IRC Logs

*******************************************************************
[03:00] <bluejay> mathrick|code: He might have meant that IE does something screwy (non-standard) with HTTP so it's faster with IIS.
[03:01] <alley_cat> ephy is missing two things for me: cookie/popup management and tabs on the left
[03:01] <alley_cat> i have to use galeon, it's the only browser that supports tabs on the left side :P
[03:03] lilo ([email protected]) left irc: "Lost terminal"
[03:03] <mathrick|code> bluejay: i don't know, but IE really is faster wrt. to dynamic things, and feels faster, but that's because it doesn't fully implement single standard. So it's cheating essentially :)
[03:03] lilo ([email protected]) joined #gstreamer.
[03:04] <mathrick|code> anyway, gonna sleep now, will give intelligence report from #StrMod tomorrow :)
[03:04] <jimmy_dean> mathrick|code: night
[03:05] <jimmy_dean> I think the gecko engine feels way faster than IE
[03:05] <mathrick|code> jimmy_dean: yep
[03:05] Nick change: mathrick|code -> mathrick|sleep
[03:12] Miko5881 ([email protected]) left irc: "ChatZilla 0.8.31 [Mozilla rv:1.4/20030928]"
[03:13] Rotty ([email protected]) left irc: Remote closed the connection
[03:14] alley_cat ([email protected]) left irc: "May the Source be with you!"
[03:38] <jimmy_dean> ds-work: hey, I got the FFTW library working on the data here...it seems to work ok though I have no idea what typical values I should see for each frequency component
[03:41] <jimmy_dean> ds-work: I seem to have 288 (0 to 287) samples for 16 bit int data type for source input...would this be correct?
[03:42] thaytan ([email protected]) left irc: Read error: 110 (Connection timed out)
[03:42] foser ([email protected]) left irc: "[ I want to believe ]"
[03:43] danb ([email protected]) joined #gstreamer.
[03:49] <ds-work> how would I know what is correct?
[03:50] NFusi0n ([email protected]) left #gstreamer.
[03:53] <jimmy_dean> ds-work: well 288 comes from this: num_samples = GST_BUFFER_SIZE(buf)/sizeof(double);
[03:53] <ds-work> yes, but how would I know how big your buffer is?
[03:53] <jimmy_dean> ds-work: I thought it was a standard size each time
[03:54] <jimmy_dean> 2 bytes worth of data
[03:56] <jimmy_dean> ds-work: I graphed in gnumeric the first 20 data points - looks like a frequency spectrum...I'm just wondering how I figure out the frequency range on each the components
[03:56] <ds-work> did you do a real fft?
[03:57] <ds-work> (actually, it doesn't matter)
[03:57] <ds-work> sample_rate/fft_size
[03:58] <jimmy_dean> ds-work: yes, I did real to real dft
[04:00] <jimmy_dean> ds-work: what kind of method do sofware equalizers usually employ to adjust the amplitude of each component (or only the adjusted components from the GUI).  Meaning, what kind of a scale do they use for the adjustments?  Do I use a 20log(gain) to go to the dB scale?
[04:01] <ds-work> why don't you research that yourself instead of asking me?
[04:01] <jimmy_dean> ds-work: well I've tried...there's not really any sites that discuss a software equalizer's design
[04:02] <jimmy_dean> ds-work: I've tried searching google a lot
[04:02] <ds-work> what does xmms do?
[04:02] <jimmy_dean> ds-work: and you seem to know what you're doing
[04:02] <jimmy_dean> ds-work: I've looked, couldn't quite figure out what was going on and nobody in their chatroom would answer me on several occasions
[04:17] <jimmy_dean> ds-work: ok, finally got an xmms response...they cheat :)
[04:17] <jimmy_dean> ds-work: they only *equalize* mpeg data sources because it's already in the frequency domain...any thing else like ogg is just sent out without doing any equalization
[04:17] <jimmy_dean> or in other words pcm data bypasses the equalizer
[04:19] monkeyiq ([email protected]) joined #gstreamer.
[04:59] <taaz> isn't there a ladspa eq you could just use?
[05:00] <jimmy_dean> taaz: ladspa?
[05:00] danb ([email protected]) left irc: "new kernels..."
[05:00] <taaz> http://www.ladspa.org/
[05:01] <taaz> http://plugin.org.uk/ladspa-swh/docs/ladspa-swh.html#tth_sEc2.33
[05:02] <jimmy_dean> taaz: that's only 3 freq. bands
[05:02] <taaz> http://plugin.org.uk/ladspa-swh/docs/ladspa-swh.html#tth_sEc2.65
[05:03] <jimmy_dean> taaz: ok, there's no link to the source code...do you know where the source for this might be found?
[05:04] <taaz> http://plugin.org.uk/
[05:04] Action: taaz thought that was obvious... ;)
[05:05] <jimmy_dean> well I mean specifically, like you specifically found those docs
[05:05] <jimmy_dean> :)
[05:06] thaytan ([email protected]) joined #gstreamer.
[05:07] <taaz> if you install those plugins in the ladspa plugin path somewhere and rerun gst-register then they are all available
[05:07] <taaz> gst-inspect ladspa-mbeq
[05:08] <taaz> looks like it has 15 bands
[05:11] <jimmy_dean> taaz: wait, it works with gstreamer?
[05:12] <taaz> it should
[05:12] <jimmy_dean> why didn't somebody mention that before?
[05:12] <jimmy_dean> all of this work :)
[05:14] <jimmy_dean> taaz: why doesn't anything like rhythmbox use it then?
[05:15] <taaz> code doesn't write itself
[05:15] <monkeyiq> heh
[05:15] Action: jimmy_dean falls over dead
[05:33] <jimmy_dean> taaz: hey, actually this plugin does it almost exactly how I did my equalizer so far...uses the exact same fourier transform library
[05:33] <jimmy_dean> I may see if I can't get Rhythmbox to use this for a graphical equalizer
[05:34] <taaz> those may be two seperate things
[05:34] <taaz> which is kind of sucky i guess
[05:35] <jimmy_dean> what do you mean?
[05:35] <jimmy_dean> two separate things?
[05:35] <taaz> just a eq with sliders not that hard.  just hook up to element params.
[05:35] <jimmy_dean> taaz: yeah, that's what I want to do
[05:35] <taaz> if you pretty frequency domain graphs though, you'd need to do some more work
[05:36] <jimmy_dean> taaz: oh no, that's way later...actually my plugin could provide the framework for that almost right now
[05:37] <jimmy_dean> taaz: how does one connect a GUI element in say Rhythmbox to the equalizer plugin from LADSPA?
[05:37] monkeyiq ([email protected]) left irc: Read error: 60 (Operation timed out)
[05:37] <taaz> probably the same way the volume slider works
[05:37] <jimmy_dean> I know you said element params, but what does that mean more specifically...are there docs?
[05:37] <jimmy_dean> taaz: ok, so look at how they did that in RB?
[05:37] thaytan ([email protected]) left irc: Remote closed the connection
[05:38] <jimmy_dean> taaz: the equalizer plugin from LADSPA looks almost just like a GStreamer plugin except it has it's own data types instead of the Gst ones...is there a coincidence there?
[05:38] danb ([email protected]) joined #gstreamer.
[05:39] <taaz> make a volume slider like class for the eq.  but a button next to the volume that pops up a eq widget.  put the eq in the pipeline near where the volume one is.
