| TheMuso | ScottL: If you feel switching back to generic is safer for now, re ubiquity etc, fine by me. | 00:43 |
|---|---|---|
| len-nb | ScottL, don't know if you will see this tonight... just looking at: https://wiki.ubuntu.com/UbuntuStudio/Roadmap | 04:08 |
| len-nb | mudita24 desktop file is now there can be called done, lightdm configured... is there more to do? It is no longer broken the bug #922424 | 04:10 |
| ubottu | Launchpad bug 922424 in ubuntustudio-meta (Ubuntu) "lightdm won't start I have console(s) only" [Undecided,Fix released] https://launchpad.net/bugs/922424 | 04:10 |
| len-nb | is taken care of so this could be called done as well. | 04:10 |
| len-nb | ScottL, I have also thought about your comment about loading a number of files to configure my workflow application. It would allow easy adding of workflow when a new workflow is added, that would be what you meant... I'll do that but keep the same format so that one file could have more than one workflow if needed. | 04:20 |
| ScottL | hi len-nb , i'm about to go to bed, i just about have the menu where i like it, should have pictures tomorrow, just fighting a few tweaks | 04:59 |
| ScottL | len-nb, yes, i need to clean up the roadmap wiki, maybe this weekend | 05:00 |
| ScottL | i'm actually taking tomorrow (or is it today now :0 ) off from work | 05:00 |
| ScottL | i'll be spending part of the time with the family for spring break, but i should get some time tomorrow to work on stuff along with the weekend | 05:01 |
| ScottL | so, i think my next steps are complete the menu and push to bzr, work through your panel stuff (which shouldn't take near as long), and get the slideshow updated | 05:02 |
| micahg | ScottL: BTW, you'll want to update the headers for lowlatency to the pae ones if they're not already | 05:02 |
| ScottL | i really, really, really, really, really, really, really want to get all that and a few more things done this weekend | 05:02 |
| ScottL | micahg, oy vey, i had forgotten about that! | 05:03 |
| len-nb | ScottL, fun stuff. | 05:03 |
| ScottL | thank you very much, i'll do that now | 05:03 |
| micahg | just for i386 though | 05:03 |
| ScottL | aye, i386 only, good point, i'm tired and probably wouldn't have thought of that :P | 05:03 |
| ScottL | i've already dropped kino, which brought in phonon, kde-runtime, konqueror, dolphin, and a bunch of other crap | 05:04 |
| ScottL | errr, digikam | 05:04 |
| ScottL | not kino | 05:04 |
| ScottL | crasy stuff | 05:04 |
| ScottL | micahg, so, i see 'linux-headers-lowlatency-pae:i386' as a package in precise, i need the full name including the ':i386' part, no? | 05:07 |
| micahg | no | 05:08 |
| ScottL | okay, just 'linux-headers-lowlatency-pae' then | 05:09 |
| ScottL | maybe i should be more clear because i might be misleading you, uno momento, por favor | 05:09 |
| ScottL | these two lines are currently there: | 05:10 |
| ScottL | * (linux-headers-lowlatency) [i386] | 05:10 |
| ScottL | * (linux-headers-lowlatency) [amd64] | 05:10 |
| ScottL | i plan to replace the first line with... | 05:11 |
| ScottL | * (linux-headers-lowlatency-pae) [i386] | 05:11 |
| ScottL | not... | 05:11 |
| ScottL | * (linux-headers-lowlatency-pae:i386) [i386] | 05:11 |
| micahg | right | 05:11 |
| ScottL | righty-o | 05:12 |
| ScottL | doing it now then and going to bed, i've been up for almost 19 hours today | 05:13 |
| ScottL | micahg, when the meta files are finally updated, you can use bug #956250 for the lowlatency headers in the changelog | 05:20 |
| ubottu | Launchpad bug 956250 in livecd-rootfs (Ubuntu) "default to lowlatency kernel for ubuntustudio builds" [Undecided,Fix released] https://launchpad.net/bugs/956250 | 05:20 |
| ScottL | okay, goings to bed now, for reals | 05:20 |
| jussi | good morning all, anyone about? | 05:27 |
| jussi | ScottL: sure you arent here anymore? | 05:29 |
| astraljava | jussi: I am now. | 09:43 |
| ailo | I tried installing only ubuntu-minimal just to see if there would be any difference in performance. None really, even though the desktop did interfere with jack | 11:08 |
