[00:45] <andyzweb> _guitarman_: I am running into those troubles with jack as well
[00:47] <andyzweb> only getting those errors with the realtime kernel
[00:49] <andyzweb> and no I am not -_-
[00:49] <andyzweb> *now
[00:57] <_guitarman_> andyzweb: is that because you rebooted?
[00:58] <_guitarman_> andyzweb: yes, mine works now
[01:04] <andyzweb> _guitarman_: yeah that's all I did.
[01:04] <andyzweb> _guitarman_: have you tried starting and stopping jack and seeing if it still works?
[01:04] <_guitarman_> will do
[01:05] <andyzweb> also in in Ardour are all the denormal handling options disabled or is that just me?
[01:05] <_guitarman_> its working for me now andyzweb 
[01:06] <_guitarman_> my system is now a real unclean mess though... i've installed kxstudio, so i can't vouch at this point how it is in ubuntustudio exactly since its now installed falk's stuff
[01:07] <_guitarman_> the pulseaudio to jack bridge is just awesome this solves a lot of problems for linux podcasters
[01:07] <_guitarman_> and of course for people who want to grab audio from vieos etc
[01:08] <andyzweb> yeah what is kxstudio? and are you running 10.04?
[01:08] <_guitarman_> yes 10.04
[01:08] <_guitarman_> kxstudio is based on ubuntu 10.04
[01:08] <_guitarman_> i was just tring it out so i could blog about it
[01:08] <_guitarman_> it uses all the cutting edge stuff
[01:08] <_guitarman_> jack2
[01:08] <_guitarman_> ladish
[01:08] <_guitarman_> the vst stuff
[01:09] <_guitarman_> so its not as stable
[01:09] <_guitarman_> based on kde
[01:09] <_guitarman_> for the desktop environment
[01:09] <_guitarman_> instead of gnome.
[01:09] <_guitarman_> its a nice setup, jack is on by default when it boots i believe
[01:09] <persia> Many of the technologies being previewed in kxstudio will appear in 10.10
[01:09] <_guitarman_> yeah, i think thats going to be a really killer release for ubuntustudio
[01:10] <_guitarman_> this stuff should have stabalized a bit more by then
[01:10] <_guitarman_> although so far it hasn't been terribly crashy
[01:11] <andyzweb> Yeah. I have given up on getting UbuntuStudio to work fully on my macbook for now.
[01:12] <andyzweb> I am playing with my other PC instead.
[01:13] <andyzweb> let me check something real quick
[01:19] <andyzweb> yeah, I get 5.8ms latency 4 in 2 out with the built in hardware on my macbook pro when @ 44.1 24-bit
[01:19] <andyzweb> I want to be able to get the same (or hopefully better) with UbuntuStudio on the same machine
[01:23] <_guitarman_> is the hardware supported?
[01:36] <andyzweb> welp...the internal sound card is the intel-HDA thing. which is very...difficult
[08:03] <jussi> Ive asked this guy to join here: 
[08:03] <jussi> [09:28:09] <macinnisrr> I have a distro for beta release which contains more current versions (and more convenient scripts) of software, as well as realtime kernels with PAE support. I can package quite well, but have no idea how to make a source package or contribute upstream. How can I help?
[08:24] <andyzweb> abogani: what is the difference between the linux-rt linux-realtime and linux-lowlatency kernels in 10.04?
[08:26] <abogani> andyzweb: -lowlatency use the same source of -generic one but a more aggressive configuration for reduce latency. -rt and -realtime use PREEMPT_RT patchset (that is linux realtime support) but the former is based on Ubuntu kernel source instead the later is based on the vanilla one (aka kernel.org).
[08:30] <andyzweb> abogani: ok I see. In the kernel.org rt project there are no patches for 2.6.32 was there no realtime kernel for that version?
[08:31] <abogani> andyzweb: No. It is the reason of -lowlatency kernel.
