[00:01] ScottL, That looks like the beginning of an extensive manual, sort of. Some of things will thankfully not require editing between releases like explaining FOSS and open formats. [00:02] ScottL, Should we do something similar to what Ubuntu does for their documentation, a release based manual? Most of it may just be a matter of copy and paste. [00:03] Once the first one is written down, of course [00:42] ailo, what else do you think needs to be removed from the seeds...i agree with padevchooser [00:44] ailo, right, the beginning of the documentation is a bit extension and i would concede that it is elementary to us that use ubuntu studio already [00:44] ailo, but i'm really hoping it lowers the threshold for people to become users (and hopefully end up helping us ;) ) [00:46] ScottL: http://dl.dropbox.com/u/1333955/us_audio_stream.png [00:47] inkscape file http://dl.dropbox.com/u/1333955/us_audio_stream.svg [00:48] rlameiro, cool, this will be very helpful to help people understand [00:48] i was thinking about using colored rectangles :P [00:49] yours is a lot nicer :) [00:49] I will send you a tg.gz with all the source files [00:50] ScottL: http://dl.dropbox.com/u/1333955/us-image-files.tar.gz [00:51] I started a script for a video [00:51] I will write it down on the wiki now [01:06] ScottL: I added some stuff to the wiki. Please chack my english, i know i suck at it... :D [01:07] bah, you and aboganni worry about your english and you guys rock...both of you write english better than many people speak it here :P [01:07] certainly better than i can read, write, or speak italian or portuguese ;) [01:08] but i'll check nonetheless :D [01:08] thanks :D [01:33] http://dl.dropbox.com/u/1333955/jack_routing.png [01:36] well i did the second depiction [01:36] now maybe i need to talk about multiple inputs? or not? [01:37] or maybe multiple outputs and inputs [02:13] ScottL: Sorry already done. [02:49] howdy [03:34] TheMuso, that's alright, it hasen't broken anything up till this point :P [03:34] i'll make notes to get it during ocelot [03:43] and thanks for updating the seeds, TheMuso [04:05] np [08:33] Hey gang... sorry for my absence! Life has been freakin' crazy, what with the divorce and moving and stuff. Ready to come back now, though. [08:34] I assume at this point of devel cycle it's about .iso testing and stuff. [08:34] I will take a look at what is underway, but are we still having meetings and stuff? [08:57] divorce? [08:59] Yes. [09:14] I'm sorry. [10:07] Thanks. [11:06] ScottL: ping [11:15] ScottL: There are the things that I should do theoretically: 1) Investigate on -lowlatency bug reported by Ronan, 2) work on 2.6.39-lowlatency (irqs threads), 3) Update the stable -realtime kernel 4) work on the beta version of the 2.6.38 realtime kernel. What should I do? [11:16] ScottL: Please let me the main priority. [11:16] *know [12:56] abogani, isn't #3 keeping people from installing he -realtime kernel from your ppa (or is it the -rt kernel in your ppa)? [12:58] as much as i want progress on #2, i would say that this is the lowest priority currently, since we will not be using this until ocelot (right?) [13:00] do we really absolutely need a 2.6.38 realtime kernel? it will not be put into the official repos and you already have a relatively current -rt kernel in your ppa, right? (therefore, we are not denying anybody a -rt kernel, just not the absolutely latest one) [13:00] i probably would consider #4 last then [13:01] abogani, so most likely look at #1 unless any of my assertions for the considerations are wrong [13:01] BUT, which do you think is most important, abogani ? [13:01] and, good morning, abogani :) [13:02] ScottL: Good morning to you. [13:02] :-) [13:05] The mailing-list thread showed very little interest on 2.6.39-lowlatency kernel. In fact last emails seems show a decreased interest on -lowlatency kernel in general. [13:06] So I think that we should drop -lowlatency kernel *completely*. [13:07] ScottL: ^ [13:17] abogani, i would agree with you if we can get 2.6.39 to include the ability to adjust irq [13:17] i think the addition of this functionality is probably going to outweigh any performance increase between -generic and -lowlatency [13:23] Ocelot'll have that feature for sure. [13:27] * abogani wonders if anyone would have objections about drop -lowlatency kernel [13:27] abogani, I'm will. I want someone to prove we can use -generic for realtime audio first. Still, I don't see any. [13:28] ailo: I would let you know that I *completely* agree with you. [13:28] but seem to me that no one use this kernel. In fact the only tester is you. [13:29] abogani, Are you sure? One reason could be that it is not in the main repo. [13:30] abogani, I think -lowlatency is reliable to a high degree, while -generic has not been. Perhaps the irq threading part will make -generic more reliable? [13:32] Obviously I couldn't be sure. But people should know that if they don't provide help for "out of archive" things these could disapear abruptly. And AFAIK the most of the work on -lowlatency is made by you. [13:33] It'll sure do but I don't know how much. [13:37] abogani, Maybe I'm just being a little pedantic then. I suppose Natty will not change much now, so -generic should probably do quite well. [13:38] ailo: You aren't pedantic at all. [13:38] ailo: I have ever found your opinions