| ailo | I'd like to pick the whole system apart somehow to find out what it is. Perhaps try an older kernel | 11:09 |
| ailo | Trying out alessios realtime kernel too see how it performs as well | 11:10 |
| ailo | Massive difference in performance :(. I say :(, because I really thought -lowlatency was enough. Still wondering how the changes can be so big. -generic was never fully up to par, but for me -lowlatency has been very reliable all up until now | 11:18 |
| astraljava | ailo: The kernel would my primary suspect for sure, as it experiences most changes of the involved software atm. | 12:28 |
| ailo | astraljava: It would seem that different kernels perform extremely well with -lowlatency configs, while others are crap | 12:29 |
| ailo | I mean, some perform extremely well.. | 12:30 |
| ailo | I'm going to try building a 3.0 kernel to see if that makes any difference | 12:30 |
| ailo | With -lowlatency configs | 12:30 |
| astraljava | ailo: I don't know kernels that much, but I do understand how crucial their performance (and configuration overall) is to this sort of work flow. | 12:31 |
| ailo | astraljava: My experience is that most of the time, everything everything else does not seem to influence performance at all | 12:31 |
| astraljava | ailo: Well, jackd obviously has a huge influence to it. | 12:32 |
| ailo | But, I haven't really brought things down to a really basic and minimal level | 12:32 |
| ailo | astraljava: jackd1 and jackd2 are quite different too in some ways, and I don't expect either to have a negative influence on performance after an update. Would be very unlikely | 12:33 |
| ailo | While I'm doing this, I'm waiting to hear from the library. Going to have to study through all of high school math these next couple of months and do tests in order to be able to get into the university, which has been long overdue for me | 12:35 |
| astraljava | ailo: Sounds good, I wish I'd have time to carry my studies forward. | 12:36 |
| ailo | When the timing is right.. :) | 12:37 |
| jussi | astraljava: ScottL: I wanted to point this out: http://www.openoctave.org/oomidi_2011 | 12:44 |
| ScottL | jussi, they have done some great stuff, i've been keeping up with that loosely for over a year since i read an dave phillips article in linux journal about alex stone (a non linux guy) making his transition to linux on a mac | 12:51 |
| ScottL | alex stone is a classical composer and had certain requirements for number of instruments and sizes of samples/instruments in his libraries | 12:51 |
| astraljava | jussi: Noticed it on linkedin. Looks promising. Wish I'd have 10k, I could get the similar workstation than what'shername. | 12:52 |
| ScottL | he was using reaper and there was a great (and long) article of alex's transition into linux and the struggles (and victories) with certain obstacles | 12:52 |
| ScottL | i think about a year go he started with open octave with another bloke and they have done some amazing work | 12:53 |
| jussi | we should get it out of the kxstudio ppa and into the repos though... | 12:53 |
| ScottL | i've been curious how much coding alex does. or if he's the "user" working with a "dev" to flesh it out | 12:53 |
| ScottL | jussi, that would be a major coup, indeed | 12:54 |
| * jussi assigns... eerrr, who are our packagers now? | 12:54 | |
| astraljava | jussi: You have to let go of such a mindset. Anyone can package for [ubuntu|debian]. | 12:55 |
| astraljava | jussi: Well okay, maybe not you... *smirk* | 12:55 |
| jussi | astraljava: I still have ops here, I can SMITE you :P | 12:55 |
| astraljava | jussi: But yeah, I could talk to quadrispro, he could mentor me if I'd take on that task. | 12:56 |
| astraljava | jussi: But you wouldn't, would you?! *blink* *blink* | 12:56 |
| jussi | hehe | 12:56 |
| jussi | mind, I have had packages accepted into the archive before ;) | 12:56 |
| astraljava | Fair enough. | 12:57 |
| ScottL | jussi, can you help set the topic back proper? somehow i did that while talking to someone the other day | 13:00 |