[05:39] <taaz> make sure it does passthrough when at unity
[05:39] <jimmy_dean> taaz: yeah, ok
[05:40] Action: taaz has never needed an eq...
[05:40] <jimmy_dean> taaz: all of my songs are flat without it...and I have a nice surround sound hookup...I want to hear my oggs to the fullest quality possible :)
[05:41] <jimmy_dean> taaz: plus RB is an iTunes clone, iTunes has a great eq
[05:41] <taaz> i've never seen iTunes
[05:41] <jimmy_dean> it's nice
[05:42] <taaz> and you do realize that using an eq is going to degrade the quality right?  fft/ifft are not totally lossless. (if i recall dsp correctly...)
[05:42] <taaz> not alot of course...
[05:43] <jimmy_dean> it's minor
[05:43] <jimmy_dean> it deals in a double precision float
[05:44] <jimmy_dean> most audio is int
[05:44] <jimmy_dean> you won't hear it
[05:52] <jimmy_dean> taaz: so if I build the LADSPA plugins, where should they be installed to again?
[05:53] <taaz> it's probably in the docs
[05:54] <taaz> somewhere in LADSPA_PLUGIN_PATH or something like that
[05:54] <taaz> i just apt-get install swh-plugins
[05:54] <jimmy_dean> taaz: ok...there's nothing special to do to use them with GStreamer then?
[05:54] <jimmy_dean> other than gst-register?
[05:54] <taaz> in theory
[05:54] <jimmy_dean> it hasn't been tested?
[05:55] <taaz> i tried some of them maybe 18+ months ago in gst-editor
[05:56] <jimmy_dean> taaz: excellent...well I can say that I know for a fact that a lot of xmms people will start using RB after it gets an equalizer
[05:56] <jimmy_dean> I'm one of them :)
[05:57] <jimmy_dean> taaz: plus something like totem with the gstreamer backend could utilize the equalizer for it's sound, correct?
[05:58] charles ([email protected]) joined #gstreamer.
[06:00] mathrick|sleep ([email protected]) left irc: Remote closed the connection
[06:20] <jimmy_dean> taaz: ahh nice, I have those plugins in portage for Gentoo :)
[06:22] egir ([email protected]) joined #gstreamer.
[06:22] <egir> hey everyone. can anyone point out the homepage for gst ladspa wrapper?
[06:25] ChrisHJW ([email protected]) joined #gstreamer.
[06:28] charles ([email protected]) left #gstreamer ("Client exiting").
[06:35] ChriHJW_log ([email protected]) left irc: Read error: 110 (Connection timed out)
[06:42] lilo ([email protected]) left irc: "brb"
[06:42] lilo ([email protected]) joined #gstreamer.
[06:44] lilo ([email protected]) left irc: Client Quit
[06:44] lilo ([email protected]) joined #gstreamer.
[06:51] <taaz> egir: gstreamer.net? ;)
[06:58] xcomp ([email protected]) left irc: Read error: 60 (Operation timed out)
[06:59] harshy ([email protected]) left irc: "I quit for now"
[06:59] harshy ([email protected]) joined #gstreamer.
[08:13] sub_pop ([email protected]) left irc: "Client exiting"
[08:15] egir ([email protected]) left irc: "leaving"
[08:15] Misirlou ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[08:20] ChrisHJW_log ([email protected]) joined #gstreamer.
[08:55] sub_pop ([email protected]) joined #gstreamer.
[08:56] Nick change: harshy -> harshyZ
[09:42] mkl ([email protected]) joined #gstreamer.
[09:52] Rotty ([email protected]) joined #gstreamer.
[10:12] Uraeus ([email protected]) joined #gstreamer.
[10:14] <Uraeus> morning
[10:17] <kmaraas> morning
[10:21] <Uraeus> kmaraas: typisk norsk å være oppen tidlig på en lørdag?
[10:22] bluejay ([email protected]) left irc: "Leaving"
[10:23] <kmaraas> Uraeus, når du har små barn ja :)
[10:23] <Uraeus> hehe :)
[10:37] <Uraeus> yoann: are you the Yoann who works for Mandrakesoft and who hacked on Nautilus for a while?
[10:51] <kmaraas> Uraeus, yes, the very same :)
[10:58] <Uraeus> guess the name is to special for it to possibly be someone else
[11:09] <yoann> Uraeus: yes :)
[11:09] <yoann> Uraeus: though I do not work for Mandrakesoft anymore
[11:11] <Uraeus> yoann: so where do you work these days? anything Linux related at all?
[11:12] <yoann> Uraeus: prelude-ids.org, sponsored by random company
[11:12] <Uraeus> yoann: so what brings you to #gstreamer ?
[11:13] <yoann> Uraeus: tried to trace some race yesterday
[11:16] <Uraeus> yoann: in gstreamer?
[11:17] <yoann> Uraeus: well, finally it sound more like gst-player issue
[11:18] <Uraeus> yoann: yeah, I have big gst-player issues myself, found out today that all the mpegs that freeze for me in gst-player works fine with gst-launch-ext
[11:18] <yoann> Uraeus: guess you"re going to use totem for a while :)
[11:19] <Uraeus> yoann: well I compile Totem with GStreamer too, and I have the same issue there :)
[11:19] <Uraeus> yoann: I guess I nag Julien when he returns on Sunday to fix them both :)
[11:30] Nick change: Uraeus -> Ura_out
[11:41] mathrick ([email protected]) joined #gstreamer.
[11:41] <mathrick> re
[12:08] iain ([email protected]) joined #gstreamer.
[12:09] <mathrick> hi iain
[12:10] <iain> hi ya
[12:10] <iain> someone on the alsa list said that they didn't think that it should be easier to get alsa set up
[12:10] <iain> because if you want an easy to set up system you should use windows
[12:11] <iain> luckily it wasn't a developer :)
[12:11] <mathrick> and what do devs think about setting alsa up?
[12:11] pb_ ([email protected]) joined #gstreamer.
[12:14] <iain> they're too busy writing lisp interpreters for the configuration file
[12:14] <iain> :)
[12:14] <mathrick> heh :)
[12:15] Action: mathrick haven't used alsa beyond old 0.9 series
[12:15] <mathrick> s/haven't/hasn't/
[12:15] <iain> I think I'm using a version just before they released 1.0
[12:16] <mathrick> but it was tricky to set even back then :)
[12:29] <hyriand> tricky? what's so tricky about setting up alsa?
[12:30] alley_cat ([email protected]) joined #gstreamer.
[12:33] water ([email protected]) joined #gstreamer.
[12:34] <mathrick> hyriand: well, it was always fun to discover proper combination of aliases / chosen cards / whatever after setting up fresh box. And alsa-config (or whatever is in there in Debian) was of limited help, although necessary to get it running at all
[12:34] thaytan ([email protected]) joined #gstreamer.
[12:35] natyaht ([email protected]) joined #gstreamer.
[12:36] <mathrick> hi thaytan, natyaht, whichever is true one ;)
[12:37] natyaht ([email protected]) left irc: Client Quit
[12:37] <thaytan> must remember - single click to launch xchat :)
[12:37] <hyriand> mathrick: ah okay.. I must be really lucky, gifted or something then :)
[12:49] Action: mathrick is really amazed by SCO's stupor
[12:50] sublett ([email protected]) joined #gstreamer.
