Back to Home / #mythtv / 2003 / 04 / Prev Day | Next Day
#mythtv IRC Logs for 2003-04-14

00:00<mdz>could make ConfigurationDialog do the help text
00:00<thor>Chutt, ok ... you're going to do the cvs.sql update?
00:00<Chutt>thor, dunno yet
00:00<mdz>all of those use that, I believe
00:01<Ndiin>ok, valgrind is completely useless.. it wont work because libGL.so is nvidias..
00:01<Chutt>i think those are the last users of ConfigurationDialog
00:01<Chutt>ndiin, using the latest nvidia driver?
00:01<Ndiin>yeah
00:01<mdz>with the default dialogWidget which just puts one configurable in a mythdialog
00:01<Ndiin>switching to xfree gls for the moment
00:01<Chutt>ndiin, there's a way to turn it off
00:01<Chutt>'sec
00:01<Ndiin>maybe it'll work there
00:01<Chutt>don't munge up your gl stuff
00:02<mdz>Chutt: were you thinking just help text for the screen, or to have the help be sensitive to which item in the list is focused?
00:02<Chutt>__GL_FORCE_GENERIC_CPU=1
00:02<Chutt>just help text for the screen
00:02<thor>Chutt, my code tree is way off CVS ... if I haven't seen the host stuff go in a couple of days, I'll do it all
00:02<Chutt>ndiin, so, "__GL_FORCE_GENERIC_CPU=1 valgrind mythmusic"
00:02<Chutt>thor, ok, cool
00:02<Chutt>thor, i just forsee the beginning of my week being busy with real work :(
00:03<mdz>damn
00:03<Chutt>ndiin, with the latest nvidia drivers, that is
00:03<mdz>finally finished building, and now I can't test it because something's recording
00:03<mdz>can't open the v4l device
00:03<Chutt>heh
00:03<Chutt>v4l2 does multiple opens, finally
00:03<thor>Chutt, fine. I want to get stuff done early this week as I'll be overseas the next
00:03<mdz>maybe I should just commit it and see if anyone complains
00:03<Chutt>i don't know what i'm going to do about the mandrake issue and v4l2 :(
00:04<Chutt>i wonder if there are any different defines i can key off of
00:04<mdz>mandrake has some old broken version of v4l2 or something?
00:04<Chutt>apparently
00:05<Ndiin>heh, dont think it was the same one, but I got this:
00:05<Ndiin>==26796== Invalid read of size 1
00:05<Ndiin>==26796== at 0x40FE795E: __GI__dl_close (in /lib/libc-2.3.1.so)
00:05<Ndiin>==26796== by 0x40FE84FE: do_dlclose (in /lib/libc-2.3.1.so)
00:05<Ndiin>==26796== by 0x4000A508: _dl_catch_error_internal (in /lib/ld-2.3.1.so)
00:05<Ndiin>==26796== by 0x40FE842E: dlerror_run (in /lib/libc-2.3.1.so)
00:05<Ndiin>==26796== Address 0x1E7 is not stack'd, malloc'd or free'd
00:05<Ndiin>Segmentation fault (core dumped)
00:05<-- mdz(~mdz@216-15-124-77.c3-0.smr-ubr3.sbo-smr.ma.cable.rcn.com) has left #mythtv ("Client exiting")
00:05<paperclip>provide a mdk kernel rpm =)
00:05<Ndiin>gonna try for some more useful info.
00:05-!-mdz [~mdz@216-15-124-77.c3-0.smr-ubr3.sbo-smr.ma.cable.rcn.com] has joined #mythtv
00:05<Chutt>ndiin, i'd compile mythmusic without the gl stuff
00:05<mdz>that ^W closing the channel thing has GOT to go
00:05<Chutt>just in case that's it
00:05<Chutt>mdz, heh =)
00:05<mdz>I just went to delete some words and left 3 channels
00:06<mdz>Chutt: is your tree in a commitable state again?
00:06<moegreen>Chutt: well, it got better when I turned off the debugging info from the ivtv driver. But still a bit choppy at the higher resolutions. I guess it's a slow hard drive issue
00:06<Chutt>mdz, yup
00:06<Chutt>my tree == cvs
00:06<mdz>Chutt: mind giving this input probing patch a quick test?
00:06<Chutt>of course
00:06<Chutt>moegreen, sucks
00:07<Chutt>could still be my network code, too =)
00:07<mdz>sent
00:07<mdz>thanks
00:07<Ndiin>Well, I just got a new bt, using the same method.
00:07<mdz>it's dead simple really
00:07<mdz>but I hadn't used qstringlist et al before
00:08<Ndiin>#3 0x08073096 in MadDecoder::deinit() (this=0x84855d8) at maddecoder.cpp:131
00:08<Ndiin>#4 0x08073fc7 in MadDecoder::run() (this=0x84855d8) at maddecoder.cpp:448
00:08<moegreen>Chutt: heh - well, the quality of the input is so much better even at the same resolution as my ati wonder ve - so... :)
00:09<Chutt>moegreen, it'll probably get better once the bitrate selection stuff goes in
00:09<mdz>brb
00:09-!-mdz [] has quit ["Client exiting"]
00:09-!-mdz [~mdz@216-15-124-77.c3-0.smr-ubr3.sbo-smr.ma.cable.rcn.com] has joined #mythtv
00:09<-- mdz(~mdz@216-15-124-77.c3-0.smr-ubr3.sbo-smr.ma.cable.rcn.com) has left #mythtv ("Client exiting")
00:09-!-mdz [~mdz@216-15-124-77.c3-0.smr-ubr3.sbo-smr.ma.cable.rcn.com] has joined #mythtv
00:09<mdz>dammit
00:09<Chutt>heh
00:09<mdz>setting gtk2 to emacs keys didn't fix it
00:09<Chutt>which email address did you send to?
00:10-!-hfb [~hbarrett@pool1027.cvx12-bradley.dialup.earthlink.net] has joined #mythtv
00:10<mdz>ijr@po.cwru.edu
00:10<Chutt>hrm
00:10<Chutt>haven't gotten it yet
00:10<mdz>my ISP's relay accepted it at 00:07:12 -0400
00:10<mdz>hey I'll just paste it to the channel
00:11<Chutt>ijr@speakeasy.net
00:11<Chutt>will go directly to me
00:11<mdz>people.debian.org/~mdz/temp/probeinputs.diff
00:12<mdz>I'll use that address in the future
00:12<Morph>ah build is done.
00:12<mdz>I need to try to RMA this CPU again
00:14<Ndiin>(gdb) frame
00:14<Ndiin>#0 0x0809ebe7 in ~Buffer (this=0x5683514) at buffer.h:21
00:14<Ndiin>21 if(data)
00:14<Ndiin>(gdb) p data
00:14<Ndiin>Cannot access memory at address 0x5683514
00:14<Ndiin>Why in the world would that crash.