| jussi | ScottL: paste what you want the topic to be | 13:01 |
| === ScottL changed the topic of #ubuntustudio-devel to: this is the wrong topic | ||
| === ScottL changed the topic of #ubuntustudio-devel to: Ubuntu Studio Development Channel | Ubuntu Studio code: https://code.launchpad.net/ubuntust?gs first Sunday of each month | ||
| ScottL | oh, that's probably how i did it then | 13:02 |
| ScottL | i'll fix the text later, i'd like to make a few changes to the menu settings that i realized how to fix this morning while i was half asleep laying in bed | 13:02 |
| astraljava | knome: Yet another interview a week from now. Spare me a couch seat the night between 22nd and 23rd? What shift is your wife on then? Are you even in town? :) | 14:14 |
| * astraljava wonders if knome already got a day job, he hasn't responded in 1h 26min, which is near-alarming considering he should be up by now. :D | 14:20 | |
| ailo | Just trying out a -generic 3.0 version. Works pretty well | 14:51 |
| ailo | I'm having big doubts about the current kernels. | 14:52 |
| ailo | ScottL: Is it too late to change to an older kernel? | 14:52 |
| ailo | I think it might be the best approach | 14:52 |
| holstein | ailo: the *-lolatency one? | 14:52 |
| ailo | I would assume -lowlatency performs better than -generic, but even an older -generic is better than what we have now | 14:53 |
| ailo | The ones we have now are not really usable | 14:53 |
| holstein | in what way? | 14:53 |
| holstein | RT? or other? | 14:53 |
| ailo | holstein: Have you tried the lates kernels? | 14:54 |
| ailo | xruns | 14:54 |
| holstein | i have used a few 3.x + ones | 14:54 |
| holstein | not with audio | 14:54 |
| holstein | and i have only fired up the live CD pre lowlatency kernel and tested with my FW device | 14:54 |
| ailo | It's a clear difference. 3.2 is not very usable for low latency audio (unless it's a realtime kernel) | 14:54 |
| ailo | At least for me | 14:55 |
| holstein | ailo: you tried with FW and your internal devices? | 14:55 |
| ailo | Only two so far, but just switching to an older kernel, on Precise, makes a huge difference | 14:55 |
| holstein | interesting | 14:55 |
| ailo | Presently running -generic, jackd -p 64, while compiling a kernel. No xruns | 14:56 |
| holstein | TBH... ive been just fine with *-generic | 14:56 |
| holstein | i think its more about transitioning folks for me | 14:56 |
| holstein | and trying to get a nice "middle of the road" kernel | 14:56 |
| holstein | folks dont try it.. they just see that we dont have an RT kernel and move on | 14:57 |
| ailo | I've been testing the difference between -generic and -lowlatency, and the results have clearly shown that -generic was not able to run stable at low latencies | 14:57 |
| holstein | i would have done hte same | 14:57 |
| ailo | So, I wouldn't say -generic is fine | 14:57 |
| ailo | At the moment I'm running a mainline kernel, not an Ubuntu one | 14:57 |
| holstein | well... im not using it on my production rig, thats for sure | 14:58 |
| holstein | but for "normal" use... | 14:58 |
| holstein | whatever that is | 14:58 |
| ailo | I switched to -p 32. Still no xruns | 14:58 |
| ailo | Never had this good results from a -generic kernel before | 14:58 |
| holstein | yeah... thats where i hope we are heading | 14:58 |
| ailo | https://wiki.ubuntu.com/Kernel/MainlineBuilds?action=show&redirect=KernelMainlineBuilds | 14:58 |
| holstein | by the time 16.04 releases, folks will *actually* believe it maybe ;) | 14:59 |
| ailo | holstein: Well, it's not heading that way, since 3.2 is totally unusable | 14:59 |
| holstein | i think thats temporary though | 14:59 |
| ailo | There's no garantee that a kernel will become good for audio | 14:59 |
| holstein | i mean... what the hell do i know... but i think its just a glitch | 14:59 |
| ailo | That's not their major preference, when they make it | 14:59 |
| holstein | i dont think the 2 are mutually exclusive though | 15:00 |
| ailo | -p 16 now.. | 15:00 |
| holstein | sweet! | 15:00 |
| ailo | It's like a freaking -realtime | 15:00 |
