[01:08] hi falktx , just reminding you about the ubiquity patch again [01:09] it looks like cjwatson is moving forward with setting ubuntu studio up as a live-dvd now [01:09] seems like everyone needs my help... [01:09] who else? [01:10] users and some devs [01:10] ScottL: what exactly is needed? [01:10] falktx, edubuntu has a patch for ubiquity that allows basically a GUI tasksel during installation [01:11] we would like to modify it to work with ubuntu studio [01:11] going upstairs for a bit, should log on there as well [01:11] ScottL: can we postpone this after friday? I need to finish something on 20th [01:38] falktx, absolutely :) [01:38] thanks [03:08] ScottL: I can take care of the ubiquity part of the live-dvd. Sorry for being a bit vague about it. I meant to include that as part of my live-dvd tasks. [03:08] falk doesn't seem to be online now, but I'll mention this to him when he is. [04:04] astraljava, oh yeah, i'm not worried about who does it as long as it's getting done :) [04:04] by the way, i dput'd the lowlatency kernel to REVU, but i haven't gotten an email and can't see it listed yet :/ [04:05] ScottL: Yep. Just thought I'd mention it, as he seemed to indicate being rather busy at the moment. [04:05] ScottL: It might take a while. It's a slow service. [04:06] and i made all bzr pushes and filed a bug for theme-ui changes [04:06] micahg, ^^^ [04:06] and i subscribed sponsors as well [04:07] damn, i'm tired and not even seeing straight now [04:07] Heheh. Go get some well-deserved sleep. :) [04:07] i'm going to bed now, talk to you tomorrow [04:07] Good job! [10:49] ScottL: will upload when I have a chance (not sure if it'll happen before the weekend) [14:30] scott-work: Did you keep Audacity in the seeds btw? [14:33] ailo: i haven't really reworked the audio/video/graphic seeds at this point [14:33] ailo: but i shall! (both rework them and add audacity) [14:33] right now i just wanted to get the push for the theme-ui changes [14:33] and get the live-dvd pushed to cjwatson [14:34] i talked to cjwatson and he said that adding additional seeds for new work flows will be easy enough later on [14:34] so i wanted to get the theme-ui changes so we can get feedback from A2 [14:34] and the live dvd working for A2 as well [14:35] ailo: do you feel up to doing some development work on a very specific item? you'll be famous! [14:35] scott-work: I don't have a lot of time on my hands, but if it's not too time demanding.. What do you need? [14:37] ailo: we still need somebody to really focus on getting gcdmaster ported to gtk3 or qt or something else [14:37] well, i say that because i think astraljava is going to work on the ubiquity patch currently [14:37] About Audacity, I read that it is the 11th most downloaded software at SourceForge, and is perhaps only surpassed by VLC as the most popular FOSS software in the world. [14:37] I mean, audio related [14:38] But, the 11th most downloaded out of all the software projects at sourceforge [14:38] ailo: stochastic and i talked quite a bit weeks ago and i need to make some fairly significant changes in the audio seeds, one of them will defintely be including audacity, i promise [14:39] * scott-work also needs to give significant thought to the video and graphic seeds and make sweeping changes in those as well [14:43] astraljava: sorry, my comment was unclear eariler [14:43] astraljava: are you also still doing the gcdmaster port as well as the ubiquity patch? [14:44] astraljava: it seemed that you were going to focus on the ubiquity patch immediately, if so i would like to see if someone else can get gcdmaster during the same time [14:44] astraljava: and if you are handling the ubiquity (which is very hard to type quickly for me) patch, then i will let falktx know that he doesn't need to do it [14:52] scott-work: Isn't it better just to focus on another cd mastering app? This one hasn't been updated since 2005 or something [14:52] scott-work: If someone wants to do it, it's fine by me, but since there just hasn't been that many volunteers, it's on my list thus far. [14:53] I'm not really qualified to do c++ coding, if that is what it takes to port it. Or, it would take me too much time. [14:59] astraljava: i just worry about workign too many things in a linear fashion, i fear we might run out of time and if we have the capacity then it would be nice to have multiple people focusing on individual items right now [15:00] scott-work: Yes of course, but like I said, there just hasn't been that many volunteers. So I'll keep it on my TODO as long as someone else takes it, or I finish it. :) [15:00] astraljava: i don't know, it might only take you one or two days to suss out the ubiquity patch and then you are on to the next item [15:01] astraljava: we're you also