[12:52] <hyriand> mathrick: the europe license thing?
[12:54] <mathrick> hyriand: yes, especially after being impotent to provide any evidence :)
[12:55] <hyriand> I liked that IRC "log" from a couple of days ago :)
[12:55] <mathrick> hyriand: but their whole show is funny to watch
[12:55] <mathrick> hyriand: hmm? what log?
[12:55] Action: mathrick reads /. too rarely
[12:56] <hyriand> I stay away from it like the plague.. I have people scanning it for me and giving me useful urls ;-)
[12:57] <hyriand> anyway, it was a chatlog between ehm.. SuSE, IBM, SCO and some other unix / linux related companies
[12:58] LeRoutier ([email protected]) joined #gstreamer.
[12:58] <hyriand> (fake of course)
[12:58] <LeRoutier> hi
[12:58] <mathrick> hyriand: real thing? or 'creative one?
[12:58] <mathrick> hyriand: ahh, ok :)
[12:58] <mathrick> yo LeRoutier
[12:59] BBB ([email protected]) joined #gstreamer.
[13:00] Action: BBB wanted to poke thomas...
[13:00] <BBB> but he's not here :/
[13:00] <mathrick> hi BBB
[13:00] <BBB> heyu
[13:00] <BBB> -u
[13:01] <mathrick> (I want a really small job (2 hrs/week max.) that will make me $2M per
[13:01] <mathrick> month.)
[13:01] <mathrick> BBB: good one :)
[13:02] <BBB> I should've said ¬
[13:02] <BBB> just to make my point
[13:02] <BBB> :p
[13:03] <mathrick> BBB: but really, this guy using "overbloated glib" and "small, primitive memory handling" and "GC" in single sentence is really nuts
[13:03] <mathrick> especially "primitive" and "GC" part is funny :)
[13:03] <BBB> yes, he's funny
[13:03] Action: mathrick thinks about what use is primitive GC of
[13:04] <mathrick> http://lists.matroska.org/pipermail/media-api/2004-January/000206.html <-- ROTFL
[13:05] <mathrick> someone, tell him about systemwide shared text pages, and why it saves mem :)
[13:06] <iain> urgh
[13:06] <iain> there was once someone who had a great idea for "shared functions"
[13:06] <mathrick> not to say that any single "chunk" of GLib will probably require most of other ones in order to not duplicate their code
[13:07] <mathrick> iain: what would they do?
[13:07] <iain> which was basically a shared library concept but with each library containing only one function
[13:07] <mathrick> buahuahua :)
[13:07] <iain> this was his wonderful idea so that you could only link in the exact functions that you needed
[13:07] <iain> and therefore save memory
[13:07] <thaytan> brilliant!
[13:07] <thaytan> applause!
[13:07] Action: mathrick imagines 100k lib headers residing in mem
[13:08] <mathrick> iain: but this guy is even better
[13:09] <mathrick> iain: he'd like to link needed chunks into lib fitting your app's needs at compile time, therefore producing different copy for every user :)
[13:10] <BBB> he wants shlibs to die, if I read hi correctly
[13:10] <BBB> because it's overheaded code
[13:10] <BBB> ...
[13:11] <mathrick> yeah, right, and then let's introduce brand new, revolutionary OS. Ladies & Gentlemen, Sys V!
[13:11] <mathrick> ;)
[13:12] <BBB> an after that, we'll introduce a new programming language, D! much cooler and more efficient than C
[13:12] <mathrick> hell, no, let's drop overhead of VM, L & G, MS-DOS 1.0!
[13:12] <iain> mathrick: hmm, thats quite nice :)
[13:12] <iain> drat...I can't find the shared function idea
[13:12] <iain> but I did find http://www.advogato.org/person/aaronl/diary.html?start=48
[13:12] <iain> which is always funny
[13:13] Action: iain wonders what happened to aaronl
[13:14] <BBB> whee!
[13:14] <BBB> 500k!
[13:14] <BBB> l00k @ m3!
[13:14] <BBB> 1 4m s0 c001!
[13:14] <BBB> I just saved 500k by removing the gnome libs and well, my whole system is completely unusable now because I have no DE, but 1 s4v3d 500k! w00t!
[13:14] Action: BBB thinks aaronl must have been a real cool person
[13:15] <mathrick> yeah, let's use MS-DOS 1.0! Only 8k RAM for complete OS!
[13:15] <BBB> and no overheady security stupidity
[13:15] <iain> BBB: the funny thing was that his contributions to software involved ripping out the gnome dialog code
[13:15] <BBB> to gtk?
[13:15] <iain> from applications
[13:16] <iain> and then adding it back in as static functions in a file called fuckgnome.c
[13:16] <mathrick> contrabutions, maybe ;)
[13:16] <iain> missing the point of sharing the code as a library completely
[13:16] <iain> just so he wouldn't have that 500k of shared libraries loaded
[13:16] <mathrick> and instead have 440k static funcs?
[13:17] <mathrick> and after loading 10 apps, 4400k?
[13:17] <mathrick> he must have been kewl h4x0r then
[13:18] <BBB> :)
[13:18] <iain> IIRC he was an intern for VA Linux
[13:18] <mathrick> VA Linux? isn't it dead now?
[13:18] <iain> Raster told him to fuck off and stop being an annoying fan boy
[13:18] <BBB> raster?
[13:19] <iain> Oh, he disappeared from gnome hacking around 2000
[13:19] <mathrick> BBB: rasterman, author of E
[13:19] <mathrick> http://www.advogato.org/person/aaronl/diary.html?start=44
[13:19] Action: iain realises that he doesn't really care what AaronL is doing now
[13:20] <mathrick> another kewl hack to rip out X from XEmacs :)
[13:21] Action: mathrick has bright idea. Let's fasten up GMP by ripping that useless multiprecision thing and using 32bit ints only ;)
[13:21] <iain> there's a great flamewar between him and itp where he just goes on about how stupid itp is
[13:21] <iain> (and I know itp personally, and he's not exactly what I would call stupid)
[13:21] <mathrick> iain: him == aaronl?
[13:21] <iain> yeah
[13:21] <mathrick> and who is itp?
[13:21] <iain> ian peters
[13:21] <mathrick> ahh
[13:22] <iain> wrote red carpet and iagno
[13:22] <iain> and stuff like that
[13:22] <mathrick> iain: but he used gnome-dialog. That proves him stupid
[13:22] <iain> yeah, sad that
[13:22] Action: iain goes for a shower
[13:25] Action: BBB goes for continuing on ffmpeg
[13:25] <BBB> bbl :)
[13:25] BBB ([email protected]) left irc: "Client exiting"
[13:35] Nick change: Ura_out -> Uraeus
[13:45] <mathrick> Uraeus: wb
[13:45] <Uraeus> thanks mathrick :)
[13:46] <Uraeus> I went out to buy a wireless print server, turned out I had to pay in advance and that it has a delivery time of 40 days
[13:46] <mathrick> buahuahua ;)
[13:46] <thaytan> does anyone have access to mpeg specs?
[13:46] <mathrick> Uraeus: happy waiting :)
[13:48] <Uraeus> thaytan: I expect that BBB has them
[13:49] <mathrick> thaytan: try http://bmrc.berkeley.edu/research/mpeg/faq/mpeg2-v38/faq_v38.html
[13:49] <thaytan> my subtitles are the wrong colour, and I haven't figured out why yet
[13:49] <thaytan> but they draw the right shapes at least
[13:49] <thaytan> and at the right times too :)
[13:49] <mathrick> anyone knows why suddenly I can't Alt-Tab w/ MCity while dragging galeon tabs?