| ailo | Actually better than the -realtime 3.2 | 15:01 |
| holstein | yup | 15:01 |
| ailo | Well, not really. Doesn't hold water when I start audio apps, but usually I would get xruns even without opening any apps | 15:03 |
| holstein | depends on the apps too, for me | 15:04 |
| holstein | some of them are "friendlier" | 15:04 |
| ailo | Ardour is the best performing in my experience | 15:04 |
| holstein | ardour occasionally kicks one up when starting, but yeah | 15:04 |
| holstein | usually the best for me too | 15:04 |
| ailo | I had a lot of xruns at -p 64, so not usable | 15:05 |
| ailo | I'm compiling a -lowlatency 3.0 now, so let's hope that one performs better. It did very much so in the past | 15:05 |
| holstein | i havent ever compiled a kernel | 15:06 |
| holstein | its one of those things i think i should do | 15:06 |
| ailo | Not that hard really. If you just recompile a kernel, it's super easy | 15:06 |
| holstein | im not even sure how to start | 15:06 |
| holstein | ive read about it a lot | 15:06 |
| holstein | i need a .deb ;) | 15:06 |
| holstein | sudo dpki -i whatever.deb | 15:07 |
| holstein | thats about where i am | 15:07 |
| holstein | anyways... im always trying to learn things | 15:07 |
| ailo | holstein: https://wiki.edubuntu.org/Kernel/BuildYourOwnKernel | 15:07 |
| ailo | The interesting part is where you change the config | 15:08 |
| holstein | yeah... that looks easy enough | 15:08 |
| ailo | There are a lot of tutorials on the web, so if you want to do something specific, chances are there's a tutorial for that | 15:09 |
| holstein | i have a nice book to that has an overview and some screenies | 15:10 |
| holstein | "how linux works' | 15:10 |
| holstein | im interested in "streamlining" a kernel | 15:11 |
| holstein | for certain hardware | 15:11 |
| holstein | maybe making tweaks for certain hardware or whatever if needed | 15:12 |
| ailo | That would be cool, yes | 15:12 |
| ailo | For one individual, you might need almost nothing of what the kernel offers | 15:12 |
| holstein | i have found that with some older hardware, the ubuntu kernels dont support it well | 15:13 |
| holstein | in some cases at least | 15:13 |
| ailo | But I think most of it isn't reducing performance though, so it's more about understanding what does, I'm thinking | 15:13 |
| holstein | and i know its something i can "fix" | 15:13 |
| ailo | You mean drivers? | 15:13 |
| holstein | since its all open and i can build whatever (in theory) | 15:13 |
| holstein | ailo: i think... | 15:14 |
| ailo | Like drivers for wifi and such? | 15:14 |
| holstein | that too.. but even just things that have been taken out | 15:14 |
| holstein | some old graphics hardware | 15:14 |
| ailo | Sometimes I guess because of legal issues | 15:14 |
| holstein | yup.. or just too old | 15:14 |
| ailo | Right. 3.0 -lowlatency built. Now to test it.. | 16:27 |
| ailo | Nope. Not better. | 16:32 |
| ailo | The only hardware change for me is my graphic card. I should try changing that, and also retry with an older Ubuntu to see the differences | 16:32 |
| ailo | holstein: If you get the chance, please do some quick testing to see how well jack plays on your hardware | 16:33 |
| holstein | ailo: will do | 16:35 |
| ailo | hmm, not sure if the -lowlatency is configured properly | 16:35 |
| ailo | I just copied the config that came with the latest kernel | 16:35 |
| knome | astraljava, hahah, we have guests | 16:37 |
| knome | astraljava, let me check that dat | 16:37 |
| knome | *day | 16:37 |
| knome | or *date | 16:37 |
| knome | astraljava, evening + morning, so not the best. but i'll ask her. :) | 16:37 |
| ailo | ScottL: TheMuso holstein: I believe the config for -lowlatency is actually the config for -generic | 16:38 |
| ailo | cat /boot/config-3.2.0-18-lowlatency | grep PREEMPT | 16:39 |
| ailo | # CONFIG_PREEMPT_RCU is not set | 16:39 |
| ailo | CONFIG_PREEMPT_NOTIFIERS=y | 16:39 |
| ailo | # CONFIG_PREEMPT_NONE is not set | 16:39 |
| ailo | CONFIG_PREEMPT_VOLUNTARY=y | 16:39 |
| ailo | # CONFIG_PREEMPT is not set | 16:39 |