looking at packaging slomovideo? i saw a bug report last night about that, but it was late and i was looking for something else anyways and didn't look too close [15:02] scott-work: The fastest way to get that wrapped up is talk to quadrispro. He'll have it ready in 5 min... scratch that, three minutes. :) [15:06] lol, that is true [15:06] i need to research the mailing list about other packages that shnatsel asked about so i can file the RFP for all of them at once [15:07] astraljava: to be forthcoming, this whole situation (well, not the _whole_ thing but a lot of what effects us now) with gcdmaster and such is my fault [15:07] simply because i haven't stayed on top of it so it got done in a timely fashion for this cycle [15:08] not we're doing the same "hurry up, we're 1/2 way through the cycle" dance :( [15:08] same thing for the new packages [15:08] Nah, it's a team effort. And I really don't recall we ever really setting porting the gcdmaster as a release goal. [15:08] i doubt we will get the new packages into ubuntu in time to include them in 12.04 (although we can backport and include in 12.04.1) [15:08] and i have serious worries about gcdmaster now for the same reason [15:08] Why is gcdmaster so important? [15:09] I haven't followed all your discussions, so forgive me for that [15:09] ailo: It's supposedly the only capable tool of creating professional-level cds in FOSS. [15:10] Wow [15:10] ailo: scott-work: I just checked-out cdrdao project, and turned it into a bzr tree. I'll look around a bit, and register a branch on LP later tonight. [15:15] astraljava: cool [15:16] ailo: my understanding is that it allows extreme manipulation of the TOC, etc for mastering CDs, including adding 2 sec spaces, etc [15:20] scott-work, astraljava: So, what does it have that k3b doesn't? k3b at least supports adding silence between tracks. Seems like a very alive and well designed app to me [15:22] At a first glance it seems to me that 3kb has everything gcdmaster does, as well as a good deal more [15:22] It uses the same backend right? [15:22] ..cdrdao [15:24] Well, I dunno, I've never even used it before. Send an email to the mailing list, if you want to wrestle some arms with the guys in the know. :D [15:25] scott-work, astraljava: I think just by using them side by side on an older Ubuntu system should reveal some facts [15:25] I have Lucid, so I can do that [15:25] 3kb looks solid to me though [15:26] k3b* [15:32] https://code.launchpad.net/~ubuntustudio-dev/+junk/gcdmaster-gtk3-port [15:34] btw, how is US handling gtk2/3 themes now? [15:35] do we use some engine that supports both toolkits? [15:35] falktx: We use whatever Xubuntu uses. If you wanna do the tech talk, I suggest heading off to #xubuntu-devel. [15:37] I'm just curious [15:38] there are only a few themes that handle gtk2/gtk3 at the same time [15:38] afaik, xubuntu using xfce will only care about gtk2 [15:39] falktx: I'm fairly sure that's a no. There's been talk of utilizing gtk3 themes. But like I said, I haven't participated in that area. [15:40] I'll ask them then [16:04] falktx: ochosi had been making updates to their theme to handle gtk3 i believe [16:07] yeah, cool [16:46] ailo: re gcdmaster/k3b. Adding 2sec gaps is easy actually. The harder part is taking a live performance with no breaks and adding track marks, or shortening the between song time while keeping that continuous always audio live album feel. [16:48] afaik, audacity can be used to split 1 audio into several tracks [16:49] All this can be non-destructively with gcdmaster. I am not at all sure that k3b can start track1 with a pregap long enough to hold a whole song... if at all. Can it add indexes? CD-TEXT? Can it add ISRC? [16:50] Can k3b take those tracks split by Audacity and run them together seamlessly with no break between them? [16:50] k3b does support cd-text and pregaps [16:50] and no-gaps between tracks obviously [16:51] oh, and overburn too, which I use quite often [16:51] I don't know what ISRC is though [16:52] It is mandatory for air play. === scott-work is now known as scott-work-afk [16:58] It is hard to describe all the things that cdmaster does. Using audacity to create a bunch more files when cdmaster works with the master file seems like extra effort. An extra unnecessary bunch of work. I will take another look at k3b... but last time I looked it was one of those burner that takes a bunch of tracks and stuffs them on a CD. CDMaster allows joining two audio files together while visually seeing that the samples are at zero crossing f [16:58] or no click without actually editing the sound file itself. [17:00] It allows one to hear the whole CD as it will be when burned. [17:03] A professional CD is the one that takes care of all the tiny things that most