interesting. [13:41] ailo: It isn't technical thing: the lowlatency kernel is better than -generic one. No one can prove the contrary. But not matter how much is better if no one use it. [13:41] So I'm +1 to drop -lowlatency completely [13:43] At the end the UKT will be very happy about my failure :) [13:47] ailo, ScottL ? [13:48] I think the problem is that Linux enthusiasts will rather build their own kernel just because they can, or use -realtime because they can (except for some firewire people), while new users and people who just want a functional system would rather just push a button to get everything set. Not many know about -lowlatency yet. [13:49] But, if 2.6.39 will change that, this is only a matter for Natty. [13:51] There are some who use the wiki to install Ubuntu Studio, or just some packages for it. If we included a PPA there, people would be more likely to install -lowlatency. [13:52] Hopefully -generic is as balanced on other systems as on mine, in which case the difference will not be that huge. [13:59] I think I would be interested in building -lowlatency on my own PPA, just for testing purposes. Especially now that 2.6.39 is coming out. I will want to host my version of the -controls there too, but it will take a while before that happens. On -lowlatency maybe you could help me get going on that ScottL? [14:01] ScottL: I guess this is a good place to start https://wiki.ubuntu.com/ScottLavender/scratchpad [15:01] astraljava: i'm sorry to hear about your divorce :( [15:02] i missed seeing that this morning [15:02] * scott-work was sleepy eyed and only saw abogani 's red text [15:03] to answer your questions: [15:03] yes we are mostly testing [15:03] and we have started meetings again with the intent that they become regular [15:06] morning :) [15:19] morning holstein :) [15:21] * scott-work is working his way back through IRC log [15:22] abogani: i would still like to compare -generic vs. -lowlatency before we decide to drop anything [15:23] abogani: i think -generic handling irq conflicts is a part of this evaluation [15:23] abogani: but if -lowlatency can provide reliably better low latencies, then we would be hard pressed to drop -lowlatency [15:24] obviously [15:25] abogani: and if we decide that -lowlatency provides better functionality we will continue to push it into the repos and it will be the default kernel [15:25] abogani: see eseentially every ubuntu studio user will be testing it :) [15:26] abogani: and since -lowlatency is built off of the -generic, then the -lowlatency will be able to adjust irq conflicts as well, no? [15:27] abogani: after building the 2.6.38 -lowlatency kernel i was going to do more testing but since you pointed out the 2.6.39 kernel i felt we should probably wait [15:27] especially since we will not get the -lowlatency kernel into natty [15:29] abogani: but i commit to you, i will assist in testing 2.6.39 -lowlatency kernel for ocelot and i will aggressively pursue getting it into the repos if it provides appreciably better performance than -generic [15:29] * scott-work admist that he is sorely tempted to pursue getting -lowlatency into the repos if for no other reason that a lot of work has been already invested into it :P [15:29] scott-work: I'm with you too [15:30] hi falktx :) [15:30] scott-work, I can agree on testing 2.6.39, -lowlatency vs -generic. We still need to work out how to use irq priorities. It seems the guy who wrote the rtirq is testing that kernel now, so hopefully he will solve that. I learned that it is possible to change priorities for irq manually too [15:30] with the lack of a recent realtime kernel, we really need that lowlatency working [15:31] * scott-work is still working through IRC logs [15:31] ailo: i will provide any assistance i can for building the kernel, although i'm very much inexperienced at this point [15:32] ailo: but that wiki page you linked is pretty much a step-by-step guide [15:32] ailo: the only thing i haven't put in there is to avoid using a dash "-" in the version number, rather you should use the tilde "~" [15:33] scott-work, It's a good start at least. Hopefully we won't need the -lowlatency, in which case I won't need to learn any more than that. [15:34] falktx: agreed, except if the -generic continues it's improvements and the delta between -generic and -lowlatency continues to reduce [15:35] as well [15:35] but at this point i would emphatically say that i feel the -lowlatency is the best path at the moment [15:36] at least unless more testing is done between -lowlatency and -generic, but for 2.6.39 [15:36] scott-work: isn't natty using 2.6.38 by default ? [15:36] falktx: i believe it is [15:37] and -generic at that [15:37] scott-work: why is 2.6.39 so important? [15:38] falktx, Right now I am getting failry good results with 2.6.38 -generic. I don't feel 100% assured by it, since it is not only the kernel that decides latency, it would seem, while with -lowlatency you get more stable performance. [15:38] falktx, On 2.6.39 they added another realtime feature to adjust irq priority [15:39] I think it's probably better to stick with the normal ubuntu kernel, since it's better supported [15:39] falktx, Well, -lowlatency is the "normal" kernel too [15:39] hehe [15:41] falktx: yes, ailo is correct, aboganni starts with the -generic kernel, does NOT apply any patches, and the only changes are run time compile