[13:50] <iain>  cos galeon has a keyboard grab when you drag?
[13:51] <mathrick> iain: it worked earlier, and didn't bump galeon version
[13:52] <iain> duno then
[13:52] <mathrick> iain: it just stopped to work:\
[13:52] <mathrick> and how am I supposed to drag tabs between windows? doing many small ones and dragging is PITA
[13:56] Nick change: iain -> iain-last-exam
[13:56] <iain-last-exam> woohoo, back in 6 hours
[13:57] <mathrick> iain-last-exam: g'luck :)
[14:03] yoann ([email protected]) left irc: Remote closed the connection
[14:15] kmaraas ([email protected]) left irc: Read error: 110 (Connection timed out)
[14:19] Nick change: apoc-away -> apoc
[14:25] kmaraas ([email protected]) joined #gstreamer.
[14:28] <Uraeus> morning apoc
[14:32] <apoc> morning Uraeus
[14:33] <mathrick> hello world! ;)
[14:35] <Uraeus> hmmm, wonder if ds-work commited the new autoplugger to SF cvs on purpose :)
[14:35] TD ([email protected]) joined #gstreamer.
[14:35] <mathrick> Uraeus: maybe he didn't update CVS root there. It's in gst-sandbox, right?
[14:35] <TD> does anybody know of a writeup of the problems basicomega has on i686 glibcs?
[14:36] <mathrick> TD: ohh, welcome, DirectShow guru ;)
[14:36] <TD> haha. far from it
[14:36] <TD> BBB knows more about directshow than me
[14:36] <mathrick> TD: no way, we talked about it yesterday :)
[14:37] <TD> hmmm, it is a sad state if the best guru we have on directshow has never written an app using it ;)
[14:37] <mathrick> TD: it turns out noone has really clue about how it works ;)
[14:37] Action: TD chuckles
[14:37] Zahir ([email protected]) joined #gstreamer.
[14:37] <TD> ah well. i know who *does* know about dshow though. if you have questions i can ping them to rob shearman
[14:37] <TD> he's busy reimplementing dshow for wine
[14:38] <Zahir> hi
[14:38] <mathrick> TD: well, it's basically about some chit-chat and stealing ideas, as well as pumping our ego about how much better we are ;)
[14:38] <mathrick> hi Zahir
[14:39] <TD> :)
[14:39] <TD> so, schedulers. i'm curious to find out exactly what goes wrong on i686 glibcs with basicomega
[14:47] Zahir ([email protected]) left irc: " adeu"
[15:00] <thaytan> hah
[15:00] <thaytan> so, apparently nav events make audio stutter for a very good reason
[15:01] <thaytan> there is a missing 'break;', and it is falling through
[15:01] <thaytan> and then causing a flush for every nav event
[15:01] <thaytan> how embarassing :)
[15:02] <mathrick> thaytan: it's nothing compared to 1,5 day debugging just to find out that -g got mysteriously swallowed in Makefile
[15:02] <thaytan> suxor!
[15:04] <mathrick> thaytan: is it nice and smooth now? Can I pick up girls with dvdnav yet? ;)
[15:04] <thaytan> it is here
[15:04] <thaytan> not quite finished yet - menu overlay needs a little more work
[15:05] <thaytan> oh, and figuring out colours
[15:05] <thaytan> and testing on loads of DVDs ;)
[15:05] <thaytan> and ...
[15:06] <mathrick> thaytan: heh, good i don't own dvd, nothing to worry about ;P
[15:06] <mathrick> but, maybe i'll d/l some dvd .iso's off the net, they should do the trick :)
[15:09] <mathrick> thaytan: is it possible to read .vob with dvdnavsrc?
[15:09] <thaytan> not sure
[15:19] <Uraeus> TD: afaik so does nobody know what makes basicomega fail in i686 glibc (which is why nobody has fixed it)
[15:19] <Uraeus> TD: and omega who made it hasn't been active since before we discovered the issue
[15:20] Zahir ([email protected]) joined #gstreamer.
[15:21] <Uraeus> TD: but what was said at the time is that there is threading is done with a separate implemenatition on i686 compared to i386 and non-86 archs
[15:21] <Uraeus> TD: which is why we thought that maybe NTPL would solve the issue, but no such luck
[15:22] <Uraeus> TD: or rather I thought it worked for a while with the first beta's that RH did with NTPL, but then it started to show again, could be that they turned on the optimization that cause the issue or something
[15:23] <TD> hmm, ok
[15:24] <Uraeus> TD: I did some testing back when we first discovered the issue, let me see if I can find the mail again
[15:32] <Uraeus> TD: I found an old mail from Wingo which might contain the answer:
[15:32] <Uraeus> This is the reason that cothreads was borked on your redhat boxen,because they provided 686 packages which enabled this new feature. Thething is, getrlimit returns -1 for the max stack size for od libc's, sothis isn't of a whole lot of use. Maybe I can get the cothreads tests topass reliably if I think hard enough about it, though... I wuold beinterested about the possibility of 8M+ stacks, come to think of it.
[15:33] <Uraeus> um, not sure that small blurb really says anything
[15:33] <Uraeus> TD: I can forward you the mail if you want
[15:33] <mathrick> Uraeus: let me guess, you use Evo
[15:34] Action: mathrick kicks Evo for eating newlines when copying
[15:35] <Uraeus> mathrick: yes, I use evo :9
[15:36] <jimmy_dean> does anyone know how to use a LADSPA plugin from a program like rhythmbox?  I got it registered with gst-register-0.7 but I only see ladspa - 93 features...and I guess the thing  I want to access is one of those features
[15:38] <mathrick> jimmy_dean: gst-inspect-0.7 ladspa or so
[15:38] <mathrick> jimmy_dean: you will get elements names
[15:38] <jimmy_dean> mathrick: ahh ok
[15:39] <jimmy_dean> mathrick: there it is...there's a multiband EQ in there already...so I don't have to finish writing mine :)
[15:39] <jimmy_dean> mathrick: and it uses the same fftw library for the DFT as I did, so I was basically just recreating this same plugin
[15:40] <mathrick> jimmy_dean: heh :)
[15:41] <mathrick> maybe it's worthy to think about some kind of iface for that, after all, not everyone will have LADSPA installed
[15:42] <mathrick> small, standalone plugin may be useful, but i think there are other better suited to decide that :)
[15:42] <jimmy_dean> mathrick: iface?
[15:42] <mathrick> jimmy_dean: interface
[15:43] <jimmy_dean> mathrick: ahh, well what I might do is if I get it working in rhythmbox and it works well, I'll finish my plugin using it's code as a guide so gstreamer can have it's own
[15:43] <mathrick> jimmy_dean: gst uses interfaces for inedepndent impls of similar things, like xvimagesink vs. ximagesink
[15:44] <mathrick> or volume iface
[15:44] <jimmy_dean> mathrick: so are you saying it's an interface that can generically reference any backend plugin to do the job or it might even not force the user to have it at all?
[15:45] <mathrick> jimmy_dean: sort of, it provides user with fallbacks
[15:45] <jimmy_dean> ok, so rhytmbox does this with the volume control?