people don't think of. GCDMaster gives a master copy that could be printed for distribution. If US is trying to be something that can take sound in, mix it master the tracks... do the whole thing, in other words why stop at the last step? [17:05] scott-work-afk: You sure it's _micro_seconds? That's a little... absurd. Who cares for such precision? No one's going to notice. [17:06] For doing straight from studio to cd with a gap between each track, probably one could get away without gcdmaster (or something similar), but for live recordings thats different. === scott-work-afk is now known as scott-work [17:18] good catch astraljava ! [17:18] :) [17:25] ailo: ping [17:25] you around later? [17:25] i would like to take a few hours (not all at once) with you [17:25] and test some kernels again [17:25] i just want to make sure im not forgetting anything [17:28] falktx: Just took another look at k3b. It is by far one of the better toc editors I have seen... [17:29] len_: i forget how it failed for me [17:29] have you actually tried it?... it seemed like it should and failed [17:29] this was a year ago i bet [17:30] but, it is not as good as GCDMaster. K3b does allow editing the toc in all the right ways just as if you were editing the file by hand but easier. [17:33] well at least it sounds like we have a path to edit the TOC even if k3b does pose a larger installation footprint [17:33] but if k3b is used to master cds _and_ as an image burner, then this is mitigated slightly [17:33] However, it does not let you hear the transition from song to song without burning a copy, or creating a pre-edited cdlength file before hand with another app. It does not let you visually see the file transition without using another app to pre-edit a cdlength file first. It would be possible to use k3b, but harder and require more disk space (minor) and time (not so minor) and it would still take burning it to hear if your track starts were accura [17:33] te. [17:33] oh :( [17:34] len_: Since you seem to be comfortable using cd mastering tools, I suggest you take a closer look at k3b. It can split tracks, which in my view is not something you actually require from a cd mastering tool. A cd mastering tool is to burn cd's, not edit audio files [17:36] I have used gcdmaster in the past... and just went down and tried taking a few wave and editing a toc with k3b. I'm not sure this is quite as good as it could be. [17:36] From my point of view, a cd mastering tool should be chosen which is able to burn cd's professionally. And that is all. Additional features that involve audio editing does not seem important [17:37] holstein: How come you want to test kernels? [17:37] it was presented to me as a direct replacemnt, and i didnt find it that [17:37] ailo: i would like to document, and get something scott-work can link to [17:37] i just want to make sure i dont miss anything [17:37] ailo: and i want to do it for 12.04, with the latest *-generic in there [17:38] We are not actually talking about editing audio files so much as accurately specing track starts and file transitions. The only way to do that is by hearing what your change sounds like. k3b does not offer that feature [17:39] len_: If you make a master, you do the cutting in a DAW or whatever you use for mastering the music [17:39] I understand it is a nice feature to have, but not essential [17:40] holstein: What sort of tests were you thinking about? The ones we were doing before Oneiric? [17:40] ailo: yup [17:40] i check at the beginning of oneiric too [17:41] holstein: The results will depend on your hardware, and can only serve as examples I guess [17:41] ailo: i was going to try FW and USB and internal [17:41] holstein: The best is if we can have multiple testers using different kinds of hardware. I have FW, internal and PCI [17:41] i just want to help make or break the case [17:42] i have no PCI though [17:42] The case is already pretty clear, I think [17:42] ailo: yeah, you would think [17:42] But you want to have some numbers to show to people, right? [17:42] but we are *not* going to get that kernel im afraid [17:42] and i have some time today [17:42] and i want "clear* #'s and pasted lines from JACK's window [17:43] len_: What do you think? Is k3b not good enough for the job? [17:44] ailo: Once you have used both for setting up a live concert to cd, not having a good toc editing tool becomes obvious real quick. [17:44] I would like something that will let me hear what I am about to burn. [17:45] I've mastered a few records, but never paid much attention to the burning, since I've just edited the music before burning it [17:46] i used 64studio, and GCD master was working there [17:46] so far, i have avoided the process.. just making mastered tracks [17:46] im going to buy