[08:34] <andyzweb> so -lowlatency just uses the Preemptable Kernel instead of the Voluntary Preemption which is enabled in -generic
[08:35] <abogani> andyzweb: That is one of the config changes.
[08:36] <andyzweb> abogani: without bothering you anymore, is there a diff or list of the config differences for them?
[08:37] <abogani> andyzweb: What is the reason of that request?
[08:38] <persia> abogani: Do we have word of -rt for 2.6.35 yet, or is that still for discussion?
[08:39] <andyzweb> abogani: just trying to grow my knowledge of kernel development and maintenance
[08:39] <abogani> persia: Will be 2.6.35 the final kernel version for Lucid?
[08:39] <abogani> Maverick
[08:39] <persia> Dunno.  it's the announced version from UDS, at least.
[08:39] <persia> -kernel would have more details though.
[08:42] <abogani> persia: At the moment the latest upstream rt release is 2.6.33. they (upstream) don't seem have intention to upgrade soon so we can: 1) stick on 2.6.33 2) forward port it on 2.6.35 (like I done for Intrepid and Jaunty).
[08:44] <persia> Forward-port is likely to be most compatible, of course.  There may be laggard topic branches that need to be supported.
[08:44] <abogani> andyzweb: Configurations changed in -lowlatency kernel flavour:
[08:44] <abogani> # CONFIG_NO_HZ is not set
[08:44] <abogani> CONFIG_PREEMPT=y
[08:44] <abogani> CONFIG_HZ_1000=y
[08:44] <abogani> CONFIG_HZ=1000
[08:44] <abogani> CONFIG_TREE_PREEMPT_RCU=y
[08:44] <abogani> CONFIG_SLAB=y
[08:44] <abogani> # CONFIG_DEBUG_KERNEL is not set
[08:44] <abogani> andyzweb: ^^
[08:45] <persia> andyzweb: The other trick is that all the kernels install the config to /boot on install, so you can run diff between arbitrary flavours.
[08:45] <andyzweb> abogani: many thanks!
[08:47] <abogani> andyzweb: You are welcome!
[08:50] <andyzweb> my usefulness will increase once I write my guide for installing UbuntuStudio 10.04 on a MacBook Pro 5,3.
[09:00] <abogani> persia: Frankly I think that the only one viable solution to -rt painful maintenance is to "adopt" a more tuned generic based kernel kernel like -preempt or -lowlatency. Then put it into main and use as default kernel for Studio. In this way we can let -rt for only people that really needs it. And avoid a lot of criticism (from community) to me.
[09:02] <persia> Oh, I agree entirely.  I'd like to have -rt in the archives, but I don't think it ought be default for Studio until it hits mainline: it's too high a volume of bug reports for the available maintenance team.
[10:49] <falktx> so?
[10:56] <jussi> hi falktx
[10:57] <jussi> youll need to talk to the packaging lads, perhaps persia can help
[10:58] <persia> With which?
[10:58] <falktx> backport packages to lucid
[10:58] <falktx> maybe through an ubuntustudio ppa
[11:00] <persia> I only know the official process: file a bug against the backports project, get three people to test, wait for it to be published.  This doesn't work for libraries at all.
[11:00] <persia> ScottL was talking with jdong about some potential process for more complex stuff, but I don't know the details.
[11:00] <persia> !backports
[11:00] <persia> That page should detail the official process, for some more detail.
[11:01] <persia> Given timezones, ScottL probably won't be around for a few hours yet.
[11:05] <falktx> that what i don't like about it
[11:05] <falktx> I have to wait for other people to review it
[11:06] <falktx> and that can take a really long time
[11:06] <falktx> upload to a ppa is quick
[11:07] <falktx> I have done this before so I already know how this works
[11:07] <falktx> (check the 'zynadd' package)
[11:08] <falktx> aren't you guys interested on a ubuntustudio ppa?
[11:09] <persia> There is one.  We just actively don't recommend it to users.
[11:10] <persia> Basically, anything that is going to get branded as part of what we do needs peer review anyway (even if just between each other).