flags [15:41] falktx: and viola...you have the -lowlatency kernel [15:41] some things we aren't thinking about probably is how the magic "100 line patch" will affect the kernel's performance or if audio users will even notice it [15:42] scott-work, I highly doubt it from what I read about it [15:42] scott-work: I checked abogani git repo a few days ago. I think I can make the 2.6.38 kernel work for lucid and maverick [15:43] For curious: http://kernel.ubuntu.com/git?p=abogani/ubuntu-natty-lowlatency.git;a=blob;f=debian.lowlatency/config/config.flavour.lowlatency;h=208ac36381710432d2bfd39fd55b3f116ac18bea;hb=HEAD [15:43] scott-work: Thanks! Alright, I'll get right on with it as soon as I get home tonight. [15:44] falktx, I think it would be great if also things like nvidia drivers would work with that too. [15:44] falktx: and sadly it would seem that the Ubuntu Kernel Team (UKT) has been incredibly misinformed as well about how to build a -lowlatency kernel, despite aboganni's assurances of how he made it :/ [15:45] ailo: falktx abogani : can someone explain what is necessary to get nvidia drivers (and/or ATI ?) to work with kernels? [15:45] scott-work: for lowlatency, nothing [15:46] Changes are necessary only for the -realtime one. [15:46] scott-work: problem is the realtime kernel headers, lowlatency has no issues with drivers [15:46] yep [15:46] falktx, But, they need to be built against the kernel, right? So, you would need to build those too? [15:46] scott-work: the only issue can be that nvidia does not support the kernel yet (as ATI does for 2.6.38) [15:46] ailo: dkms takes care of that [15:46] falktx, Err, I have nvidia drivers working on 2.6.38 [15:47] ailo: I said ATI [15:47] afaik, canonical is working with ATI now [15:55] so if we went with either -generic or -lowlatency there would be no further work required for video drivers then? [15:55] either for nvidia or ati? [16:04] falktx, When you say dkms, won't you still need to build the drivers specifically for that kernel version? [16:06] scott-work: yes [16:06] that is awesome! [16:06] ailo: ah, yes. I meant that dkms usually does this job for us [16:07] ailo: I think there's a way to force dkms to build for all kernel versions. this way user don't get surprises [16:07] between not needing to rebuild video drivers and a possible elimination of the -rt kernel, those are some serious improvements :) [16:07] So, for Lucid and Maverick, using 2.6.38 will at least mean that one needs to build graphic drivers for those. I wonder what else will be affected? [16:08] ailo: I have a KXStudio-Team Kernel PPA now. If I put kernels there, I'll ensure compatible drivers will be uploaded too [16:08] ailo: but that is if they choose to install the -rt kernel? [16:08] scott-work: ailo: for rt kernels, patches are needed [16:09] it's just a matter of creating a new package with the patch included, and tell dkms to use it if kernel is realtime [16:09] scott-work, No. It's not a matter if it is -generic or -lowlatency. It's a matter of kernel version. For -rt I assume the changes are done on the -rt kernel for the drivers to work [16:10] * falktx still wonder why rt guys changed the kernel headers... [16:12] ailo: disregarding -generic, -lowlatency, -rt - if you change kernel version numbers then you must rebuild video drivers [16:12] ailo: is that correct? [16:13] scott-work, Yes. But, falktx said something about making a driver work for all kernels. Don't know how that works, but I don't think Ubuntu are like that, since they will only work with the kernel version they were built against [16:14] scott-work, And as stated, since -generic and -lowlatency are the same, no need to patch anything to get anything working. [16:15] ailo: right, because UKT or somebody else will have already rebuilt them for the version, right? [16:15] ailo: something like: [16:15] sudo dkms build --all [16:15] But, 2.6.38 is not meant for Lucid or Maverick. I wonder what else, besides graphic drivers would be affected when using that kernel for those releases [16:16] scott-work, correct [16:17] Or, I should say, Lucid and Maverick were not meant to work with 2.6.38. [16:17] On Lucid, there would be a change regarding firewire at least, which the user needs to be aware of, I guess. [16:18] dammit, are FW users getting the fuzzy end of the lolly-pop again? [16:20] holstein, falktx proposed to add 2.6.38 -lowlatency for those releases in his PPA [16:20] only if you upgrade kernels in maverick and lucid apparently, holstein [16:20] and all the lollipops i've ever bought never had a "fuzzy end" , maybe you need a different supplier :P [16:20] scott-work: yeah? [16:21] ailo: holstein: I was using 2.6.38 on lucid and it worked fine (way better than 2.6.32) [16:21] * falktx is now on Natty [16:21] when your done with it... the stick is 'fuzzy' where the pop was [16:22] falktx, I haven't had much experience in using a different kernel for a release. I wonder about things like Virtualbox, things that have kernel modules. Will everything work? [16:23] as long as we make notes [16:23] virtualizing can *not* be a goal of ubuntustudio i think [16:23] hosting or guesting for that matter [16:23] ailo: official VirtualBox releases were working fine. I never tested the full-open variant though [16:26] holstein: ahhh, gotcha [16:27] i think when -lowlatency 2.6.39 is ready, there will be a