00:15<Chutt>p this
00:15<Ndiin>(gdb) p this
00:15<Ndiin>$1 = (Buffer * const) 0x5683514
00:15<Chutt>p *this
00:15<Ndiin>(gdb) p *this
00:15<Ndiin>Cannot access memory at address 0x5683514
00:15<Chutt>ah hah
00:15<Chutt>buffer's bad, there
00:15<thor>buffers bad in the destructor?
00:16<mdz>uninitialized maybe?
00:16<Chutt>the this value for buffer is bad there
00:16<Chutt>could be
00:16<Chutt>mdz, nice how you ignored the v4l2 stuff =)
00:16<mdz>Chutt: I can't even compile it
00:16<mdz>better to leave it alone than to break it, I say :-P
00:16<Chutt>yeah, but it's similar to the v4l stuff
00:17<Chutt>lemme fix it
00:18<mdz>oh, I didn't notice it autodetected whether v4l2 was there
00:19<Chutt>yup
00:19<mdz>I probably would have done it then
00:19<Ndiin>btw, I have an if in recycled.cpp's destructor: if (bufferwhatever[i]) { delete ... }, so wouldnt that make sure that the buffer is valid at the time?
00:19<Chutt>ndiin, no, it'd just make sure it's non-null
00:19<Ndiin>hmm, true
00:20<mdz>I took one look at #ifdef/#endif and ran the other way
00:20-!-PeteCool [~pete@modemcable019.237-201-24.mtl.mc.videotron.ca] has joined #mythtv
00:22<thor>Ndiin, don't have the code in front of me, is something doing an assignment to buff pointers that you can see?
00:23<Chutt>mdz, think i've got it all updated, compiling now
00:23<Ndiin>thor: looking..
00:23<mdz>Chutt: thanks
00:24<Chutt>finally got the email =
00:24<Chutt>err, =)
00:24<mdz>Chutt: it is in fact safe to use qstringlist as a return type, right?
00:24<mdz>I assumed it was since there were those static split functions which seemed to return one
00:24<Chutt>i believe so
00:24<Chutt>right
00:24<Ndiin>thor: only the *decoder.cpp's even get() a Buffer..
00:25<Ndiin>but they do memcpy()'s
00:26<Chutt>hmm
00:26<Chutt>could be writing past the end..
00:26<Chutt>mdz, works perfectly
00:26<Chutt>mind if i commit it?
00:26<mdz>nope
00:27<Chutt>done
00:27<Chutt>thanks =)
00:28<thor>Ndiin, you might want to do a quick grep on the variable names that are of type buffer. Could be they rarely get &point'ed somewhere else or that something else gets to pointed to one of them (and then the original destructs first)
00:28<Chutt>so the pvr-250 is pretty seamless, now
00:28<Chutt>need to setup some per-card profile settings
00:28<Chutt>somehow
00:28<thor>Chutt ... good timing, mine arrives tomorrow
00:28<Chutt>then do stuff to support edit mode
00:28<Chutt>and add support for seeking in playback
00:28-!-NonToxic is now known as zZzToxic
00:28<Chutt>but it can record and stuff
00:29<Chutt>watch live tv
00:29<Chutt>but the setup for it's all there =)
00:29<Ndiin>thor: that'd be a lot easier if they weren't all named "b" ;)
00:29<thor>they is neither Chutt's nor my fault :-)
00:29<moegreen>Chutt: is it possible to prefer one capture card over another? I originally configured my old card first - so it was picking that card instead of the mpeg2 card
00:29<mdz>Chutt: profile selection looks like it's going to be harder than I originally thought
00:29<thor>(MQ3)
00:29<mdz>there are a lot of criteria that would be useful to consider in the selection
00:30<Chutt>moegreen, not yet
00:30<Chutt>maybe eventually
00:30<Chutt>i've just modified mainserver.cpp to return the last free encoder, which is the pvr card
00:30<moegreen>Chutt: ok, well ... i just worked around it by configuring it first - so it doesn't matter much
00:30<Chutt>first =)
00:30<thor>ah moegreen, you're back ... have a sec
00:30<Chutt>mq3's decode pipeline was nice
00:30<moegreen>thor: yeah
00:30<Chutt>it was weird when i was looking for a general purpose decoding library
00:31<thor>starting to theme-ify mythmusic playback
00:31<thor>....
00:31<Chutt>looking at mq3, and thinking 'boy, does this look familiar'
00:31<Chutt>turned out to be _heavily_ based on freeamp, just cleaned up a lot
00:31<thor>got the feeling I was duplicating a lot of parsers
00:31<Ndiin>thor: Be nice to it, don't make it too pretty to be useful ;)
00:31<thor>was following mythfrontend/playbackbox
00:32<thor>and just keep feeling ... why am I parsing this .... I should be inheriting something that already knows how to parse this
00:32<thor>am I (1) lazy, (2), doing something wrong, (3) out to lunch
00:33<Morph>Chutt: should have a back trace for you in a second.
00:33<Chutt>morph, thanks
00:33<moegreen>thor: if you look at the progfind.cpp file in libmythtv - it uses more of the parsing code that is in the xmlparse. The playbackbox does more parsing to support extra widgets (the popupbox colors, video placement)
00:33<Morph>sure.
00:33<Chutt>hopefully somethin simple =)
00:34<thor>moegreen, ah ... that's probably why you told me to look there in the first place
00:34<moegreen>thor: heh, yeah - it's a bit easier to follow too
00:34<Chutt>moegreen, if you get a chance, check out the progfind patches that were posted to the lists
00:34<Chutt>that's all your code, wasn't sure if you liked it or not
00:35<Chutt>the stuff to only show subsets or whatnot
00:35<thor>moegreen, good ... thanks .
00:35<moegreen>Chutt: I took a look at the movie patch and it looked fine, but then the guy changed it a few times - thought I might let his patch settle down a bit first :)
00:35<Chutt>ah
00:35<Chutt>heh
00:36<Morph>Chutt: email work?
00:36<Chutt>sure
00:36<Chutt>ijr@speakeasy.net
00:36<Chutt>or ijr@po.cwru.edu
00:36<Chutt>both work
00:36<Ndiin>thor: I dont see anythink like that anywhere..
00:36<Chutt>or even just paste it in here
00:37<Morph>on its way.
00:37<Chutt>(the relevant parts, at least0
00:37<Ndiin>nothing getting deleted thats part of the recycler's buffers, etc..
00:37<thor>Ndiin, it may be the memcpy then ....
00:37<Ndiin>hmm
00:37<Chutt>morph, you sure you turned off deinterlacing?
00:37<Ndiin>maybe make it check Buffer->size()
00:37<Morph>Yup.
00:37<Chutt>doesn't look like it =)
00:37* Ndiintries
00:38* Morphdouble checks
00:38<Morph>int Deinterlace = 0
00:38<Chutt>err, where's that?
00:38* Ndiinadds an assert()
00:38<Morph> /usr/local/share/mythtv/settings.txt
00:39<Chutt>do the docs still reference that?
00:39<Ndiin>mythmusic: maddecoder.cpp:306: void MadDecoder::flush(bool): Assertion `b->size() < sz' failed.