[15:45] <mathrick> jimmy_dean: like, when there is no hw mixer, we use volume element
[15:46] <jimmy_dean> ahh
[15:47] <mathrick> s/volume interface/mixer interface/
[15:47] <jimmy_dean> ok, well I'll keep that in mind after I get an initial thing going
[15:47] <jimmy_dean> alright, well I'll be back, gotta reboot to change to my new kernel
[15:48] jimmy_dean ([email protected]) left irc: "Carrier Signal Lost"
[15:48] <TD> Uraeus: [email protected], thanks
[15:50] <Uraeus> TD: ok sent two mails
[15:50] <TD> ta
[15:50] <Uraeus> TD: are you considering trying to debug it?
[15:52] <TD> i don't have time at the moment, i was just curious
[15:53] <Uraeus> TD: well I am not sure you can solve it, 3 people has already broken their back on the issue without actually coming any closer to understand exactly why :(
[15:55] <mathrick> Uraeus: what are "i686 glibc's" anyway? -march=686?
[15:55] <TD> the emails don't tell me a whole lot. basicomega is cothreads?
[15:55] <Uraeus> mathrick: yes
[15:55] <mathrick> Uraeus: why would -march change anything in glibc behaviour?
[15:55] <Uraeus> TD: it use cothreads yes, you need to ask someone like Wingo for more details
[15:56] <TD> i've looked at the cothreads hack. why do you even want this?
[15:56] <Uraeus> mathrick: well that is part of why this riddle is unresolved :)
[15:56] <mathrick> TD: because they're faster
[15:57] <TD> hm
[15:57] <TD> is scheduler overhead really that big a deal?
[15:57] <Uraeus> mathrick: I was told something along the lines that glibc had some special optimisations for i686 compared with other plattforms, which could explain why things only fail on that particular configuration
[15:57] <Uraeus> TD: probably not for media playback, but for other contexts 
[15:58] <mathrick> TD: yep, context switching is what scheduler's doing most of the time
[15:58] <TD> ok
[16:05] <thaytan> alright, I'm off to bed
[16:05] <thaytan> g'night
[16:05] <mathrick> nite
[16:06] matrixise ([email protected]) joined #gstreamer.
[16:07] <Uraeus> night thaytan
[16:12] Nick change: harshyZ -> harshy
[16:13] <mathrick> wb harshy
[16:15] <harshy> ;)
[16:16] jimmy_dean ([email protected]) joined #gstreamer.
[16:17] <mathrick> jimmy_dean: how's your new kernel?
[16:17] <jimmy_dean> mathrick: excellent :)
[16:17] <jimmy_dean> mathrick: it's not 2.6 yet though
[16:17] <jimmy_dean> mathrick: waiting for my wireless USB card drivers to get into Gentoo's stable branch
[16:30] Zahir ([email protected]) left irc: "vinga, deu"
[16:49] water ([email protected]) got netsplit.
[16:49] sub_pop ([email protected]) got netsplit.
[16:49] wheels ([email protected]) got netsplit.
[16:49] sxpert ([email protected]) got netsplit.
[16:49] sxpert_work ([email protected]) got netsplit.
[16:51] apoc_ ([email protected]) joined #gstreamer.
[16:51] water ([email protected]) returned to #gstreamer.
[16:52] sub_pop ([email protected]) returned to #gstreamer.
[16:52] wheels ([email protected]) returned to #gstreamer.
[16:52] sxpert_work ([email protected]) returned to #gstreamer.
[16:52] sxpert ([email protected]) returned to #gstreamer.
[16:52] #gstreamer: mode change '+b *!*omg@*.client2.attbi.com' by leguin.freenode.net
[16:52] hallibaby ([email protected]) joined #gstreamer.
[16:53] sub_pop ([email protected]) left irc: Excess Flood
[16:53] hallibaby ([email protected]) left irc: Client Quit
[16:53] sub_pop ([email protected]) joined #gstreamer.
[17:08] apoc ([email protected]) left irc: Read error: 110 (Connection timed out)
[17:10] Nick change: Uraeus -> Ura_book
[17:27] fsfman ([email protected]) joined #gstreamer.
[17:27] matrixise ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[17:36] hallibaby ([email protected]) joined #gstreamer.
[17:41] ct_ ([email protected]) joined #gstreamer.
[17:41] <Ura_book> hi ct_
[17:42] <ct_> Ura_book: hey!
[17:42] Nick change: Ura_book -> Uraeus
[17:42] <ct_> Uraeus: finished up school and everything
[17:42] <Uraeus> ct_: so what is next for you then? got a job already? or going to do something else first?
[17:43] jimmy_dean ([email protected]) left irc: Read error: 110 (Connection timed out)
[17:49] mxpxpod ([email protected]) left irc: "Guns kill people as much as spoons made Rosie O'Donnell fat.."
[17:50] Action: ct_ is away: brunch and out for the afternoon!
[17:50] #gstreamer: mode change '+o Uraeus' by ChanServ!ChanServ@services.
[17:51] <Uraeus> New website - http://www.freedesktop.org/~gstreamer/ | GStreamer article - http://www.osnews.com/story.php?news_id=5648
[17:51] Topic changed on #gstreamer by [email protected]: New website - http://www.freedesktop.org/~gstreamer/ | GStreamer article - http://www.osnews.com/story.php?news_id=5648
[17:51] #gstreamer: mode change '-o Uraeus' by [email protected]
[17:51] Uraeus ([email protected]) left irc: "Client exiting"
[17:55] Nick change: fsfman -> matrixise
[18:00] mxpxpod ([email protected]) joined #gstreamer.
[18:07] <TD> i'd suggest s/recent older news/recent news/
[18:08] foser ([email protected]) joined #gstreamer.
[18:11] <mathrick> TD: yep, "recent older news" is cute :)
[18:21] matrixise ([email protected]) left #gstreamer ("Leaving").
[18:28] jimmy_dean ([email protected]) joined #gstreamer.
[18:31] BBB ([email protected]) joined #gstreamer.
[18:49] jimmy_dean ([email protected]) left irc: Read error: 110 (Connection timed out)
[18:51] jimmy_dean ([email protected]) joined #gstreamer.
[18:52] BBB ([email protected]) left irc: "Client exiting"
[19:03] bosko ([email protected]) joined #gstreamer.
[19:03] <bosko> Hello all
[19:04] <alley_cat>   /* paranoia check to make sure we're in the right thread */
[19:04] <alley_cat>   g_assert (ctx->thread == g_thread_self());
[19:04] <alley_cat> ^^ that assertion fails with basicomega scheduler here :P
[19:06] <bosko> I am writing a gtk2 program and I need to play a movie, is there a gstreamer widget that I can use in program for playing movies?
[19:07] <bosko> The docs mention GstMediaPlay, but doesnt explain how to use it. Is this what I should be using?
[19:07] <mathrick> jimmy_dean: hmm, one more thing about eq. When you translate into freq domain, you only have finite number of freqs. And regardless of how big that number is, there are always more components in real signal. So either it means losing significant part of signal, or doing some magic over whole bands. And I have some troubles imagining that magic ;) (I.e, I can imagine it with physical band pass filter, but not with doing integral :)
[19:08] <jimmy_dean> mathrick: yeah, it would group several frequencies together into a range and adjust the level of each individual group
[19:09] <jimmy_dean> mathrick: obviously there's too many frequencies alone to make a practical amplifier.  The ladspa eq plugin I'm trying to use has 15 bands so that means 15 frequency groupings
[19:09] <jimmy_dean> mathrick: s/amplifier/equalizer
[19:10] <mathrick> jimmy_dean: hmm, you can't adjust whole band w/o clamping it all into one frequency when emitting again, can you?