nero for the blueray support [17:46] The problem with gcd is that it hasn't been updated since 2005, and shouldn't this be maintained in the Debian repo instead of here? [17:46] holstein: k3b will get blueray soon [17:47] i need the CD thing [17:47] not sure how blueray works... [17:47] K3b is better than any of the other ones I have looked at. It has a lot of good stuff. I would certainly use it over anything else if there was no gcdmaster. [17:47] BR data can be made from the CLI, and i dont mind supporting nero [17:47] i only want it for data [17:47] can we even use k3b? [17:47] what are the deps? [17:47] kdelibs [17:47] is it heavy on KDE? [17:48] we just set some non-wanted kde packages to not install on the US dvd [17:48] like done before, I think [17:48] falktx: so, we an have it without a bunch of extras? [17:48] if k3b gets into the seeds, kdenlive would be a nice option too [17:48] im fine with includeing it [17:49] holstein: yes, it's perfectly possible [17:49] and i agree about kden [17:49] scott-work: can we just go with this? [17:49] then we get kden too [17:49] GCD master is awesome, but it is dying [17:49] as long as i can burn a TOC projet with k3b, then i say lets do it [17:50] I think because of the work to get gcdmaster back up to scratch, it isn't going to make it in time. K3b is certainly next best that I can find. [17:51] Would seem likely that k3b will continue to get more features in time. I never considered the dependencies. Do we have any other KDE dependencies? [17:52] There needs to be a new project to replace gcdmaster with something more up to date. but that will not happen in time for 12.04. [17:53] Yes, and in the mean time, I will work on porting it to GTK3. [17:53] astraljava: k3b? [17:53] ailo: maybe 20-30Mb of dependencies [17:53] holstein: Obviously not. :D https://code.launchpad.net/~ubuntustudio-dev/+junk/gcdmaster-gtk3-port [17:53] Point anyone interested that way. ^^ [17:53] astraljava: hey... i dont know what you guys mean half the time ;) [17:54] falktx: Not more? That's not too bad. [17:55] basero can come out too, though from what Csott [17:55] ailo: yep, but I think it needs tweaks to make sure we don't get unneeded kde stuff [17:55] ailo: True, but consider them all in RAM, alone for that app, if nothing else needs them. [17:55] sure... k3b can just bo our burner [17:55] its nice [17:56] Scott said brasero doesn't add anything [17:57] astraljava: I don't know that much about live CD's but to my understanding, applications themselves take the same amount of space in RAM, when they are used, and no space at all, when they are not used. [17:58] I suspect what is additional in RAM for live CD's is the filesystem [17:59] ailo: Applications use the RAM they themselves need, plus the dynamically linked libraries, when used. Libraries will get swapped at some point, when the application that needed them is shut. [18:00] astraljava: I can't believe there would be any problem anyhow. We're talking about fairly small amounts, and nowadays, 2-3GB is not enough to run a web browser [18:01] I understand that. But in the worst case scenario, if the machine ended up swapping because of that... [18:02] But obviously we don't know at this time whether the porting brings in additional GTK3 libraries. [18:02] I'm just saying we need to keep an open mind, and try to avoid adding stuff unnecessarily. [18:04] astraljava: gtk3 libs should be on xubuntu/US by default, since base ubuntu stuff uses it [18:04] falktx: I mean are all of them loaded into memory already? We're not using a lot of GNOME apps. [18:06] libs are only loaded once an app that uses it is started [18:06] if we include kde stuff, kde libs will only be loaded once we start a kde app [18:06] But of course. [18:06] The point was, are there KDE apps? [18:06] no, I don't think so [18:08] I believe the kde libs take less than 5 MB in ram [18:08] Not counting k3b itself [18:08] But falktx just mentioned it wants 20-30MB? [18:09] That's the size of the packages, right? [18:09] on cd [18:09] Oh ok. I thought we were talking about RAM. [18:10] But then, if those are libraries, then the real size is somewhat bigger. [18:10] Close to that, but slightly bigger. [18:10] .debs are gzipped, right? [18:10] most of them [18:11] http://people.canonical.com/~ubuntu-archive/cd-build-logs/ubuntustudio/precise/daily-20120118.log [18:11] new ones use lzma [18:11] that shows some kde packages but not much and i'm not actually sure they were really installed [18:11] So, kde-runtime is already in [18:11] ? [18:12] ailo: i didn't see it actually installed though [18:12] ailo: furthermore, there are things in there that probably shouldn't be, e.g. libreoffice-* [18:13] ailo: by "[12:12] ailo: i didn't see it actually installed though" i meant