substantial number of people testing it [16:28] being able to adjust irq prorities being on of the reasons [16:28] falktx, What's your KXStudio kernel PPA? Is it on your launchpad page? [16:28] and hopefully we can get a good comparitive feeling between -generic and -lowlatency [16:30] ailo: for now I use 2.6.32-preempt as default [16:31] ailo: I been so f* busy lately that I'm very late on kxstudio stuff [16:31] falktx, Ah, but you don't have a separate PPA for custom kernels? [16:31] ailo: yes [16:32] https://launchpad.net/~kxstudio-team/+archive/kernel [16:32] some notes: [16:32] rt-33 natty 64bit build failed to compile (broken GCC!!!!) [16:32] lowlatency-38 only builds on natty for now [16:33] ^ based on abogani's work [16:33] gcc got broken back in hardy [16:33] rt-31 does not work on maverick. still havent tested natty [16:33] and stayed that way [16:33] :/ [16:33] sux [16:47] falktx, I think you've done quite a lot already. I have been busy with other things, but I am eager to look through some things that KXStudio offers. I guess we are all busy with different things. [16:49] * ailo is going running. The goal is 4 minutes per km for longer distances :P [16:50] ailo: I just wish I had internet at home :( [16:50] what's with the 'minutes per kilometer' lately - what's wrong with good old km/h? [16:51] falktx, Isn't there good choices for mobile internet? I have still not tried on Linux· [16:51] ailo: I need a job... [16:51] I was kinda getting into one, but it went wrong [16:51] raboof, Easier to count. 4 min / km, means 1:20 for 20 km [16:52] lol [16:53] raboof, Or easier to measure, when running [17:14] anyway running is a good idea, i'm off too :) === ubott2 is now known as ubottu [17:24] eh, i keep meaning to respond to ronan's email and keep forgetting :( [20:30] ailo: when you get back, give me a ping please, i wanted to ask you about the other pulse stuff to remove and about documentation [20:31] scott-work, I'm here. [20:31] oh, okay :) [20:32] i agree with the padevchooser to be removed, what was the other item(s) you were saying should be removed as well? [20:33] scott-work, At first I thought pavucontrol should be removed, but there is one bit of functionality on it that may be worth while given that we include a volume control, which has almost exactly the same app as "Sound preferences" [20:34] ailo: but if we move towards including the indicator applet in the top panel then we could remove pavucontrol as well, right? [20:35] scott-work, I think that would be reasonable. pavucontrol has at least one bit of functionality extra, which is being able to choose audio card per application [20:36] But, I'm sure that can be done from the command line, or something. [20:36] Other than that, it seems identical. [20:37] scott-work, For the next release, I would like to start a more systematic testing approach where each and every application is tested and reviewed. [20:38] scott-work, Also, I want to test regurlarly [20:39] ailo: yeah, i almost typed up seomthing about "card per application" but stopped since i think all my experience with it was because i had three audio cards [20:39] ailo: we have discussed trying to get a small group together to codify testing but it never seems to gel and i can't do it on top of other things i do :P [20:40] I would at least test the kernel regularly. I want to be able to pick up on changes and try understanding what causes them [20:40] which is why i liked ronan's email, it seemed fairly organized and regulated...even just seeing if jack starts may seem trivial but is an important step :) [20:41] It's very easy to think that because something worked one week ago, there's no need to do that again [20:42] I'm also interested in pulse-jack. [20:42] i'm still pretty ignornat about many things and that is one of them, hence i haven't really pushed this but i think it could be an important functionality [20:42] i just don't have time to dive into everything :P [20:43] At the very least it could be added as an option. And, if something is not included in the main repo, I don't see the problem in letting a script import it from a PPA, like the css script which imports a package from medibuntu [20:43] ailo: but about documentation, how do you feel about what i've done so far with rlameiro? [20:43] https://wiki.ubuntu.com/UbuntuStudio/IntroTutVideos [20:43] i've been thinking that it could be broken down into four large groups: [20:44] 1. intro [20:44] 2. installing [20:44] 3. simple uses (broken into audio, graphics, video) [20:44] 4. advanced uses (parsed into a/g/v) [20:44] scott-work, Haven't read through everything. It seems more like an extensive approach. I was thinking, perhaps it is smarter to first compile every detail one can think of into a whole, then decide what is left in, and also make a quick-guide based on that [20:45] i'm still very keen about making some sort of nice, (at least semi-) profession looking introductary videos for people who don't know what ubuntu studio is [20:45] scott-work, I think those should be easily accessible from the main site [20:46] ailo: i'm not sure i understand your approach as described, can you explain a little more [20:46] ailo: yes, i would want at least one, perhaps two (they would need to be supplementary to each other) on the main website page [20:47] scott-work, Yeah, sorry. I mean, what you started now, seems