| ailo | holstein: No reason to do any tests yet, then | 16:40 |
| holstein | yeah... | 16:40 |
| ailo | I'm going to recompile it and add the -lowlatency stuff to it, and see what that does to it | 16:40 |
| ailo | Later tonight. Will be back with results.. | 16:41 |
| ailo | Sometimes the obvious is a little too obvious | 16:43 |
| astraljava | knome: No worries, I'll try some other friends. Thanks, though! | 17:47 |
| ailo | TheMuso: Who are so far involved in maintaing the -lowlatency? | 18:09 |
| ailo | Besides abogani and yourself | 18:09 |
| astraljava | ailo: I think I volunteered at some point when Scott asked. | 18:15 |
| ailo | astraljava: I'm just wondering who else might be changing the source so far | 18:22 |
| astraljava | ailo: Let me check from the repository logs. | 18:23 |
| ailo | I have volunteered to take over maintenance from abogani, but I will need to learn a few things first. It'll take me a few weeks | 18:23 |
| ailo | astraljava: Which repo? The git repo? | 18:23 |
| ailo | Ah, the package log? | 18:23 |
| astraljava | ailo: The branch in LP where the package is rolled from. | 18:27 |
| astraljava | So far only Alessio and Luke have committed to it. | 18:27 |
| ailo | astraljava: LP? | 18:29 |
| astraljava | Launchpad. | 18:29 |
| astraljava | https://code.launchpad.net/~ubuntu-branches/ubuntu/precise/linux-lowlatency/precise | 18:29 |
| astraljava | Erm... that's one of those Importer branches. | 18:31 |
| ailo | It's the right package | 18:31 |
| ailo | I have a lot to learn about the whole package procedure when it comes to Ubuntu | 18:32 |
| ailo | Or, the right source :) | 18:33 |
| astraljava | Yeah but it's not the one where maintainers push. | 18:34 |
| ailo | I believe after Luke changed it, the lowlatency config was no longer used in the build | 18:34 |
| astraljava | I'm a little bit behind when it comes to UDD, too. | 18:35 |
| ailo | astraljava: I believe the maintainers push to the git repo for the specific kernel | 18:38 |
| ailo | Still need to understand how they turn the source into a package | 18:38 |
| astraljava | ailo: No I meant the lowlatency maintainers for Studio, not just the kernel maintainers. | 18:39 |
| ailo | astraljava: Same thing | 18:39 |
| ailo | There's a repo for lowlatency too | 18:39 |
| astraljava | ailo: Outside of LP? Oh, nice. | 18:40 |
| astraljava | I don't understand why, though. Isn't it supposed to be just a configuration issue mainly? | 18:40 |
| ailo | astraljava: http://kernel.ubuntu.com/git | 18:40 |
| astraljava | ailo: Yea, but that's for UKT only, isn't it? | 18:41 |
| ailo | astraljava: Only the /ubuntu kernels | 18:41 |
| ailo | abogani has been keeping his -lowlatency there the whole time | 18:42 |
| ailo | Don't know what the policy is, but there are a few custom kernels there | 18:42 |
| ailo | Ah, Luke has made his own -lowlatency repo there | 18:43 |
| astraljava | Right, good to know. (I'm afraid I was supposed to know that, due to said volunteering, but I suppose I was for back-up purposes anyway) | 18:43 |
| ailo | Crossing my fingers that the actual -lowlatency will perform as expected :P. Still building.. | 18:46 |
| ailo | astraljava: Yeah, I don't really see why -lowlatency needs to be in its own repo either tbh | 18:52 |
| ailo | I guess cause UTK is not going to maintain it | 18:53 |
| astraljava | That's the reason, yeah. | 19:11 |
| knome | astraljava, okay. if anything else fails though... :) | 19:19 |
| astraljava | knome: In the case of emergency, I'll crash in at my brother's. :) | 19:21 |
| ailo | ScottL: astraljava TheMuso holstein: -lowlatency rocks. Let's fix the config for it, and we have a killer kernel for audio | 19:30 |
| ailo | I'm just doing some testing, and I can instantly see that it's about as good as -realtime | 19:32 |
| astraljava | Sweet! | 19:33 |
| ailo | Seems like qjackctl is totally tied into jackdbus? I don't understand. I thought I started jackd1 with it yesterday | 19:39 |
| ailo | I'm not able to start jackd1 with qjackctl. It tries to start jackdbus | 19:39 |