that i didn't see a package actually included on the disc (i.e. later down the log) [18:13] but also, see: https://wiki.ubuntu.com/UbuntuStudio/Sandbox#package_size_testing [18:13] that documents k3b and kdenlive package requirements for an installation on an xubuntu 12.04 daily instal [18:13] isntall [18:15] Would be good to find out how big those packages are. Most of them tiny, I would think [18:16] ailo: which packages? [18:16] scott-work: The dependencies for k3b, and furthermore kdenlive [18:17] Both share a few dependencies [18:17] That said, k3b shares a few dependencies with gcdmaster as well [18:17] ailo: to add to this: https://wiki.ubuntu.com/UbuntuStudio/Sandbox#package_size_testing [18:18] ailo: i was going to install k3b and then see what would be further required for kdenlive to find the delta [18:18] ailo: which should be the shared dependency size [18:18] i was hoping to do similar with a kde install against a few things as well [18:18] scott-work: Yeah, duh. That was pretty elementary [18:19] if we get kde stuff into the dvd, we'll need to add some default-settings to make kde follow the gtk and icon theme [18:20] falktx: if we add kde stuff, and i'm not saying that i'm onboard with this yet, can you help us make those changes? [18:20] scott-work: what would you need? [18:21] falktx: what ever changes you mentioned about making "dke follow the gtk and icon theme" [18:21] ah, sure, it's just a file [18:21] ~/.kde/config/kdeglobals I think [18:22] we just theme to Gtk2, and icons to [18:22] we first need to be sure if kde stuff gets in or not though [18:23] falktx: right, it is still a decision that needs to be considered and evaluated [18:23] k3b, 75.2 MB of archives. Then the questions is how much additional space kdenlive would need. I would assume both together are at least 100 MB [18:23] astraljava: is cjwatson wanting you or us to create the live and live-dvd seeds in our bzr branch? [18:24] Are you making a live CD, or live DVD? [18:24] ailo: check the installed size, not the downloaded size [18:24] ok [18:24] ailo: it would be a live dvd, i really doubt we would be able to get down to cd size [18:24] ailo: half of that is not needed [18:25] --no-install-recommends helps there, not sure how to do it in seeds [18:25] falktx: but can we exluce the half that is not needed wiht the seeds [18:25] oh, nevermind ;) [18:25] note - k3b recommends 'dvd+rw-tools', which should become a US depends [18:26] we can also add 'libk3b6-extracodecs' for mp3 support [18:26] falktx: i thought brasro did the same with dvd+rw-tools? [18:26] brasero misses functinality [18:27] nautilus as well (pulling dvd+rw-tools) [18:27] it may use the same base app for recording, but it doesn't have the options [18:27] falktx: sorry, i misunderstood what you were getting at...i thought you were complainging about k3b pulling that in, i was just saying so does brasero and nautilus, therefore we would have it anyways [18:28] ok [18:45] totem can be such a buggy piece of crap sometimes :P [18:46] holstein: So, you want to write an informational page about the difference between -generic and -lowlatency for users, or what is your plan? [18:51] ailo: no.. just for us [18:51] for me and for scott-work to use if he wants [18:51] if a kernel dev says "why do you need it?" i want him to be able to say, see this... [18:52] holstein: It's just too bad our test is not the most professional, even though I think it counts. Especially if we have enough testers [18:52] ailo: well, we can link it where ever [18:53] maybe i start a new blog somewhere, just for this purpose [18:53] one that other can add to or can be copied to where ever [18:53] ailo: im going to select a machine here in a bit, and get on it [18:53] unless i get distracted, im about 2 hours from doing some tests [18:54] ailo: you think i should start with xubuntu daily? [18:55] holstein: I suppose so. It would make sense for each tester to have the same exact kernel, and preferably the same build of the iso [18:55] I can do some tests as well [18:55] Can't find the pages I made back then... [18:56] I found this on edubuntu's wiki https://wiki.edubuntu.org/generic_vs_lowlatency_testing [18:57] ailo: i want the most recent generic [18:57] and whatever abogani is hosting for lowlatency [18:58] not a mainline kernel.. [18:58] i mean, i dont think i want a mainline... what do you think? [18:59] holstein: What is a mainline? The -lowlatency should be exactly the same as the -generic, and both should ideally be the same version [18:59] The only difference would be the config file, with a few options altered [18:59] This way we know we are only testing the diff between the two configs for the kernels [18:59] https://wiki.edubuntu.org/generic_vs_lowlatency_test_results [19:00] ^^ thats interesting, and probably good enough for scott-work to use [19:00] thats why we need it ^^ [19:00] holstein: Here's a page with results from back then https://wiki.edubuntu.org/generic_vs_lowlatency_test_results [19:01] ailo: parole has problems too. It shows some videos really fat... not the right height to width ratio. It also seems to need libs we don't normally install to work right. I personally like xine even though it is quite old. But then I don't watch that many videos anyway. I think we are including one for completeness. I think the video editing apps have their own viewer... is that correct? [19:02] len_: I don't know much at all about video. I like totem, as long as is doesn't freeze just from searching the video file (fast forwarding). And I use mplayer or vlc when I need more options [19:03] holstein: We should really try testing for a bit longer periods, and preferably use a script. It shouldn't be too hard [19:05] ailo: my tests for my personal needs take overnite [19:05] The key is to make sure we get 0 xruns, before approving a setting is functional [19:05] 8 tracks and 24/96 into ardour overnite [19:05] with a zero xrun policy [19:06] ailo: Should we test a bunch of video players then? It seems they all use the same back ends anyway. I think we chose totem because it was the gnome or ubuntu standard. So it would be available. [19:06] That'll take you a few nights to complete with a few different jack settings [19:06] And multiple cards [19:07] len_: I really don't have an opinion about that myself. Whatever is easy to use, and is functional is ok with me [19:07] ailo: that was my thought/comment as well. [19:09] holstein: At a later stage, we did tests while compressing a file. That was a nice easy way to use up all of the CPU. Pretty simple to add that to a script [19:09] I could look at that tomorrow. Now I gotta sleep. Good night folks [19:10] I'm pretty excited about getting the -lowlatency in, and it's impressive how well US has progressed [19:13] eh, i'm still struggling to get it into REVU [19:13] i'll be honest, this is taxing both my patience and reserves, of which i tend to have both [19:13] er, "of which i tend to have plenty" [19:13] scott-work: -lowlatency, or US in general? [19:14] At least -lowlatency is in progress [19:15] lowlatency [19:16] i have had to explain _what_ it is and _why_ we need it to so many people at each step of the progression [19:16] and it seems like each step is difficult as well [19:16] like getting it into REVU, e.g. [19:16] first REVU was down and it appeared that no one knew it [19:17] then it turns out REVU is being deprecated, but is still used in some cases [19:17] and i don't think a kernel is a paritcularly good candidate for the latest new package process, but i might be mistaken on this [19:17] then when i squeek (as in squeeky wheel) REVU is restarted [19:17] then i can't upload because it seems to be not catching my key for some reason [19:18] i'm probably a moron and i'm doing everything wrong [19:18] scott-works: I see. Well, let me and holstein give you some test results shortly to give you some data to show for whoever is approving it. We need to explain what the limits are, and why we need to have our performance stable within certain limits [19:19] scott-work: line that link from edubuntu [19:19] link* [19:20] you can replace kubuntu.org or whatever it was, with ubuntu.com and the link is the "official" one, methinks [19:20] holstein: It was originally on Ubuntu wiki. I made those pages back then, but don't know what happened to them [19:21] Right https://wiki.ubuntu.com/generic_vs_lowlatency_testing [19:21] ubuntu.com, not ubuntu.org [19:22] ok.. im queing up a xubuntu daily alternate iso [19:22] IT IS DONE! http://revu.ubuntuwire.com/p/linux-lowlatency [19:22] sirestart is my hero :) [19:22] TheMuso: http://revu.ubuntuwire.com/p/linux-lowlatency [19:23] scott-work: so we are in? [19:23] we have it? [19:24] holstein: no, no...it's just in REVU, but it can now be reviewed by two MOTU (which one should be TheMuso ) and then after any (mostly likely) changes are made, it should be pushed to the archives [19:24] scott-work: cool [19:24] let me know what i can do then [19:24] but this is still a major milestone though :) [19:24] ill postpone getting crazy with tests then [19:24] i was just going to try and get you some docs.. more current ones [19:25] I'll make a script for doing tests tomorrow anyway, just for the sake of it [19:25] ailo: that would be great! [19:25] ailo: i feel like we were really getting good results [19:25] we meaning you were really helping me understand what to do [19:40] holstein: ailo: there is something very tangible that i would like to generate from testing the lowlatency kernel [19:40] i would