like an all-inclusive manual, that covers everything. I think we should proceed with that. A manual that covers everything about Ubuntu Studio. Later, we break it down, and decide what to keep and where to keep it [20:47] maybe one on the main page, and a couple of others on a dedicated page to explainin gubuntu studio [20:48] ailo: ah, yes...that sounds good, but i was also thinking we could parse it out like the ubuntu packaging guide where it's parsed by "chapters" into different wiki pages [20:48] I guess it depends on what the video is about. If it's just an intro video to Ubuntu Studio, that would seem better placed on the main site [20:48] for example: https://help.ubuntu.com/6.10/ubuntu/packagingguide/C/index.html [20:49] right (re: intro video) that would be the plan [20:49] i would expect an appreciable percentage of people who visit the main page are ignorant about ubuntu studio (need to find a better way to describe them) [20:49] and that's where we need to grab them and hit them over the head with sexy, seductive information [20:50] "it's free! did we mention it's free!" [20:50] okay, since you don't hate what i've done i'll keep pushing the documentation then :P [20:50] scott-work, Absolutely. I think we should just add to it as much as possible, and sort it out later [20:51] Later, I would want to have a "quick-guide" that just in as few words as possible lets a new user start using Ubuntu Studio [20:52] ailo: i think that is a great idea as well [20:53] And, I believe -controls will be a big help when it comes to tuning the system. I will include a sound-check for it. [20:53] this would be good for people who, like me, does something and susses it out, say midi, and then tries to do it again several months later, like i did last weekend [20:53] but can't remember _exactly_ how it happened before [20:53] scott-work, Right. It's not fun having to dive deep into the details, when you just want to try it out [20:56] ailo: just for clarifcation... [20:56] i'm thinking the parts that are already in the wiki (the stuff i basically did) requires some lengthy or perhaps wordy explanations [20:56] that is, few pictures or bullet point lists and many words [20:57] but the parts where we show how to accomplish say, adding a synth track in qtractor is more line-by-line steps and pictures with fewer descriptive words [20:57] scott-work, That's the approach I have had too. List the topics first, and then start writing them. I'll need some time to get my English right as well. [20:57] eh, your english is quite good [20:58] you and alessio and ricardo read and write it better than many of my own countrymen [20:58] * scott-work cries a little inside [20:58] i think slang and internet abbreviations are killing literature and writing [20:59] and i really don't think americans tend to value education like other countries [20:59] but that's besides the point [21:00] what i'm getting at, ailo, where does you "quick guide" fall into that dichotomy? length explanation describing philosophies, followed by succinct step-by-step guides with pictures? [21:00] Well, I'm not so sure. I think knowledge of language will be more tuned towards content, than anything else. Computers will do so much work for us. [21:01] agreed, but i find it amazing (and have great respect for) how you, ricardo, and alessio can converse fluently in multiple languages [21:01] it's not something your white, anglo-saxon, protestants do well, mainly due to pride or egocentricity [21:01] The quick-guide should basically be: One, two, three, four - rock'n'roll. Only information that is needed in order to install, start an application and get sound out. [21:02] ailo: okay, so the quick guide would be almost a separate documentation, just extracting the bare minimum to get the job done? [21:02] I think Alessio should have some credit. Italians don't usually speak English very well [21:03] scott-work, Exactly. I think the main documentation could be more like a reference, where everything you need is collected, even links to other sites [21:04] I grew up with two languages, and in Sweden we get a lot of English from our medias. [21:04] So, I speak three languages + a little German [21:05] The quick-guide would be much easier, if there was a live DVD [21:06] scott-work, This way, the user could get some experience very fast. [21:07] ailo: ooohhhh, that's a good point [21:07] that's a great way to get people interested [21:07] i wonder if we could even include the quick guide on the disc [21:07] i bet we could [21:07] think of it.... [21:08] person never used ubuntustudio before can boot up live dvd, then follow along with quick-guide tutorial and make mad sounds [21:08] without ever touching their harddrive....i would think that would really encourage people [21:09] Puredyne works pretty well, when that is concerned. Though, recording to disk is another story. [21:10] It wouldn't be bad to include some sort of demo. A script that starts up multiple apps and plays a song [21:10] Showcasing a few applications at the same time [21:13] * ailo needs an ergonomic chair... [21:13] that would be cool :) maybe if we get ladish into the repos then that wouldn't be that hard to do as well [21:14] scott-work, That's another thing I don't know anything about. [21:15] i will probably still need you and rlameiro to help identify (probably as a group) what