[19:10] <mathrick> jimmy_dean: so in fact, you replace band with single freq?
[19:10] <jimmy_dean> mathrick: no, it's not clamping it into one frequency (or at least i don't think so), it's just grouping them saying that frequencies 14k through 16k will be adjusted when you adjust the 15k GUI slider
[19:13] <mathrick> jimmy_dean: but how do you preserve whole band when translating into freq domain? if you replace whole range with single 'bar', it's impossible to get that info back when emitting after adjusting levels
[19:14] <jimmy_dean> mathrick: no, it's just a mapping...it's not combining the frequencies into only one frequency...it's grouping the magnitude
[19:15] <jimmy_dean> it's like a loop then, I loop through the frequencies in a grouping and individually change their values to be the same magnitude as the *mask* frequency that you see on the equalizer's GUI
[19:15] <jimmy_dean> so say I change the 16k frequency slider on the GUI
[19:15] <jimmy_dean> and say I group 15k through 17k with the 16k slider
[19:16] <jimmy_dean> that means that I make all frequencies between 15k and 17k to be the same magnitude
[19:16] <jimmy_dean> and by magnitude I don't mean the frequency value, I mean the vertical bar that you see in the frequency domain graph
[19:17] <mathrick> jimmy_dean: then I think i need more background on FT. Do you get more components when applying *the same* FT over signal with more source components. Or is it FT params that dictate number of separate freqs visible afterwards?
[19:18] <jimmy_dean> mathrick: it depends really on your data source...if there's no sound data in the 16k frequency range, then really you're not adjusting anything by changing the slider
[19:18] <mathrick> jimmy_dean: hmm, let me rephrase
[19:18] <jimmy_dean> mathrick: k
[19:18] <mathrick> jimmy_dean: lets take that picture from before
[19:19] <jimmy_dean> k
[19:19] <mathrick> jimmy_dean: i have some FT with given params, and apply it to 1) alone. Obviously it returns only single sine wave
[19:20] <jimmy_dean> it returns a single value, yes
[19:20] <jimmy_dean> it won't return a sine wave though, it'll return the FT of the sine wave
[19:20] <mathrick> jimmy_dean: now, lets take 1) + 2) = 3), and treat it w/ *the same* FT as applied to 1). Will it give 2 sines, or some 1 sine being sort of average between 1) and 2) ?
[19:21] <jimmy_dean> mathrick: it should give 2 i believe
[19:21] <jimmy_dean> actually it does
[19:21] <jimmy_dean> look at the right graph
[19:21] <jimmy_dean> there's 2 there
[19:22] <mathrick> jimmy_dean: ahh, so when you feed it with more complex input, it gives you more detailed output?
[19:22] <jimmy_dean> mathrick: yep
[19:22] <jimmy_dean> mathrick: it just does an FT on what's there
[19:22] <mathrick> jimmy_dean: ok, so now preserving band input seems indeed possible
[19:22] <jimmy_dean> nothing magical
[19:22] <jimmy_dean> yeah, exactly
[19:23] <jimmy_dean> mathrick: does it sorta make sense now?
[19:23] <mathrick> jimmy_dean: i think i need to look at my maths ref manual :)
[19:23] <mathrick> jimmy_dean: yes
[19:23] <jimmy_dean> mathrick: I wouldn't look at the math, that'll confuse you more unless you are really devoted to study it and trying it out :)
[19:23] <mathrick> jimmy_dean: i plan on doing math degree later, so np :)
[19:24] <mathrick> jimmy_dean: it's just waves i don't like back from my physics days :)
[19:24] <jimmy_dean> mathrick: ok, well there's a ton of resources on the web on the FT...mostly from all of the world's engineering schools
[19:24] <jimmy_dean> ahh, I love waves :)
[19:25] <jimmy_dean> mathrick: ok, I gotta shower...bbl
[19:25] <mathrick> jimmy_dean: what I totally fail to grasp is (warning, non native speaker :) difference between ideal wave and actual waves
[19:25] <jimmy_dean> mathrick: ahh, we can talk about that in about 20 mins if you are still around
[19:26] <mathrick> jimmy_dean: sure, g'luck ;)
[19:26] Action: mathrick should take shower too
[19:26] Nick change: mathrick -> mathrick|bath
[19:37] <iain-last-exam> iain and ain are both in use
[19:37] <iain-last-exam> and ain is some dumb bot
[19:37] Nick change: iain-last-exam -> ^iain^
[19:39] Nick change: Jara[zZ] -> Jaramir
[19:54] sublett ([email protected]) left irc: "I like food, food is good!"
[19:54] bosko ([email protected]) left irc: "Bezig met verlaten"
[20:04] Nick change: LeRoutier -> LeRaway
[20:29] Nick change: mathrick|bath -> mathrick
[20:29] <mathrick> ahh, I'm clean now :)
[20:30] <mathrick> ^iain^: how did it go?
[20:30] <^iain^> ummm, 2 of the three questions I could talk shit for 2 pages, and the other I ran out of stuff to write about after a page and a half
[20:31] <mathrick> ^iain^: what was the exam of?
[20:32] <^iain^> Practical Theology
[20:32] <^iain^> so I did a question on the grieving process
[20:32] <^iain^> one on dementia
[20:32] <^iain^> and community
[20:36] <mathrick> ^iain^: wow, nice major you have :)
[20:36] <^iain^> its a standard BTh degree
[20:38] <alley_cat> if a video filter plugin works in rgb colorspaces, which should i support? there are so many ...
[20:41] <mathrick> alley_cat: easy one is BE ARGB, as used in X. It's kinda standard RGB in gst, was even considered as only one supported when BBB wanted to drop masks
[20:42] <alley_cat> ok, right now i have BE RGB, easy enough to expand that
[20:45] Action: ^iain^ looks at his long TODO list and realises he can't be bothered doing any of it
[20:47] Zeenix ([email protected]) joined #gstreamer.
[20:48] <Zeenix> hi
[20:48] <mathrick> y0 Zeenix
[20:54] <^iain^> stuffit's website says that a .sit is the best compression you'll get anywhere
[20:54] <^iain^> then you look at the mozilla download sizes and laugh
[20:54] <^iain^> tgz: 39meg, tbz2: 29meg, sit: 57meg
[20:56] <mathrick> ^iain^: heh. bz2 is unbeatable as usual
[20:56] <^iain^> well, a large text file will kill bz2
[20:57] <^iain^> like I had a file that took gzip 6 minutes to compress
[20:57] <mathrick> ^iain^: hmm? isn't it best suited for text exactly?