00:39<thor>Ndiin, just to throw another alternative out there ... (from memory, so probably worng) ... buffer is short (unisigned?) int's, which may be getting badly copied as ints?
00:39<Ndiin>or did I just get my assert logic backwards?
00:40<Chutt>morph, in the frontend
00:40<mdz>sure would be nice if there were universal channel identifiers
00:40<Chutt>go to tv->setup->playback
00:40<Ndiin>ie I want it to fail if sz is larger than b->size()
00:40<mdz>which were the same between sources of program information
00:40<mdz>that would make a nice chanid
00:40<Chutt>deselect 'deinterlace playback'
00:41<Chutt>(arrow right/left on the checkbox)
00:41<Morph>OH. I see.
00:41<mdz>and when my channel lineup changes, as it just did, it wouldn't break my recordings
00:41<Chutt>go through the entire wizard
00:41<thor>Ndiin, cout some sizeof's?
00:41<Ndiin>thor: I'm just trying to make sure my assert logic isnt backwards.. it fails if true, correct?
00:42<Chutt>ndiin, it fails if it's false
00:42<Chutt>so
00:42<Ndiin>ah, i did get it backwards then
00:42<Chutt>it looks like it's overwriting the buffer size
00:42<Ndiin>no
00:42<Ndiin>size() should be larger
00:42<Chutt>ah, sorry
00:42<Chutt>it's late =)
00:44<Ndiin>uhm, okay..
00:44<Ndiin>b->size(): 2048 sz: 2048mythmusic: maddecoder.cpp:307: void MadDecoder::flush(bool): Assertion `b->size() > sz' failed.
00:44<Ndiin>the opposite fails the assert, too.
00:44<Chutt>>=
00:44<Ndiin>ah, duh
00:44<Chutt>mdz, well, with the gist listings
00:44<Ndiin>...it's late ;)
00:44<Chutt>they use the tmsid as the channel identifier internally
00:44<mdz>it looked like the gist listings did the same as everyone else; make something up .gist.com
00:44<mdz>ah
00:45<Chutt>that's the same id everyone uses, since, well, it's by tms =)
00:45<Chutt>well, the beginning of it is
00:45<Chutt>there's a tv_grab_na_gist now
00:45<Chutt>in xmltv cvs
00:45<Ndiin>Well, I got yet another bt..
00:45<mdz>heh
00:46<mdz>I bet that one is a fair bit shorter than, say, tv_grab_na
00:46<Chutt>once they fix their timezone issue
00:46<Ndiin>(gdb) frame 3
00:46<Ndiin>#3 0x080730d6 in MadDecoder::deinit() (this=0x83ed088) at maddecoder.cpp:131
00:46<Ndiin>131 mad_stream_finish(&stream);
00:46<mdz>or ZapListings that is
00:46<Chutt>i'll fix up mythtv to support it
00:46<Ndiin>stream appears valid.
00:46<mdz>that will be so nice
00:46<Chutt>and add support for actors and stuff =)
00:47<thor>Nddin, what's above #3
00:47<mdz>do they have parseable pages for the provider selection and channel lineup too?
00:47<Chutt>i'm going to try to see if i can't wrangle a free subscription out of em for a link and a suggestion that others sign up
00:47<Ndiin>thor:
00:47<Ndiin>#0 0x40100f2d in pthread_mutex_lock () from /lib/libpthread.so.0
00:47<Ndiin>#1 0x40c5f4fe in free () from /lib/libc.so.6
00:47<Ndiin>#2 0x4014cb01 in mad_stream_finish () from /usr/lib/libmad.so.0
00:47<Chutt>mdz, i dunno, but if it's just as easy as going up to their site and setting it up, first
00:47<mdz>oh, right, it looks it up based on your userid
00:47<Chutt>the ui in setup can then just ask for your userid
00:47<mdz>yep
00:48<mdz>not quite as nice as being able to select it all in setup though
00:48<Chutt>no, but i'm willing to make some sacrificies for a stable listings source and the additional data
00:48<mdz>but well worth the tradeoff
00:48<mdz>yep
00:48<thor>Ndiin, this time it's a mutex?
00:48<Ndiin>This is a painful bug.. its reproducable, but with slightly different backtraces each time, heh
00:48<Ndiin>apparently.
00:48<Chutt>maybe it's mad
00:48<Chutt>that's causing problems
00:49<thor>Ndiin, no, craps out on oggs
00:49<thor>Chutt
00:49<thor>and mp3's
00:49<Chutt>ah
00:49<Chutt>well, i've never been able to reproduce it
00:49<Chutt>and i don't have any mp3s, so..
00:49<Chutt>and valgrind's giving things a pretty clean bill of health
00:49<Chutt>when i switch songs rapidly
00:50<thor>it's something very subtle, I'm sure
00:50<Ndiin>Chutt: I couldn't get valgrind to work with that env var either, only way I could was swapping out the gl libs
00:50<Chutt>ndiin, it works for me
00:51<thor>but it's definitely reproducible on at least two machines (mine and Ndiin's :-) )
00:51<Chutt>yup
00:51<Chutt>morph, working now, i hope?
00:52<thor>Ndiin, I need to go to sleep ... if you come up with any more leads, send me an email mythtv@lamedomainname.com
00:52<Morph>one sec.
00:53<Ndiin>will do thor
00:55<Ndiin>hmm
00:55<Ndiin>running through valgrind..
00:55<Ndiin>==27672== Use of uninitialised value of size 4
00:55<Ndiin>==27672== at 0x4032FBBE: III_huffdecode (in /usr/lib/libmad.so.0.1.0)
00:55<Ndiin>attached to gdb now..
00:55<Chutt>eh
00:55<Chutt>not necessarily bad
00:55<Morph>Chutt: No, floating point error now. got a back trace.
00:56<Chutt>you're more looking for doubled deletes, bad deletes, etc
00:56<Chutt>overwriting buffers
00:56<Chutt>morph, ok
00:56<Ndiin>Chutt: just kind of odd that it only happened a few times after like 20 next()s
00:57<Chutt>true, but still not necessarily bad =)
00:57<Chutt>without knowing what it's doing, i can't say for sure, of course
00:57<Morph>Chutt: http://franken.sclug.org:8888/8
00:58<Chutt>still in gdb?
00:58<Ndiin>bing bing, got an overwrite
00:58<Chutt>morph, 'print audio_sample_size'
00:58<Chutt>and 'print audio_sampling_rate'
00:59<Chutt>actually, wait a sec
00:59<Chutt>that's totally broken
00:59<Morph>k
01:00<Chutt>sorry :(
01:00<Morph>no worries.
01:01<Morph>newly support capture card, weird CPU, understable :)
01:01<Chutt>no, that code's really broken =)
01:01<Chutt>curious why it worked for me
01:02<Morph>Oh. heh. well then Im glad I helped find it :)
01:02<Ndiin>Chutt: I think it may or may not be unrelated, but take a look: http://members.cox.net/~ndiin1/overwrite
01:03-!-inman [] has quit ["Client exiting"]
01:03<Chutt>ndiin, that could very well be it
01:05<Morph>Chutt: did you need anymore info from this gdb session?