[11:10] <persia> And the backports guys tend to be friendly to known groups.
[11:11] <persia> That's why I suggest waiting for ScottL: he has started a discussion to do something like upload to a PPA, and have that end up in backports for our stuff.  I just don't know the details, and it's not a complete process yet, so the documentation is lacking.
[11:11] <falktx> ok, that sound good
[11:12] <falktx> i'll be around here today
[11:12] <falktx> when will he come?
[11:15] <persia> He usually has some time early his morning, so near the end of the day UTC.  Worst case, he'll get the message in backscroll.
[11:24] <falktx> just one super-dumb question - how do i change irc text color ?
[11:25] <falktx> sometime ppl talk in red
[11:26] <persia> It's usually something in your client configuration.  Try looking for a Preferences menu or submenu, and see if there is a colours choice.  If you use a CLI IRC client, it's probably some /SET magic.
[11:26] <falktx> i use konversation
[11:28] <persia> I don't know for konversation.  With Quassel it's Settings....Configure Quassel...Chat view
[11:28] <persia> jussi: Do you remember where to configure that in konversation?
[11:29] <falktx> i saw it, custom colors
[11:29] <falktx> but how to enable it while typing?
[11:29] <jussi> Im sorry, been using quassel for too long
[11:29] <falktx> it's ok
[11:30] <persia> How do you mean?  To control what colors you send?  Lots of folks won't see that, because they show the colors client-side.
[11:30] <falktx> it's not like i need that
[11:31] <jussi> pretty much all of ubuntu channels have colour restrictions set also
[11:35] <astraljava> irssi FTW!!
[12:31]  * jussi gags astraljava
[14:41] <falktx> hi Scott
[14:42] <falktx> i was discussing with some guys here about the ubuntustudio ppa
[14:42] <falktx> to get a way to upload to that ppa, "auto-review", and send to backports if it's good enough
[14:43] <rlameiro> falktx: boas, por aqui :D
[14:44] <falktx> ya, ola
[14:44] <falktx> fui um dia pesquizar umas cenas...
[14:44] <falktx> e encontrei ...
[14:44] <rlameiro> falktx: i finished reading the channel logs, ScottL is in a trip of about 3 days
[14:44] <rlameiro> http://irclogs.ubuntu.com/2010/05/
[14:45] <rlameiro> we had a meeting last sunday and we talked about the backports issue, go into there and you can look at it
[14:45] <rlameiro> also, WELCOME MAN!!!!!
[14:45] <falktx> nice to be here
[14:46] <falktx> ja tou no estágio, por isso ja tenho net o dia todo
[14:46] <falktx> até as 6h30
[14:46] <falktx> which is the specific log date i should check ? (sunday...)
[14:47] <rlameiro> 16
[14:47] <rlameiro> sorry folks, we are both from Portugal :D
[14:49] <falktx> i'm reading the logs...
[14:49] <falktx> what do you guys think about kxstudio?
[14:49] <falktx> does it suck, or is it a good thing?