documentation/tutorials should go under 3. simple uses and 4. advanced uses [21:15] scott-work, And, another thing I want to improve for next release, the knowledge of all the applications available [21:15] my thought is that if we can frame it in and give people a format to easily follow then we might get some help to create the documetnation [21:15] scott-work, I think holstein would be a good asset there as well [21:15] i think that people generally want to help but dont' want to also have to invest time figuring out *how* to help ;) [21:16] ailo: ooohhh, yeah, i'll brow-beat him later about this :D [21:16] hehe [21:16] yeah, im in [21:16] "knowledge of all applications" i choose this year in february to learn some midi during the rpm challenge for this very reason [21:17] holstein: sweeet :) [21:17] i'm even thinking of doing some quick wiki pages showing people how to do wiki pages! [21:17] like how to take a screen shot, crop it, and attach it to the wiki [21:17] how to special format some text that would be used often [21:18] give people the tools and the defitintive goal so that it only requires their time to accomplish stuff [21:25] although i think the 3. simple uses might need a preamble, like "getting sound out of linxu", helping newbies understand the linux audio landscape, specifically pulse and jack [21:31] scott-work, If we have a system that is very easily tunable. If we have -controls doing quite a lot already, we'll need troubleshooting for when something fails. [21:32] I'm also thinking about adding a patch to make delta cards work with pulseaudio, which could be added with the users permission from the -controls [21:33] M-Audio Delta, I mean [21:34] Trouble shooting could be a few questions, that then give links as answers on different topics [21:34] ailo: how would they work differently than they do now? [21:34] scott-work, You mean m-audio delta cards? [21:34] aye [21:35] i use a delta 44 and i think i can get sound out of it through pulse [21:35] * scott-work admits that he uses onboard for the majority of non-audio recording audio [21:35] scott-work, I haven't tried in a while, but from what I've read it's not been possible [21:36] You need an edit to a alsa config file. ice1712.conf I think it's called [21:36] ailo: what would be a use case for this (i ask because i can think of a reason not to do this) [21:36] A simple addition to that file makes the card usable with Pulseaudio [21:37] New users have a tough time getting their ice1712 cards to work with pulseaudio. [21:38] why would they want to use pulse with their ice1712 cards? why not use onboard audio for those activities? [21:38] sorry if this sounds snotty, or egalitarian [21:39] scott-work, Not everyone has onboard audio, and usually people would prefer to use the same card for all tasks, especially if it is a home studio setup, where the card is connected to some nice monitors [21:39] okay, i can see that [21:39] Just the fact, that you can't get sound out, makes people think they aren't supported. [21:40] but then some functionality is lost currently, meaning i can route audio through pluse and onboard and i can also route other audio through jack via delta44 [21:40] not saying this is a deal break, just pointing it out [21:40] I don't expect that any functionality would be lost [21:40] many people may not care, and jack-bridge may be the answer to this "loss "of functionality [21:41] I would rather the card works just as any other would [21:42] The only reason why ice1712 doesn't work with pulseaudio is because the alsa config file is not presenting a typical stereo output that PA can read [21:42] Or something like that [21:42] It's a very easy fix, but is also a hack. [21:43] But, it doesn't affect anything. The only thing the fix does is allows the card to be used with PA [21:44] ailo: my point is that even if you fix the alsa config so that delta cards work with pulse, then a user would not be able to concurrently route audio through pulse and jack at the same time [21:44] i currently can because i use different cards, one for pluse and one for jack [21:44] That will not change [21:44] It's up to the user [21:45] If the user wants, he can use one card for jack and the other for PA [21:45] i don't think we can concurrently, i think dbus with either let the card use pulse or jack, but not both [21:45] right "he can use one card for jack and the other for pa" [21:46] Or, use the same card for both, but not at the same time [21:46] exactly :) [21:46] i don't want to be misunderstood, i think this would be nice to be fixed, anything that makes it "just work" is probably good [21:46] i was just pointing out a very niche use case where this would cause a problem [21:47] of course, jack-bridge might mitigate this "problem" as well [21:47] scott-work, I don't understand. I don't see what is changed with fixing the alsa.conf [21:47] oh, the change itself isn't necessarily the problem [21:48] The card will still be usable with jack at the same time as one uses the onboard card with PA [21:48] it's user's expectations..."i can route everything through pulse now including my audio recording stuff...oh, why can't i route youtube and my recording together?" [21:49] right, but your use case was that a user wanted to use both pulse and jack through the same card with the nice speakers [21:49] Yeah, but that would only be possible