| ailo | I'm on regular Ubuntu btw | 19:40 |
| astraljava | I can't really comment on that so far, my knowledge as a user of these apps is too limited. | 19:42 |
| astraljava | I can look into it tomorrow, though. | 19:43 |
| falktx__ | ailo: disable dbus integration in it's settings | 19:49 |
| ailo | falktx__: .config/rncbc.org/QjackCtl.conf? | 19:53 |
| falktx__ | no, the GUI | 19:54 |
| falktx__ | it's an option in the last tab | 19:54 |
| ailo | Ah, I see it now | 19:54 |
| ailo | New to me | 19:54 |
| astraljava | ailo: On that k.u.c/git page I see no other lowlatency repos except Alessio's and Luke's, so I'm still not sure where it's actually maintained, or how. | 19:59 |
| astraljava | Perhaps commits are cherry-picked from those by the importer. | 20:01 |
| ailo | astraljava: abogani had his own repo for the launchpad kernels, and then I assume Luke started his own in order to make the source buildable for the main repo (something about symbolic links) | 20:05 |
| ailo | The abogani tree is probably not updated anymore | 20:05 |
| astraljava | Ok. So probably Luke cherry-picks from others' repos if need be, then uploads. That would work. | 20:07 |
| ailo | astraljava: I think Luke based his on aboganis (not sure), while it might be better to either have a very light type of repo (like a patch for the standard ubuntu kernel), or an exact copy. At least that's what I'm thinking | 20:08 |
| astraljava | ailo: Yes, well I'm not sure how it works, but probably it would include most of Ubuntu's patches, except where it'd interfere with lowlatency configuration, and then just the different configuration. That's _my_ thinking, I'm waiting to be proven wrong. | 20:10 |
| knome | astraljava, whatever goes. planning to have any free time? | 20:16 |
| astraljava | knome: I can, if there's something interesting. There would be a cool gig to go to, but I most likely don't have the cash. | 20:21 |
| knome | astraljava, what gig? | 20:21 |
| astraljava | Pain of Salvation on Saturday evening. | 20:21 |
| knome | ouch. | 20:21 |
| knome | that's not going to be cheap, not at all | 20:21 |
| astraljava | It's not that much, 22 €, but in my situation, I don't think I can shelve that much. | 20:22 |
| knome | hmm, that's less than i thought. i was thinking 30€+ ... | 20:22 |
| astraljava | They're not that popular. | 20:23 |
| knome | mmh. but they're not finnish :) | 20:23 |
| astraljava | Sure ain't. | 20:23 |
| knome | yeah, that brings some extra to the ticket prices | 20:24 |
| astraljava | Depends. I'm pretty sure Stam1na costs a lot more than most of the swedes I'm listening to. | 20:25 |
| knome | probably, i don't know much about the metal scene | 20:25 |
| astraljava | It's not the most expensive genre, how could it be. :) | 20:25 |
| knome | yeah. probably not. no grand pianos to fly around. | 20:26 |
| knome | btw, lol, do you agree with wife's brother on this metal aphorism: if it has synth, it's melodic | 20:27 |
| astraljava | Sort of. Usually it means that. It does not rule out melodic metal without synths, though. | 20:33 |
| astraljava | And please do note that melodic metal does not equate not-grunting-vocals. :) | 20:38 |
| knome | yeah, i know, but i don't completely agree with the aphorism either... | 20:39 |
| astraljava | Why, do you have contradicting examples? | 20:41 |
| knome | i can't name any, but i know there is | 20:41 |
| astraljava | Like I said, it usually means that, but it's not fool-proof either. | 20:46 |
| astraljava | But then I haven't heard most of the metal bands in the world, and certainly don't even listen to that many of what's left on a regular basis either. | 20:47 |
| knome | :) | 20:47 |
| ailo | astraljava: I guess the trick is to know if and which patches would interfer. So far it seems to me it's mostly about the kernel version itself + the lowlatency config | 21:01 |
| len_ | Hmm, no 32 bit build today. | 21:04 |
| ailo | holstein: -p32 -r48000 for an hour or so. Playing rakarrack. No xruns. | 21:43 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!