[14:56] <falktx> ok
[14:56] <rlameiro> falktx: i dont like so much kde, its a personal question, but kxstudio is a very nice build and has very neat features 
[14:57] <rlameiro> some things are planned to be on the next release, like jack2, for instance
[14:57] <falktx> everything could be ported to ubuntustudio
[14:57] <falktx> i saw the interest on pulse-jack
[14:57] <rlameiro> well, not everything
[14:58] <rlameiro> to get into ubuntustudio, the packages must be on the ubuntu repos
[14:58] <falktx> i decided to stick with lucid for at least a year
[14:58] <rlameiro> and for that they must get through the REVU process, maybe persia can help you in there
[14:58] <falktx> so I won't work for maverick
[14:59] <falktx> (always persia)
[14:59] <rlameiro> :D
[14:59] <falktx> the thing is:
[14:59] <falktx> i'll keep working on lucid
[14:59] <rlameiro> well he is a MOTU, so he can guide in there
[14:59] <falktx> whatever that thing is going to ubuntu or not, I don't think it's my job
[15:00] <falktx> i have many packages in revu
[15:00] <falktx> the only one that has been accepted is "zynadd"
[15:01] <falktx> dpkg -s zynadd | tail -n 1
[15:01] <rlameiro> well i dont know about that, but revu seems a very hard process to go through
[15:01] <falktx> _it's cool_
[15:01] <falktx> revu is all about waiting, i think
[15:01] <falktx> ppl are busy, i get it
[15:01] <rlameiro> well yea, but there are a lot of things to correct
[15:02] <rlameiro> well , time to give lesson
[15:02] <rlameiro> be back later
[15:02] <falktx> ok 
[15:02] <falktx> bye
[15:02]  * rlameiro AFK
[15:04] <falktx> i think ubuntustudio should have something like this:
[15:04] <falktx> http://kxstudio.sourceforge.net/screenshots/kcm_audio1.png
[15:04] <falktx> i did it for kxstudio, and it's very handy
[15:06] <rlameiro> https://wiki.ubuntu.com/UbuntuStudio/ControlsRedesign falktx 
[15:07] <falktx> @rlameiro: weren't going to give a class
[15:08] <falktx> ?
[15:09] <falktx> i can easily do that app in 3 weeks
[15:09] <falktx> using pyqt
[15:09] <falktx> just need to leave everything for later, which i might not do...
[15:10] <falktx> i would change some stuff though
[15:10] <falktx> shouldn't wireless be a task of networkmanager ?
[15:10] <falktx> nm-applet ?
[15:20] <persia> Historically we've used gnome-network in preference to network-manager, as it doesn't have a dynamic behaviour, so it's easier to predict system load.
[15:20] <falktx> so that's the reason
[15:21] <falktx> is networkmanager included in the dvd?
[15:22] <falktx> it should
[15:22] <persia> I don't think so.  There's a bug in lucid that it doesn't work at all, but that's different.
[15:22] <persia> Why, if it's considered to offer a degraded experience?
[15:22] <falktx> i sometimes use nm-applet even in kde
[15:23] <persia> It's *wonderful* for usecases that don't run into issues with a bit of extra latency once in a while, but this doesn't tend to be that use case.
[15:23] <falktx> it never failed me, and kde has some minor probs with networks
[15:23] <persia> It's not a reliability issue: it's a predictability issue.  With n-m, there exists no way to know "when" n-m will decide to consume processing resources.
[15:24] <persia> So one can't safely e.g. push one's machine near the limit in a performance.
[15:24] <falktx> i understand
[16:10] <rlameiro> persia: could this bug be a packaging issue? https://bugs.launchpad.net/bugs/581786
[16:25] <persia> rlameiro: It might be, but it might also be an accidental upstream release artifact creation bug.
[16:26] <persia> Check the debian package to see if the defaults are different.  Also, check to see if there is a config file in the upstream release tarball.
[16:26] <persia> It can be fixed in packaging, but it's worth tracking down the source of those values, and fixing it there so that we don't have to readjust every time we touch the package.
[16:32] <rlameiro> persia: the comments said that in ardour forum this issue was on the ubuntu package
[16:33] <persia> Sure, but do we know that it isn't also in the Debian packaging?  How about the upstream release tarball?
[16:34] <persia> That nobody ever encountered it there doesn't help us to know: maybe the Debian users all modify the configuration.  Maybe nobody actually uses the usptream example file.
[16:34] <rlameiro> well i will try to look into it later
[16:34] <persia> Thanks.  If it's really just an issue in Ubuntu, we'll want to check why it exists (maybe there's a bug in each direction, and it needs some other solution), and then just sync with Debian.
[17:49] <falktx> you're here too guitarm1n ?
[17:49] <falktx> ha
[18:03] <_guitarman_> yes - i am everywhere falktx ... rlameiro recruited me for the meeting to suggest some things
[18:03] <_guitarman_> since i use the tools in the podcast etc
[18:03] <_guitarman_> and for music