[20:57] <^iain^> bzip2 was going for ~20 before I stopped it
[20:57] <^iain^> it was like a 1/2Gb text file :)
[20:57] <^iain^> bz2 would have been smaller probably, but it was just taking so damn long
[20:57] <mathrick> ^iain^: probably it was too good at compresing it :)
[20:58] <^iain^> yeah
[20:58] <mathrick> ^iain^: just set lower ratio then
[20:58] <mathrick> s/ratio/level/
[20:58] <^iain^> I figured gzip had got it small enough for me
[20:58] <mathrick> ahh, thats solution too
[20:59] <mathrick> and best to compress are .bmp's produced by MSPaint, they got ratios like 0.001 or smaller :)
[21:01] <^iain^> bmps blow :)
[21:06] <wheels> ^iain^: bzip2 is about an order of magnitude slower than gzip
[21:06] <^iain^> I noticed :)
[21:07] <wheels> ^iain^: I did some testing on it at one point -- actually I think it was ~20x slower for compression and 10x slower for decompression, but I'd have to check again...
[21:07] <^iain^> that is quite a lot
[21:07] <wheels> (and usually yeilds about 10% smaller compressed files)
[21:09] <mathrick> wheels: usually more than 10%
[21:10] <mathrick> wheels: 15-25% from my exp
[21:10] <mathrick> like: <^iain^> tgz: 39meg, tbz2: 29meg, sit: 57meg :)
[21:14] Nick change: mathrick -> mathrick|coding
[22:04] kleppari ([email protected]) joined #gstreamer.
[22:05] <kleppari> evening, how would gstreamer be suited to display a video stream from a b878 tv card'
[22:05] <kleppari> s/'/?/
[22:07] <taaz> ChrisHJW: ping?
[22:09] <Zeenix> kleppari: We used a b878 tv card on our video-wall
[22:11] <kleppari> and that worked fine?
[22:11] mxpxpod ([email protected]) left irc: "Guns kill people as much as spoons made Rosie O'Donnell fat.."
[22:18] <Zeenix> kleppari: of course
[22:19] <Zeenix> kleppari: but it was almost an year ago
[22:20] BBB ([email protected]) joined #gstreamer.
[22:21] m_wheels ([email protected]) joined #gstreamer.
[22:21] wheels ([email protected]) left irc: Nick collision from services.
[22:22] Nick change: m_wheels -> wheels
[22:22] <mathrick|coding> yo BBB
[22:23] <BBB> hi
[22:23] water ([email protected]) left #gstreamer ("The water has parted.").
[22:24] Nick change: TD -> TD[gone]
[22:29] Action: Zeenix is finally thinking to shift to debian
[22:29] <BBB> what do you use now?
[22:30] <wheels> Windows 3.1?  :-)
[22:30] <BBB> heh
[22:30] <BBB> wheels, go fix unicode bugs in the qt/gst bindings
[22:30] <BBB> :p
[22:30] <BBB> (yes, I read the kde list)
[22:31] <wheels> BBB: Well, I assumed it wasn't hypnotic.  ;-)
[22:31] <mathrick|coding> anyone wanna hear about StrMod spying? ;)
[22:32] Nick change: jimmy_dean -> jimmy_dean|AFK
[22:32] <wheels> BBB: But honestly -- I just don't give a shit right now.  We're dealing with tagging KDE 3.2 right now and that's pretty low on the priority list...
[22:32] <BBB> :)
[22:33] <wheels> (Plus I would put reasonable odds on him having problems with his file system's encoding, but we'll see.)
[22:37] Nick change: mathrick|coding -> mathrick
[22:38] hallibaby ([email protected]) left irc: "Bye bye"
[22:41] hallibaby ([email protected]) joined #gstreamer.
[22:44] wheels ([email protected]) left irc: "brb"
[22:46] <Zeenix> BBB: RH
[22:51] wheels ([email protected]) joined #gstreamer.
[23:00] Action: ds_ moos #gstreamer into submission
[23:00] Nick change: ds_ -> ds
[23:02] <BBB> hi
[23:02] <BBB> I had a bug for you, but forgot
[23:02] <ds> my very own bug?
[23:02] <BBB> yes
[23:03] <ds> I will love him and squeeze him and call him george
[23:03] <BBB> special rice because you are my friend
[23:03] <BBB> +p
[23:03] <BBB> ;)
[23:03] <BBB> george, as in bush?
[23:03] <BBB> omg
[23:03] <BBB> :D
[23:03] <ds> ok, I'll call him osama
[23:04] Action: BBB shutsup
[23:04] <BBB> I prefer saddam
[23:04] <BBB> but anyway, I'm working on autotoolization of ffmpeg, and it sucks very much
[23:04] <ds> should I go buy a new laptop today?
[23:04] <BBB> ar eyou good at asm?
[23:04] <ds> yeah
[23:04] <BBB> there's a bug in ffmpeg where it cannot compile using -fPIC, can you fix that without #if 0'ing the whole block?
[23:05] <ds> there are some parts of ffmpeg that are not PIC compatible
[23:05] <BBB> ehm, right :D
[23:05] <ds> I'd have to rewrite all that asm
[23:05] <BBB> only a small part, right?
[23:05] <ds> whatever function is disabled
[23:05] <LeRaway> i love ASM, only z80 asm unfortunatly :)
[23:06] <BBB> dct_quantize_MMX2 and dct_quantize_MMX, as far as I can see
[23:07] walters ([email protected]) joined #gstreamer.
[23:07] <ds> ok, I'll look at it
[23:08] <ds> I should get a DSP book today, too
[23:11] <mathrick> I talked with that StrMod guy yesterday, it looks like nice I/O abstraction layer. It's similar architectural-wise to gst, as it was initially planned to be media framework :)
[23:13] <mathrick> it has Modules that map to elems, and Plugs instead of pads. basics are pretty much similar, except that Plugs are two way - ie. push as well as pull is possible
[23:13] <BBB> hm... so R/W media?
[23:13] <BBB> that could be useful for us...
[23:13] <BBB> we currently don't do both at the same time, do we
[23:14] <ds> I thought about adding bidir pads
[23:14] <ds> it's scheduling hell
[23:15] <mathrick> it seems good suited for anything that involves serializable IO
[23:15] <mathrick> like net, ant IPC even
[23:15] <mathrick> s/ant/and/
[23:15] <mathrick> also, building multicast is easy, as you just use proper module and it casts to many sockets at once
[23:17] <mathrick> it may be nice to have some genreal serial IO plugin, that would be able to do any type of IO, as changing it is matter of swapping Mosules, just like elems
[23:19] Miko5881 ([email protected]) joined #gstreamer.
[23:20] <mathrick> the only tricky part is that it's written using C++ and templates, so it would take some work to map it to GObject. OTOH, shouldn't be extremely hard
[23:21] <BBB> I noticed that some ffmpeg codecs have started option mapping
[23:21] <mathrick> BBB: hmm, what is that?
[23:21] <BBB> so it should be possible to fully describe some ffmpeg elements in gstreamer in future
[23:21] <BBB> mathrick: same as gobject gparamspecs
[23:21] <mathrick> BBB: ah, nice
[23:21] <BBB> well, no, because they're just using basic types
[23:21] <BBB> so no nice enums or so
[23:22] <BBB> just integers and strings and floats
[23:22] <BBB> :(
[23:22] <mathrick> well, someone should tell them about GObject then :)
[23:22] <BBB> I rather live
[23:22] <mathrick> BBB: are they that bad?