with jack-pulse bridge [21:49] yeah, but it's probably a very small use case that most people aren't going to worry about and i probably shouldn't have brought it up ;) [21:50] scott-work, But that isn't possible even before the fix, without jack-PA bridge [21:50] i wonder how we would be able to "fix" this permantly [21:50] ailo: with two cards it is [21:50] You mean, you can route the jack card to PA? [21:51] Or, the other way around? [21:52] As things are now, no matter what, without jack-PA bridge, I don't think you can use PA and jack with the same card at the same time [21:52] http://dullass.blogspot.com/2010/06/wisely-and-slow-they-stumble-that-run.html [21:52] i wanted to play a youtube video while playing my guitar [21:52] i routed the youtube video through pulse and my onboard audio into computer speakers [21:52] i routed my guitar through jack and my delta 44 into studio monitors [21:53] it was when they moved to (jack2 i think, and) dbus that allowed this to happen in maverick [21:53] scott-work, This would not change [21:54] Allthough, wait.. [21:54] Now that jackd is suspending PA, I don't think you can do this anymore on Natty anyway [21:55] but i found i couldn't route both through the same card though [21:55] On Lucid qjackctl suspends jack, so there you would use jackd from a terminal to achieve this [21:55] the pasuspend is in qjacktl and that's how i started jack in maverick so i don't really understand how it happened either [21:57] scott-work, No, I'm lying. [21:57] my delta card is in the computer with lucid on it now, so i can't test this again currently [21:57] scott-work, On Natty, as well as on Lucid, qjackctl suspends jack [21:57] ailo: i dont' think you are the lying type [21:58] But, I believe on Maverick it doesn't [21:58] suspends PA, I mean [21:58] oh, i gotcha [21:59] So, on Natty, if I start my second card using jack, while PA is using my onboard card, it will work, as long as I start jackd from the terminal [21:59] that's wierd that it wouldn't do this on maverick, though [21:59] It will work on Maveric too, even if you use qjackctl ( I think ) [21:59] Because there qjackctl is not stopping PA [22:00] Anyway, the fix will not change this. I can use any cards like this [22:01] The only thing that you can't do is use jack and PA on the same card at the same time, without PA-jack bridge, which makes PA a jack client I believe, and therefore PA is not using any card, only giving jack the audio streams [22:03] to reiterate: i think it will be a good fix and anything we can make "work out of the box" is probably a good thing [22:03] but i wonder how we would be able to make this fix permanent? [22:03] maybe we should file a bug about alsa? [22:04] scott-work, As I understand it, it's like a design issue for alsa [22:04] There are multiple bug reports on it, and they are more than 3 years old now [22:04] but i would imagine there are ubuntu-specific config changes that are already being done, this would just be another one [22:04] oh [22:04] maybe we should push some people about this then [22:05] Everyone blamed PA first, but in a way PA is forcing a change in the way the alsa.conf is written. Or, it is not done right. I cannot tell, since I don't know that much more about it [22:06] Not alsa.conf, ice1712.conf.. [22:08] I'm not sure, but there may be other multi-channel cards affected. At least, I would like to add a fix for ice1712 [22:12] oh, yeah, i've heard david t. chen and others say the same thing about pulse exposes alsa's warts [22:12] Daniel, not David. [22:12] ack, yes, my mistake, thank you astraljava :) [22:13] :) [22:26] right, i'm going home :) [23:24] hi holstein [23:26] ronj: o. [23:26] o/ [23:26] ScottL mentioned https://lists.ubuntu.com/archives/ubuntu-studio-devel/2011-April/003021.html that you did some fiddling with usb installs of studio. I'm a tad surprised to hear that installing from usb is not reliable and not really supported, because that's what I've been doing for years. What were your findings? [23:28] ronj: let me get the bug report i filed [23:28] yup [23:29] https://bugs.launchpad.net/ubuntustudio/+bug/697774 [23:29] Ubuntu bug 697774 in Ubuntu Studio "cant install ubuntustudio from USB stick" [Undecided,New] [23:29] this is mentioned after my ranting there [23:29] http://die-klapsmuehle.org/2010/02/21/ubuntu-alternate-von-usb-stick-installieren/ [23:29] i got my information from the ubuntuserver wiki [23:30] ok thanks, so that means the official usb-creator-gtk is not supported [23:30] AFAIK the alternate style installers dont work [23:30] with that or unetbootin [23:30] they both fail at finding the install media for me [23:31] ronj: we are talking about the installer [23:31] and where we want to go [23:31] IF we go to a live CD [23:31] then the USB install will be easier [23:32] as is stands [23:33] to be precise: I tried to dump a recent version of studio / natty to a usb key through Ubuntu's usb-creator-gtk, in order to do some testing. usually it works like a charm but there it did not [23:33] i personally put the responsibility on the tool [23:33] the USB creator [23:33] i did not try unetbootin [23:33] its what is not supporting alternate installers [23:33] unetbootin is not an official ubuntu tool, so that could be ignored [23:33] "its what is not supporting alternate installers" >> well, it used to work for me till now..... [23:33] For