01:06<Chutt>morph, no
01:06<Chutt>i'll have it fixed soon
01:06<Chutt>couple minutes
01:06<Ndiin>the buffer though, only gets a copy of output_buf in its data section
01:06<Morph>okay cool.
01:06<Chutt>ndiin, yeah, but overwriting ram is bad =)
01:07<Ndiin>granted ;)
01:08<Chutt>that's going to munge up the value of output_bytes, for one thing
01:08<Chutt>then that'll mess up what it's writing to the Buffer buffers
01:08<Chutt>etc
01:09<mdz>Chutt: do you know if v4l2 can be built as modules, or does it require a patch?
01:10<Chutt>requires a patch, as far as i know
01:10<mdz>that is, modules without modification to common code
01:10<mdz>hmm, no kernel-patch-v4l2
01:10<Chutt>though
01:10<Ndiin>hmm
01:10<Chutt>the only code it touches is in the drivers/media/video/ dir
01:10<Ndiin>trying to think of the best way to test this
01:12-!-_bDove [ipo@p508145DB.dip0.t-ipconnect.de] has joined #mythtv
01:22<Chutt>morph, it should (hopefully) work now
01:22<Morph>cool.
01:22* Morphupdates
01:23<Chutt>you may want to recompile without debugging at some point =)
01:23<Chutt>debugging mode combined with the slow cpu is going to be, well, painful
01:23<Ndiin>Chutt: btw, just a few problems with that mythmusic setup patch update I posted; spaces close the pages, even on LineEdits, and a type in music_settings.xml (settings_scan should be music_scan)
01:23<Morph>yeah..
01:23<Chutt>ndiin, i'll take a look at it later
01:23<Chutt>thanks, though
01:23<Chutt>how is the vis setup?
01:24<Ndiin>basically just a lineedit to type in, like the original had for treelevels.. I was lazy and didnt feel like mucking with the way the original had it
01:24<Ndiin>I may go back and do it properly if I get a chance
01:25<Chutt>is there a list visible anywhere?
01:25<Ndiin>It lists in the help
01:25<Chutt>i think a bunch of checkboxes would work
01:25<Chutt>and a toggle for random or something
01:25<Chutt>i dunno
01:25<Ndiin>probably
01:27<Chutt>i hope someone fixes the audio in the ivtv driver soon
01:27<Chutt>this whine is so very annoying :/
01:33<Morph>You're still getting that?
01:33<Chutt>yup
01:33<Chutt>unless i load the old driver first, then don't load the msp3400 module
01:33<Chutt>but i need the msp module loaded for my other card (standard wintv-radio), so...
01:33<Morph>it happens all the time or just when watching live tv?
01:33<Chutt>it's not really _that_ bad of a whine, but it's headache forming
01:33<Chutt>all the time
01:34<Chutt>well, when it's playing back a recording
01:34<Morph>weird...I dont get that..
01:34<Chutt>live or pre-recorded
01:34<Chutt>there's two or three channels it doesn't happen on
01:34<Chutt>i think it may be a stereo/mono issue
01:34<Morph>even when I played that short clip you sent to the ivtv list.
01:34<Chutt>morph, you must either have bad ears or speakers =)
01:34<Chutt>it's a high-pitched whine
01:35<Morph>heh
01:45<Chutt>anyway, i'm off to bed
01:46<Chutt>morph, thanks for helping to point out my broken code =)
01:46<Morph>suer thing :) happy to help
01:48<Chutt>lemme know if it works/doesn't
01:48<Chutt>hopefully it will
01:48<Morph>will do.
01:48<Morph>just finished buiding actually.
01:49<Ndiin>Chutt: hmm, one problem with the checkbox idea.. you can't really order them that way.. or does it choose randomly from the list?
01:49<Chutt>it'd be random
01:49<Chutt>if you chose more than one
01:49<Ndiin>hm, ok
01:49<Chutt>have like an 'all' toggle or something
01:49<Chutt>i dunno
01:50<Ndiin>All should be ~= Random, no?
01:50<Chutt>yup
01:50<Morph>hrm.
01:52<Morph>Chutt: still failing.
01:52<Morph>wait.
01:53<Morph>maybe not.
02:04-!-hfb [] has quit ["Huh! Wasn't me!"]
03:12-!-PeteCool [] has quit ["Client exiting"]
04:04-!-choenig [~choenig@p50859859.dip.t-dialin.net] has joined #mythtv
04:59-!-PhracturedBlue [] has quit [calvino.freenode.net irc.freenode.net]
04:59-!-sc00p [] has quit [calvino.freenode.net irc.freenode.net]
04:59-!-thor [] has quit [calvino.freenode.net irc.freenode.net]
04:59-!-Captain_Murdoch [] has quit [calvino.freenode.net irc.freenode.net]
05:01-!-PhracturedBlue [~tcpaiqgvn@cs6668127-96.austin.rr.com] has joined #mythtv
05:01-!-sc00p [~oldendic@129.252.132.142] has joined #mythtv
05:01-!-Captain_Murdoch [~buy_more_@ip68-107-147-80.hr.hr.cox.net] has joined #mythtv
05:01-!-thor [~thor@208.185.11.34] has joined #mythtv
05:25-!-thor_ [~thor@208.185.11.34] has joined #mythtv
05:27-!-thor [] has quit [Read error: 110 (Connection timed out)]
05:29-!-PBlue_away [~tcpaiqgvn@cs6668127-96.austin.rr.com] has joined #mythtv
05:29-!-PhracturedBlue [] has quit [Read error: 111 (Connection refused)]
08:24<nziarek>i saw people talking about Gist/XMLTV above; did you see that XMLTV has a link on Gist's home page?
09:26-!-hurdel [~greg@wnpgmb08dc1-res-104-154.mts.net] has joined #mythtv
09:28-!-nziarek_ [nziarek@pcp010412pcs.unplugged.mu.edu] has joined #MythTV
09:50<nziarek_>wow, those gist XML listings are awesome
10:24-!-PBlue_away [] has quit [Read error: 110 (Connection timed out)]
10:29-!-Captain_Murdoch_ [~cpinkham@linoleum.engrs.infi.net] has joined #mythtv
10:29<Chutt>nziarek, yeah, once they fix up their timezone issue
10:29<Chutt>i'll work on getting their stuff all supported nicely
11:03<rkulagow_>chutt, are you here?
11:04-!-hfb [~hbarrett@lsanca1-ar2-4-60-005-210.lsanca1.dsl-verizon.net] has joined #mythtv
11:16<nziarek_>Chutt: what timezone issue?
11:18<nziarek_>Chutt: think you'll still go through XMLTV, or no, since you can just download that XML page?
11:21<rkulagow_>nziarek: zap2it doesn't send a timezone with the program information, so there have been issues during the daylight savings time crossover when it's difficult to determine what time a program is actually going to be scheduled for.