really like to be able to give users expectations on what they should experience [19:41] this goes for everyone i guess as well, not just you two :) [19:42] i would really like to spend a little time testing kernel performance using the same procedure (as arbitrary and non-explicit as it might be) and document the results for the users [19:43] i have four machines i can do this with which are all rather different (i386, i386 with hyper threading, amd64 dual core, amd64 six core) [20:08] yeah, i think that near impossible though scott-work [20:08] i think so much of it depends on hardware [20:09] i have USB devices that, on the same machine and kernel that i can get 1.2ms with FW, can barely get 20ms stable [20:09] why?... i mean, we can assume driver support [20:09] but that doesnt mean we can put that in a document [20:09] i think the best we could do is offer a database where users can offer there experiences [20:09] make it editable for when someone finds something new [20:10] or when the kernel rev's and changes *everything* for that one hardware case ;0 [20:11] holstein: i'm not saying we just give a number [20:11] just a general expectation? [20:11] holstein: i would like to give a chart given our hardware and the latency we expereienced [20:11] scott-work: im into that.. certain hardware cases documented [20:12] holstein: aye [20:12] i think with my FW device, some internal ones, and ailo 's 1010, we should be good [20:12] i have a simple USB one too [20:12] i agree, and i think we have a broad range of hardware [20:13] even though this is not an explicit testing suite either, we still should be able to categorically demonstrate a performance expectation [20:13] and if we chose, we could even contrast this againt the generic kernel as well [20:13] sure [20:14] im in... ive recently aquired a few "not so bad" laptops devoted to testing [20:14] couple cores [20:14] 2 and 4 gb's of ram [20:14] not sure about testing FW on them though [20:14] i need to see about the chipset [20:25] Re k3b, bassero etc. Been playing with these two just now. I can not get k3b to put out a toc file before burning. or even while burning from what I can tell... brasero at least creates a bin and cue file, but when I try to split a track it only gives me the last 3 seconds to split. [20:26] Having a toc or cue file would allow manual editing to add a second language to the CD-TEXT stuff for example. [20:27] k3b splits tracks using an index mark, not making two tracks. some cdplayers will search for tracks but not index [20:30] brasero does offer an audition feature, but it does not seem to allow auditioning transitions. It just lets you hear what a track you are adding sounds like. [20:31] I will have to see (when I warm up) if a toc file created externally will be accepted by either program. [20:42] BTW a bin and cue file take more space than a toc file as the bin is the size of the CD. Both k3b and basero create files the size of the cd. A plain toc file uses the wav files as they are in place. A minor point with the size of disks now. [20:44] A bin and cue file or a toc with wav files is portable. cdrdao can run on different platforms so just these sets of files can print the same CD on almost any computer. [20:46] http://www.xbloome.com/drupal/howto/toc_cd_master gives a pretty good over view of a lot of what has been covered here. [20:49] Anyway, it does not appear, after further research, that k3b offers a great advantage over brasero. perhaps not enough to justify the kde libs being added. [21:25] Further testing reveals that basero will burn a cd from a TOC file, but k3b will not. Brasero is not bullet proof and when given a toc tries to find the size of all files involved... it failed or was taking a very long time, but when I killed it and tried the same toc with it again it did print it. [21:26] len_: k3b burns toc files here [21:26] or at least it did some time ago, I don't burn cds for 2months now [21:27] I just got a can't read that from it. The toc file was created by gcdmaster [21:28] I am installing todays iso (12.04) I'll install k3b fresh and try the same file on that. [21:28] hm, incompatible formats? [21:28] is TOC a standard? [21:29] K3B seems to have it's own internal format in XML. I think toc is standard only in that anything that uses cdrdao should be able to pass the toc to cdrdao. [21:30] I think toc is a cdrdao thing, cue/bin is more of a standard at least for windows apps. [21:30] as long as its the one that get spit out of ardour [21:30] i can ask las... [21:31] Ardour spits out wav files. It doesn't that I could find put out a ready for cd set of stuff. [21:31] len_: you can get toc from there [21:31] you can get only toc [21:32] if you want [21:32] but, las is going to be the expert [21:32]