me unetbootin worked as far as creating the installer, but after installing, not everything had been installed [23:33] ronj: yeah? [23:33] yup [23:34] never worked for me [23:34] no ubuntustudio [23:34] no server [23:34] no alt [23:34] always failed [23:34] even did studio natty alpha2 like that [23:34] weird [23:34] 32bit version [23:34] ok [23:34] yes [23:34] since 9.10 [23:34] yes [23:34] ronj: ive been loud about it [23:34] ? [23:34] on several occasions ;) [23:34] oh [23:35] I'm not always listening as much as I should :P [23:35] ronj: well, not at you [23:35] just in here really [23:35] ok... I'll give it another try and may try to add some meat to your bug or possibly create a new one on usb-creator-gtk [23:36] thanks holstein ailo [23:36] ailo, can you precise "after installing, not everything had been installed"? [23:36] ronj: please do :) [23:37] ronj, After installation, I booted into the system, but it just stopped at one point, and the screen went to sleep. Ctl + Alt + Del rebooted. [23:37] ronj, Then, I used recovery mode, which was no problem [23:38] ronj, And there, I realized, ubuntustudio-desktop had not been installed, so I installed it [23:38] ronj, After that, everything was fine [23:38] w0w [23:38] thats *wat* better than i ever got [23:38] ok thanks [23:38] way* [23:38] But, I remember from before, I could not even create the installer [23:39] ailo, could not create the key? [23:39] you mean [23:39] ronj, Right. [23:39] it would be fine if in the tool documentation it said 'doenst support alternate CDs' [23:39] which tool? what did it say? [23:40] OR, if when you had an alernate iso [23:40] it would fuss about it [23:40] holstein, right, that's what i was suggesting: if implementing support for alt cds into usbcreator is too much work, at least warn the user [23:41] yeah, thats where i think it should be [23:41] i have other reasons for wanting US to have a live CD [23:41] and that would just be an added bonus [23:41] wonder who makes that tool? [23:41] that should be easy enough to add [23:42] A live DVD / CD seems to me like the only way to go, really [23:42] ronj: i have a couple days off here [23:42] maybe i'll try and grind that axe a bit more :) [23:42] welll sorry for interrupting... [23:43] rlameiro: o/ [23:43] but I was thinking the same problem [23:43] how to deploy an ubuntustudio image.... [23:43] for a lot of computers and not Linux versed [23:43] (Universities / Academic) [23:44] rlameiro: like a net install scenario? [23:44] LiveDVD would be great [23:44] holstein: that would be god on a second base [23:44] but you can do it already wit the alternate install [23:44] true [23:44] The only downside with a DVD is that it doesn't cover every scenario, why I think it would be really nice to have a CD as well [23:44] I will give the example [23:45] * holstein BBL [23:45] Composing teacher: Today we will lear how to create an orchestra in supercollider, or a synth. [23:45] please put the DVD on the computer and restart [23:46] go to applications and open this X Appp [23:46] this could be great for academics to strat using linux.... at least on this area [23:46] rlameiro, It would even better, if one could install onto the live-usb. Like you can with Puredyne [23:47] I mean, install programs, change settings and so on [23:47] ailo: That would be AWESOME [23:47] Puredyne already does this, so we could see how they achieve that [23:47] even better if the image has all the MAC drivers [23:48] ailo: I think they create a install and then make it a live cd/usb [23:49] rlameiro, They have a script, that installs the iso to the usb stick, and when you boot into it, you can add new users, install programs. There is two partitions. One for the image, and the other for saving data [23:50] hummm [23:50] seems easy implementation [23:50] But, the interesting part is how the iso is created [23:50] So, it's in two steps [23:50] ailo: what do you think of the script untill now? https://wiki.ubuntu.com/UbuntuStudio/IntroTutVideos#Objectives: [23:51] do you know rootstock? [23:51] rlameiro: rootstock? [23:51] ailo: https://launchpad.net/project-rootstock [23:51] its for arm [23:52] you create an root filesystem [23:52] you can add users and seeds at build time [23:52] I am sure there is a similar tool for normal computers [23:54] Puredyne uses debian-live, but I don't know how that works. Haven't investigated further. [23:54] hi rlameiro and ronj, just go in, will read backscroll in a minute or so [23:54] One thing that is missing from Ubuntu Studio is giving the user realtime prio right from the start [23:55] I think.. [23:55] https://help.ubuntu.com/community/LiveCD#Making%20a%20Customised%20LiveCD [23:55] ScottL: hi there :D [23:55] ailo: that would be managed if jack is installed or not? [23:56] rlameiro, Yes, but user needs to be in audio group too. Don't think this happens automatically. We should add that for creating new users, if possible [23:57] Probably easier to do on a strictly audio based distro [23:57] ubuntustudio-settings? [23:58] we could add a startup script to do $dpkg-reconfigure jackd [23:58] Anyway, the live user must have realtime prio, especially if one cannot save settings, otherwise it's not really fun using the liveDVD [23:59] rlameiro, I think it must be some kind of default thing, when you add a new user, the user becomes a member of certain groups