11:21<rkulagow_>i believe that xmltv tries to make guesses and workarounds, but sometimes it doesn't work.
11:22<nziarek_>rkulagow: ah...but, gist.com just sends all of the times in GMT (says 20030414110000 GMT). I guess the DST thing could still apply, though
11:23<rkulagow_>well, if they actually do GMT correctly, then it's actually a lot better than the data that zap2it sends.
11:25<nziarek_>nziarek: plus it has better descriptions, actor lists, directors, if audio is stereo, the aspect ratio, etc
11:25<nziarek_>um, i didn't mean that to me :)
11:25<rkulagow_>nziarek and nziarek_: yeah, that's what i assumed. :)
11:30<mdz_>nziarek: the timezone issue is that when you ask for a certain day's data, you get that day in GMT
11:35<_bDove>hello
11:36<_bDove>im having some problems... QGDict::hashKeyString: Invalid null key , Connection timed out. while starting... using gentoo
12:14<Chutt>the descriptions are the same as w/ zap2it
12:14<Chutt>same data
12:15<Chutt>nziarek, main issue is if you ask for a date, you get that date in GMT
12:15<Chutt>not the date in the local timezone
12:15<nziarek_>chutt: well, it seems more complete :)
12:15<nziarek_>chutt: so, why not just add that offset?
12:16<Chutt>because i'll need to grab 2 days to get one days worth of data
12:16<nziarek_>oh...good point
12:16<Chutt>i've already asked if they can fix it, and it's on their todo list
12:16<nziarek_>wow...it is amazing that little things I would hav never thought of
12:17<Chutt>they don't have channel icons, though
12:17<nziarek_>i did see that. not a big loss, to me.
12:18<nziarek_>zap2it's were too low quality to be of much use
12:18<Chutt>this is the same data zap2it provides
12:19<nziarek_>no, i meant the icons...did zap2it provide the actor names, etc?
12:19<Chutt>yup
12:19<Chutt>it's just not on the pages that xmltv scrapes
12:19<Chutt>zap2it == tribune media services, which provides data to everyone
12:19<Chutt>including gist
12:19<nziarek_>ah, so really, the gist thing will just make it much quicker
12:20<Chutt>yup
12:20<Chutt>one thing to download
12:20<Chutt>they're going to make it a pay service, though
12:20<Chutt>same price as their gold subscription
12:20<Chutt>well, for all the data
12:20<Chutt>there's going to be a free version with less data
12:21<nziarek_>Chutt: yeah, i bought the gold package
12:27-!-wigginjs [] has quit ["Client exiting"]
12:46-!-StarHeart [edgan@64-42-21-228.atgi.net] has joined #mythtv
12:54<rkulagow_>chutt: are you here?
12:54<Chutt>yup
12:55<moegreen>Chutt: how much is that subscription? (for the guide data)
12:55<rkulagow_>i've got a situation where the frontend / backend have lost "sync" with each other. backend seems to think the frontend is still there, but it's not. when the frontend tries to reconnect, "nothing happens", even though the frontend is trying to connect.
12:55<Chutt>like a buck a month
12:56<rkulagow_>i've got an strace of the frontend.
12:56<Chutt>hrm
12:56<thor_>moegreen 3.95 every 3 months
12:56<rkulagow_>gettimeofday({1050326777, 636044}, NULL) = 0
12:56<rkulagow_>select(11, [3 7 10], [8], [], {0, 0}) = 2 (in [7], out [8], left {0, 0})
12:56<rkulagow_>read(7, "\0", 1) = 1
12:56<rkulagow_>connect(8, {sa_family=AF_INET, sin_port=htons(6543), sin_addr=inet_addr("0.0.0.0")}, 16) = 0
12:56<rkulagow_>getsockname(8, {sa_family=AF_INET, sin_port=htons(34391), sin_addr=inet_addr("192.168.10.102")}, [16]) = 0
12:56<rkulagow_>getpeername(8, {sa_family=AF_INET, sin_port=htons(6543), sin_addr=inet_addr("192.168.10.101")}, [16]) = 0
12:56<rkulagow_>gettimeofday({1050326777, 639952}, NULL) = 0
12:56<rkulagow_>gettimeofday({1050326777, 640140}, NULL) = 0
12:56<rkulagow_>gettimeofday({1050326777, 640204}, NULL) = 0
12:56<rkulagow_>ioctl(3, FIONREAD, [0]) = 0
12:56<rkulagow_>gettimeofday({1050326777, 640350}, NULL) = 0
12:56<rkulagow_>select(11, [3 7 8 10], [], [], {0, 0}) = 0 (Timeout)
12:56<Chutt>rkulagow, the frontend really won't help
12:56<rkulagow_>gettimeofday({1050326777, 640603}, NULL) = 0
12:56<rkulagow_>nanosleep({0, 50000}, NULL) = 0
12:56<rkulagow_>ioctl(7, FIONREAD, [0]) = 0
12:56<rkulagow_>write(9, "\0", 1) = 1
12:56<rkulagow_>brk(0) = 0x8474000
12:56<Chutt>if the backend's not responding properly
12:56<rkulagow_>brk(0x847d000) = 0x847d000
12:56<rkulagow_>write(8, "18 ANN Playback pvr 1", 26) = 26
12:56<rkulagow_>ioctl(7, FIONREAD, [1]) = 0
12:56<rkulagow_>select(9, [8], NULL, NULL, {0, 5000}) = 0 (Timeout)
12:57<rkulagow_>ioctl(8, FIONREAD, [0]) = 0
12:57<rkulagow_>ioctl(7, FIONREAD, [1]) = 0
12:57<rkulagow_>nanosleep({0, 50000}, NULL) = 0
12:57<rkulagow_>backend is in gdb, just sitting there. never get the "adding" message...
12:57<rkulagow_>backend is still doing something though, because there are multiple threads, but it's not writing anything to the video directory as far as i can tell
12:57<Chutt>can you do a bt full of all the threads in the backend
12:57<Chutt>and email me that?
12:57<rkulagow_>sure. ctrl-c then bt full, etc?
12:58<Chutt>ctl-c, then info threads
12:58<rkulagow_>rihgt,
12:58<Chutt>switch to each thread one by one, with 'thread X'
12:58<Chutt>then bt full for each
12:58<Chutt>thanks
12:59<rkulagow_>18 threads; 1-5, 61, 107-121
12:59<Chutt>yeah, which is why i asked for email =)
12:59<rkulagow_>ok, will do
13:09<Chutt>damn people using my cvs server and all my bandwidth
13:10<moegreen>heh
13:13<Chutt>moegreen, current cvs might be a little better at live-tv mode w/ the hardware encoding
13:13<Chutt>playback was buggy =)
13:14<Morph>Chutt: morning.
13:15<Morph>Chutt: playback is working on this board now...a bit jittery but thats the CPU :) thanks.
13:16<moegreen>Chutt: alright, I'll check it out a bit later. I should have this weather in cvs later today too
13:28<rkulagow_>chutt, mailed you the backtrace for what we discussed earlier.