[23:23] <BBB> some of them
[23:23] <BBB> ;)
[23:23] <BBB> some are quite ok
[23:23] <BBB> others are just... well... bad
[23:23] <mathrick> BBB: how many points in MPlayer devs units? ;)
[23:23] <BBB> all the bad ones are mplayer devs too, actually
[23:23] <BBB> I wonder if it's related
[23:24] <BBB> some of the good ones are mplayer devs too, though
[23:24] <mathrick> BBB: you probably need to be evil to get MPlayer CVS RW ;)
[23:24] <BBB> don't want that ;)
[23:25] Action: mathrick wonders if MPlayer G2 will still require users to recompile on every mouse move
[23:25] Misirlou ([email protected]) joined #gstreamer.
[23:25] <mathrick> lo Misirlou
[23:26] <Misirlou> hi mathrick
[23:38] <Zeenix> should'nt this pipeline work: gst-launch filesrc location=~/downloads/alien.mpg ! mepgdemux video_%02d! { queue ! ffdec_mpeg1video ! ffcolorspace ! cacasink } ?
[23:38] <Zeenix> or s/cacasink/xvideosink
[23:38] <ds> you forgot a .
[23:39] kleppari ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[23:39] <Zeenix> ds: . where?
[23:41] <BBB> .video_00
[23:41] <BBB> instead of video_%02d
[23:43] <mathrick> Zeenix: and better separate that ! as some shells don't like it otherwise
[23:44] <Zeenix> cool, xvideosink jerks & cacasink runs so fast :)
[23:45] <mathrick> Zeenix: xvideosink is deprecated :)
[23:45] <Zeenix> mathrick: yeah but is there any alternative? non-xvideosink? ;)
[23:46] <mathrick> Zeenix: x[v]imagesink?
[23:51] bluejay ([email protected]) joined #gstreamer.
[23:52] hallibaby ([email protected]) left irc: "Bye bye"
[23:59] <thaytan> good morning all
[00:00] --- Sun Jan 18 2004
[00:01] <bluejay> Hello. Do the CVS gst-plugins fail to configure for anyone else?
[00:01] <BBB> morning?
[00:01] <BBB> night, you mean
[00:01] <BBB> no
[00:01] <BBB> how do they fail?
[00:02] <bluejay> the gst/colorspace directory is missing.
[00:03] <mathrick> BBB: he's aussie ;)
[00:03] <mathrick> bluejay: latest autobuild indicates that yes
[00:03] <mathrick> bluejay: gstcolorspace?
[00:03] <bluejay> mathrick: where do you see the autobuild results?
[00:03] <mathrick> bluejay: gst-daily list
[00:03] <BBB> eh?
[00:04] <bluejay> oh, /me isn't subscribed.
[00:04] <BBB> I've got code there
[00:04] <bluejay> It doesn't seem to be in CVS anymore
[00:05] <mathrick> bluejay: ds nuked large parts of it, maybe he did something nasty
[00:05] <BBB> bluejay: maybe someone removed it
[00:05] <BBB> anyone wanna fix, go ahead
[00:05] Action: BBB goes to bed
[00:05] BBB ([email protected]) left irc: "Client exiting"
[00:07] <ds> should i get an opteron notebook?
[00:08] <bluejay> yes.
[00:08] <bluejay> ;)
[00:09] <bluejay> ds: do you know anything about the gdt/colorspace directory going missing?
[00:09] <ds> bluejay: you didn't update the cvs directory
[00:09] <ds> cvs up -d
[00:10] <bluejay> whoah. I'd just been doing cvs up.
[00:11] <bluejay> I see. Is -d the only extra option I need to be using?
[00:12] Action: alley_cat uses cvs up -Pd
[00:18] <mathrick> bluejay: "Use the -d option to create any directories that exist in the repository if they're missing from the working directory"
[00:19] <bluejay> silly cvs. Why would I want to do anything else?
[00:22] <mathrick> bluejay: "but it has an unfortunate side effect.  If you  deliberately  avoided certain directories in the repository when you created your working directory (either through use of a  module  name or  by  listing explicitly the files and directories you wanted on the command line), then updating with -d will create those  directories, which may not be what you want."
[00:24] <bluejay> Won't cvs up alone get the files you explicitly skipped anyway? 
[00:24] Action: bluejay likes arch better. ;)
[00:25] <mathrick> bluejay: nope, not by default
[00:26] <bluejay> mathrick: mm, then that's something cvs can do that arch can't.
[00:28] mkl ([email protected]) left irc: "Leaving"
[00:30] <walters> maybe, but if you find yourself only checking out certain files often, that seems to imply that they shoudl have been separate categories in the first place
[00:32] <bluejay> I agree.
[01:15] thaytan ([email protected]) left irc: "See y'all"
[01:28] Nick change: apoc_ -> apoc
[01:28] <apoc> walters : ping ?
[01:34] <walters> apoc: yeah?
[01:35] <apoc> walters : I have a patch that add replaygain support to rhythmbox ... Is it too late to add this to arch/cvs for gnome 2.6 ?
[01:36] <walters> apoc: very cool!
[01:36] <walters> apoc: well, that depends.  does it affect the UI?
[01:36] <apoc> walters : no 
[01:36] <walters> apoc: i don't see why not then
[01:37] <walters> apoc: could you put it in bugzilla, under the current replaygain bug?
[01:37] <apoc> walters : yep ... I'll made the patch tomorrow
[01:38] <apoc> walters : thx
[01:38] <walters> apoc: cool, thanks :)
[01:40] Action: apoc goes to bed now ...
[01:40] <apoc> night all
[01:40] apoc ([email protected]) left irc: "Leaving"
[01:57] asdlfjkklsj ([email protected]) joined #gstreamer.
[01:57] <asdlfjkklsj> Anyone know of a gstreamer fedora RPM greater than version 0.6.3?
[02:00] md` ([email protected]) left irc: Read error: 60 (Operation timed out)
[02:00] foser ([email protected]) left irc: Remote closed the connection
[02:06] <wheels> asdlfjkklsj: well, the current stable branch is only at 0.6.4...
[02:07] md` ([email protected]) joined #gstreamer.
[02:11] md`` ([email protected]) joined #gstreamer.
[02:15] pb_ ([email protected]) left irc: "Client exiting"
[02:25] md`` ([email protected]) left irc: Connection reset by peer
[02:26] md`` ([email protected]) joined #gstreamer.
[02:26] <asdlfjkklsj> wheels: ah, gotcha... having issues in rhythmbox, where moving between songs causes a 1-6sec lockup when using gstreamer but not xine for the backend... happens in other media players
[02:26] <asdlfjkklsj> using fedora core 1, gs 0.6.3
[02:28] <LeRaway> asdlfjkklsj, it's a known problem, even with current CVS HEAD
[02:29] <LeRaway> so, no need to upgrade at all cost to newest, unless you want to discover new bugs
[02:33] md` ([email protected]) left irc: Connection timed out
[02:34] Zeenix ([email protected]) left irc: Read error: 60 (Operation timed out)
[02:38] <mathrick> nite
[02:41] wheels ([email protected]) left irc: Read error: 113 (No route to host)
[02:42] Nick change: mathrick -> mathrick|sleep
[02:48] alley_cat ([email protected]) left irc: "May the Source be with you!"
[02:57] lmjohns3 ([email protected]) joined #gstreamer.
[02:58] Nick change: md`` -> md`
[02:58] <lmjohns3> howdy
[02:58] <lmjohns3> ds-work: are you about ?


-------------------------------------------------------
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration
See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
http://www.eclipsecon.org/osdn