So what do you think sucks about Linux audio ?
On Sun, 2013-02-10 at 07:05 +0100, hermann meyer wrote:
Am 10.02.2013 06:37, schrieb hermann meyer:
So it isn't a Hentai picture, since Hentai is porn. Of course "school
girls", "pretty magical girls", "vampires" etc. pp. especially in
mangas
and animes have a sexual side. Art always has a sexual side too, it's
close to courtship display, we don't need art for feeding etc..
Sexuality is in the eye of the beholder.
For me this picture isn't offensive and Linux audio doesn't suck,
regarding to bad default wallpapers.
;)
Am 10.02.2013 03:58, schrieb david:
and here is a entry in Wikipedia which explain what the meaning of this
picture is:
http://en.wikipedia.org/wiki/Hatsune_Miku On 02/09/2013 04:18 PM, Chris Bannister wrote:
Before you go on with dissing and bashing, you may wane have a look at
the picture in question:
http://kxstudio.sourceforge.net/tmp/scr124thumb.png
Besides from that I didn't like it as well, it is clearly just a "ART
Picture".
On Fri, Feb 08, 2013 at 04:17:38PM -0500, shane
richards wrote:
> 4. And now to the biggest gripe. On my laptop, I decided to create
> a login for my partner so she could quickly check her mail without
> me having to unblock facebook etc. When I finally got her to log
> in, the DEFAULT BACKGROUND WALLPAPER HAD A HENTAI CHICK IN A
> SUBSERVIENCE POSE. WTF!? Partner not amused. Are all the distro
> people 15 year old losers?
Is this after an upgrade from Ubuntu Lucid? So what is the name of
Ubuntu that is installed on your machine now?
P.S. The person who did/condoned that should be relegated to the
back ranks and not have anything to do with public exposure/HR/PR
etc, etc.
P.P.S Always keep backups of your configuration files! :)
P.P.P.S Ubuntu should be informed, this doesn't sound like something the
Ubuntu business would like.
4439
days inactive
4443
days old
linux-audio-user@lists.linuxaudio.org
42 comments
17 participants
participants (17)
-
Alexandre Prokoudine -
Charles Z Henry -
Chris Bannister -
Dan MacDonald -
david -
Folderol -
Harry van Haaren -
hermann meyer -
Len Ovens -
Louis Gorenfeld -
Nils Gey -
Paul Davis -
Raffaele Morelli -
Ralf Mardorf -
Ricardus Vincente -
shane richards -
SxDx