13:29<Chutt>thanks
13:31<rkulagow_>btw, should i make any changes to the HOWTO WRT MPEG2 encoding?
13:31<Chutt>not yet
13:31<Chutt>maybe once it's a tad more mature =0
13:31<yebyen>is there any option to cvs update to display commit log messages as it updates?
13:32<rkulagow_>chutt: ok.
13:33<rkulagow_>chutt: any thoughts to built-in lirc rather than irxevent? i remember that (mythmusic?) had a patch submitted where lirc was enabled, and you took it out. i
13:33<rkulagow_>'m ready to get really really frustrated with focus issues and various WMs. i though fvwm was working right, but it looks like it might not be.
13:33<Chutt>mythvideo
13:34<Chutt>it wasn't done right, either =)
13:34<Chutt>i dunno
13:34<Chutt>i'd really rather not have to go to all that trouble
13:35<rkulagow_>ok; i'm going to keep messing with it. it seems like the remote will work for a while, and then just "stop". i thought with simpler WMs there'd be less of a problem, but even fvwm has bailed on me after a while. i had such high hopes. :)
13:36<Chutt>heh
13:40<yebyen>hrm, is mythtv.org down?
13:40<rkulagow_>do you want me to stay within gdb on the backend, or can i punt and reset the backend and the frontend?
13:42<Chutt>yeah, just reset it
13:42<Chutt>i think i know what's up, but not quite sure what to do about it
13:43<mdz_>rkulagow_: are you running any other processes in your session other than mythfrontend, your wm, and irxevent or whatever?
13:44<yebyen>odd, I can't get to mythtv.org from my desktop
13:44<yebyen>works from gripz.com shell though
13:45<rkulagow_>mdz: no, other than the usual system stuff.
13:50<yebyen>wow
13:50<yebyen>actually my firewall is b0rked
13:58-!-choenig [] has quit [Remote closed the connection]
14:00<rkulagow_>chutt: any thoughts on maybe moving README and FAQ to docs/ so that newbs that don't know any better will explore the filetree and maybe say, "hey, i'm getting a QMYSQL3 not found error. i wonder if there's documentation?"
14:01<yebyen>Chutt: is mythtv behind a firewall?
14:03<yebyen>like, in some way changed recently
14:04<rkulagow_>yebyen: i'm getting it just fine as of 10 seconds ago.
14:04<yebyen>rkulagow_: works fine from my firewall box, but anything behind it can't get to it
14:04<mdz_>rkulagow_: what is the usual system stuff?
14:05<mdz_>you do not need anything other than those programs I listed in your X session
14:05<mdz_>a GNOME/KDE panel or anything like that could be causing your problems
14:05<mdz_>I recommend trying it with the bare minimum of programs in your session
14:17<rkulagow_>mdz: i'm running mythfrontend on a dedicated frontend; login as mythtv, using blackbox; open two eterms, one has irxevent, the other has mythfrontend. blackbox focus policy is click to focus and "focus new windows". remote will work for a while, then seem to stop. keyboard continues to respond. wish (not a feature request) there was a debug mode that would write events to files with a timestamp so that i wouldn't have to try to reme
14:17<rkulagow_>i'm going to try again with fvwm and the fvwmrc that's in contribfiles
14:22<mdz_>seems like an lirc problem then
14:23<mdz_>try getting rid of the eterms and starting everything from your xsession or xinitrc
14:41-!-orangey [~orangey@dsl-207-112-62-143.tor.primus.ca] has joined #mythtv
14:41<orangey>hey guys! Is the current CVS broken?
14:45-!-Chang [~coffman@fw.pqbs.com] has joined #mythtv
14:50<rkulagow_>mdz: i'll give it a go.
15:06<Cloak>itvc15 support in cvs?
15:08<nziarek_>orangey: i DL'd it last night, and it was fine
15:08-!-orangey [] has quit [Remote closed the connection]
15:44<yebyen>blum
15:45<-- nziarek_has quit ()
16:17<rkulagow_>chutt: are you here?
16:17<Chutt>yup
16:20<rkulagow_>different issue: i have a 48 meg 1 hour MPEG4 program that was supposed to be "iron chef battle". obviously, based on the size, something bad happened. i tried to "watch recording" on the frontend, screen went black, then nothing happens. i've got the "static preview" option, and it just shows a big black rectangle. (first few times i thought that maybe the recording just happened to start right at the fade from black; later i check
16:21<rkulagow_>issue: is there some way to determine if a file is valid before trying to play it?
16:21<Chutt>hm
16:21<Chutt>not really
16:28<Chutt>like, not more than trying to play it, really
16:28<thor_>sort of Church-Turing-Godel thesis thing
16:30<rkulagow_>if i select the file in "watch recordings", screen goes to full screen black and stop responding to ESC. have to CTRL-ALT-F1, killall mythfrontend, then return to X. When I return, the eterm is full of "audio buffer overflow, audio data lost!" all the way back to the top of the scroll buffer.
16:33<moegreen>http://untzuntz.com/mythtv/weather/
16:33<moegreen>I'm committing this now...if the cvs server will respond this time :)
16:34<Chutt>oh, hold on
16:34<hurdel>does anybody else have connect failures when using mythweather? this is from about a month ago, but can never resolve the hostname, if i change the hostname to the ip address, then no probs
16:34<Chutt>there
16:35<moegreen>hurdel: rkulagow_ had a similar problem with one of his machines - not sure if it ever was resolved or if he just used the IP address.
16:35<hurdel>hmm, maybe some broken library for resolving hosts
16:37<moegreen>ok...let me know if it breaks for anyone
16:37<moegreen>of course I need to commit the xmlparse and uitypes stuff
16:39<hurdel>for redhat 9, is it still necessary to comment out the strip lines in the themes Makefile?
16:41<Tack>hurdel: I didn't have to.
16:42<Tack>hurdel: In fact I didn't even have any lines that began with "strip" in my themes/Makefile under RHL 9
16:44<hurdel>argh, now I'm getting the moc_dialogbox.cpp errors "This moc file was generated....It's too different to use this version of QT", somthing ilke that, brand new install of RH9
16:44<hurdel>thanks tack
16:49<hurdel>qtdir wasn't in the path, trying to make again
16:51<hurdel>well that sure didn't work
16:52<moegreen>hurdel: did you do a 'make distclean'
16:53<hurdel>no, i've never had a build complete tho,
16:53<hurdel>i'll try it tho
16:53<moegreen>hurdel: make distclean will remove all the moc_ files from your tree, then when you do a make the next time they will be regenerated with the new qt
16:53<hurdel>ok, makes sense, thanks
16:54<hurdel>i am 100% unaware of moc_ files and what they are/do
16:59<-- Chang(~coffman@fw.pqbs.com) has left #mythtv
17:06-!-nziarek_ [nziarek@pcp010412pcs.unplugged.mu.edu] has joined #MythTV
17:08<moegreen>nziarek_: I just committed the mythweather changes
17:08<nziarek_>moegreen: at this rate, I'll never get iulius.4 out the door :)
17:09<moegreen>well...hmm...put weather into 0.5...heh. The default theme looks pretty good w/ iulius - a little dark I suppose
17:11<nziarek_>i do want to get the new playback box out...so maybe weather will wait...along with the finder ;)
17:11<nziarek_>i've been slacking a little bit :)
17:15<moegreen>heh, well I'm onto the conflict resolution screen next
17:18<nziarek_>moegreen: does mythweather allow you to put whatever on whichever screen?
17:19<moegreen>nziarek_: yeah, well I don't have the pictures on each page yet - but all the weather data fields can go on any page. If you look at the weather-ui.xml file, the 'updatetime' field is like that
17:20<moegreen>nziarek_: hey I like the ui with the updated fonts -- looks really nice
17:21<nziarek_>i think you might be the first :)
17:21<nziarek_>i don't think anyone else likes those fonts.
17:21<nziarek_>i just wanted something i could distribute with the theme, were free, and looked nice
17:21<nziarek_>but, thanks ;)
17:22<moegreen>Chutt: playback/livetv is better now, I still get some studdering at 640x480, but that could be a host of other things
17:23<Chutt>but, overall it's a little better?
17:23<moegreen>yeah
17:25<moegreen>I am hearing that high pitched sound that you were complaining about
17:26<Chutt>yay
17:26<Chutt>someone else =)
17:27<moegreen>Chutt: did you try out the updated mythweather? There are a few more minor changes I need to make, but I wanted to get it out of my tree
17:27<Chutt>not yet
17:28<Chutt>been busy with work
17:51<-- Captain_Murdoch_has quit ()
17:58<Chutt>heh
17:58<Chutt>the gist guy's silly
17:58<Chutt>he _almost_ fixed the timezone issue
17:58<Chutt>now, instead of getting 8pm to 8pm
17:59<Chutt>i get 4am to 4am =)
17:59-!-zZzToxic is now known as NonToxic
18:00<mdz_>it should let you just ask for listings from one date/time GMT to another date/time GMT
18:00<mdz_>and forget deciding on what a particular "day" is
18:01<Chutt>ooh
18:01<Chutt>wait
18:01<Chutt>he did fix it
18:01<Chutt>it just still gives the time in gmt
18:01<Chutt>GMT-0400, so, 4 am is midnight
18:02<mdz_>that's reasonable
18:02<Chutt>his previous fix had the times in local time
18:02<mdz_>still silly to have the server side decide when the day starts and ends
18:02<Chutt>yeah, but this works
18:02<Chutt>as long as a day is a day
18:02<Chutt>you give it a timezone offset now
18:02<Chutt>tz_offset=-0400
18:03<Chutt>i'm considering bypassing xmltv entirely for this, btw =)
18:03<Chutt>just have my own grabber stuff
18:04<nziarek_>Chutt: i think that'd be nice...gets rid of one thing to install at least ;)
18:05<moegreen>Chutt: is it just an XML file with show data?
18:05<Chutt>moegreen, yup
18:05<moegreen>nziarek_: true, and XMLTV has a bunch of perl module dependencies too
18:05<Chutt>so no perl needed
18:06<mdz_>you can still use all of the xmltv utilities (other than the grabbers) with it though
18:06<Chutt>yup
18:07<mdz_>it'd be nice to have a configurator with the same interface as tv_grab_na though
18:07<moegreen>i guess I should get a gist account, heh
18:07<mdz_>to manage the provider selection and channels
18:07<Chutt>moegreen, the free one is good enough for now
18:07<mdz_>channels, if ever you can limit the channels to fetch
18:08<mdz_>but it's so fast there's not much reason to
18:08<Chutt>mdz, right
18:08<Chutt>just downloading a file per day =)
18:08<mdz_>and a small one at that
18:08<mdz_>I wonder what percentage of tv_grab_na's traffic is overhead
18:08<nziarek_>how big is each file? i thought mine looked around 1.24MB or so
18:08<Chutt>mdz, i liked my answer to that register guy better
18:08<mdz_>those pages must be huge
18:09<mdz_>nziarek: a few hundred k with gzip
18:09<nziarek_>i was just using IE
18:09<nziarek_>and then saving as text
18:09<Chutt>nziarek, it should be like a tenth that or so
18:09<mdz_>Chutt: you have no room to talk, using a prerelease compiler yourself :-)
18:09<Chutt>it's automatically compressed/decompressed
18:09<Chutt>mdz, yeah, but it's not _that_ prerelease
18:09<mdz_>I started to tell him to go bug suse about their compiler
18:10<mdz_>and I felt a rush of good will
18:10<Chutt>like, 3.2.3 isn't all that much different than 3.2.2
18:10<Chutt>not like 3.3 =)
18:10<mdz_>that assert is probably a real mythtv bug
18:10<Chutt>yup
18:11<Chutt>but, i don't care all that much about 3.3 right now
18:12<mdz_>maybe -fPIC used to imply -DPIC and doesn't anymore
18:12<mdz_>that will probably break half the inline assembler out there
18:12<Chutt>does that dir compile with -fPIC for you?
18:12<mdz_>not by default
18:12<Chutt>i didn't think it did at all
18:12<mdz_>it probably should
18:12<Chutt>since it's compiling as a static lib
18:12<mdz_>oh
18:12<Chutt>it doesn't need it
18:12<mdz_>but it's linking into a shared lib
18:12<Chutt>no it's not
18:12<mdz_>it isn't?
18:12<Chutt>libmythtv is static
18:13<Chutt>yes, i'm bloating my binaries a little =)
18:13<Chutt>moegreen, Map Unavailable
18:13<mdz_>hmm, so it is
18:15<mdz_>he must have run configure with --enable-shared or something then
18:15<Chutt>i thought i disabled that
18:15<Chutt>oh well
18:15<moegreen>Chutt: what's you location? It might have just failed that once - try and update again. It stores as /tmp/weather.jpg so you can see if that file is there
18:15<mdz_>I just tried it
18:16<mdz_>and it seems to add -fPIC
18:16<mdz_>no idea if it actually builds a shared lib or not
18:16<Chutt>moegreen, 'sec, someone's using my cvs server =)
18:16<Chutt>USOH1054
18:17<moegreen>hmm...the city name might be easier
18:17<Chutt>willoughby, ohio
18:17<moegreen>i've got a map here (clear skies)
18:17<Chutt>what can i do to debug, then?
18:18<Chutt>weather.jpg is 0 bytes
18:18<moegreen>well, if you've tried to update again and it's still 0 bytes, changing the 'debug = false' to a true in the weather.cpp file will print out a lot of info about what it's downloading, etc.
18:18<Chutt>ok
18:19<Chutt>updated 3 or 4 times =)
18:19<moegreen>heh
18:19<moegreen>Chutt: you could also try and up your aggressiveness level
18:19<moegreen>that will give it more time to get the map
18:20<Chutt>it says it read a bunch of data
18:20<Chutt>it got the url and everything
18:21<moegreen>when it was reading data did it display a bunch of '!'s
18:21<Chutt>yup
18:21<Chutt>!!!! once
18:21<Chutt>then a dozen lines of !
18:22<moegreen>what does your [image size = 31947] say?
18:22<Chutt>27172
18:22<moegreen>does it reach 'MythWeather: COMMS : Socket is closed.'
18:22<Chutt>yup
18:23<moegreen>hmm...
18:24<moegreen>well...that means it got the 27172 bytes of data for the image
18:32<-- nziarek_(nziarek@pcp010412pcs.unplugged.mu.edu) has left #MythTV
18:32-!-nziarek_ [nziarek@pcp010412pcs.unplugged.mu.edu] has joined #MythTV
18:35<Chutt>moegreen, if the connection times out for the initial weather, it displays a blank screen until a key is hit
18:35<Chutt>then it displays the 'timeout limit' stuff
18:35<Chutt>and, i've fixed the map stuff
18:36<Chutt>made it use a QFile instead of an ofstream
18:36<Chutt>and that's seemed to have worked
18:37<Chutt>just checked that in
18:37<moegreen>alright, cool - I'll look at the timeout stuff in a little bit
18:37<Chutt>no rush
18:37<Chutt>there was just someone doing a checkout right then, so i noticed =)
18:37<Chutt>doesn't happen all that often
18:39<nziarek_>has anyone ever used Google Sets?
18:41<nziarek_>i was just looking at it, and it takes a set of words (so banana, pear and lime) and then gives you a list of wrods that relate to all three (so apple, grape and watermelon)
18:41<nziarek_>and it works with TV shows, too, so far as I can tell
18:42<nziarek_>i just typed in a few show names and hit it and all the shows that came up I like...
18:42<nziarek_>i remember someone saying that they wanted a feature like this in Myth, maybe this is something to look at, if that person is here
18:49* NonToxicyawns
18:51<Cloak>chutt, you get the wintv 250 yet?
18:58-!-PBlue_away [~tcpaiqgvn@cs6668127-96.austin.rr.com] has joined #mythtv
18:59<Tack>Anyone here use a Geforce 2 GTS with nvtv?
19:04-!-hfb [] has quit ["Huh! Wasn't me!"]
19:23-!-Tack [] has quit [Remote closed the connection]
19:26-!-Tack [~tack@h24-70-129-87.su.shawcable.net] has joined #mythtv
19:26-!-choenig [~choenig@p50859859.dip.t-dialin.net] has joined #mythtv
19:48<Tack>What video cards do people use for tv-out here?
19:48<moegreen>Tack: Matrox G400
19:49<Tack>Can you do overscanning (or do you need to) with the G400?
19:51<moegreen>didn't need to
19:51<moegreen>it's a nice bright picture, clean and sharp too. It works with the kernel framebuffer (it's accelerrated)
19:53<Tack>Hmm. Well I'm not too happy with the output from my nvidia card.
19:53<Tack>mythtv works okay with the framebuffer?
19:53<moegreen>Tack: yeah
19:54<Tack>Another option is to try one of those VGA to NTSC converters. Might be cheaper than buying a new video card.
19:55<moegreen>Tack: that's an option - with that if you want to upgrade your video card to something else you don't need to worry about it TV out and what not
19:55<Tack>I'm just wondering what will yield the best quality tv-out. What tuner card do you use?
19:57<moegreen>Tack: i've got an ati tv wonder ve - it's tuner isn't very good. I've also got a wintv pvr250
19:57<Tack>I wonder how the newer tv cards compare to my WinTV PCI, since I'm not excessively happy with that quality either. :)
19:58<Tack>Basically, if I can tell the difference between myth output and my original tv signal, less a little mpeg4 artifacting, I won't be happy. :)
19:59<moegreen>Well, I don't really have a computer that is fast enough to record at a high enough resolution. On certain channels the ATI tv wonder VE is great (minus it being a mono card).
20:00<moegreen>If you commit enough CPU power and bitrates/resolution - you should be able to reach your goal
20:02<Tack>I'm capturing at 640x480, mpeg4, with 2800kbit, and I'm not happy. Even when I view it on the monitor (ignoring tv-out, then), frame-rate isn't smooth (I have 30% CPU left usually), and the colors seem washed. So I suspect too the quality of my hardware is an issue.
20:03<Tack>So I'm particularly interested in the wintv pvr-250 that Chutt is working with.
20:05<moegreen>Tack: have you tried increasing the color on the input? You can do that per channel. Other than that, there a several issues with using a computer with standard tuners and outputs to reproduce a perfect NTSC signal
20:06<Tack>I've played with increasing the color using the color control within myth. It helps a little bit. What sorts of things can one do to reproduce a perfect NTSC signal?
20:08<Timon>Start with some color bars and an analyzer :-)
20:08<Timon>Thats what the TV stations do
20:09<Tack>I don't suppose there's a HOWTO on all this? :)
20:09<Timon>Umm, not that I know of. If you had an analyzer, you would know how to :-)
20:09<Tack>Yeah. I don't, and I don't. :)
20:09<Timon>The analyzer isn't a cheap piece of equipment
20:10-!-StarHeart [] has quit ["Client exiting"]
20:19-!-Ripp [] has quit [""Bring me the hydrospanners! I don't know how we're gonna get out of this one.""]
20:20-!-decaff [~kevin@clt74-76-015.carolina.rr.com] has joined #mythtv
20:39<mdz>Chutt: I've started watching it, and the frontend is getting a little chunky in memory usage
20:39<mdz>may be leaking somewhere
20:43-!-choenig [] has quit [Remote closed the connection]
20:45<nziarek>Tack: have you changed the card type in modules.conf?
20:45<Tack>nziarek: No, should I?
20:46<nziarek>Tack: what distro?
20:46<Tack>RHL 9.
20:47<nziarek>hmmm...then I don't know. on mandrae 9.1 once i did that, the colros went from washed out to perfect.
20:47<nziarek>i am trying to find what you need to add now
20:47<Tack>That would be most excellent. :)
20:47<moegreen>Tack: If xawtv looks bad (i.e. not like your actual cable signal) then your card or tuner might not be set right - which is easily fixed by adding the type= and tuner= arguments
20:48<Tack>xawtv is fine. My feed comes in via Composite1.
20:48<moegreen>Tack: if you do a 'dmesg' you should see information about which card and tuner were autodetected
20:49<Tack>Yes, and bttv does detect the correct tuner card. (WinTV PCI)
20:49<Tack>Things work. They work adequately well, it's just not the level of quality I'm happy with. :)
20:52<nziarek>well, that all said, I can't find it, and my MythBox isn't currently accessible
20:52<nziarek>but, if everything else looks good, then that probably wasn't it...my XAWTV looked awful until I added the tuner=17 to that modules.conf too
20:52<Tack>nziarek: No problem. I'll bug you some other time when you can access your myth box. :)
20:52<nziarek>:) Deal
20:52<nziarek>it is in the archives, that is where I got my answer
20:53<nziarek>i just can't seem to find it again
20:53-!-orangey [~orangey@dsl-207-112-62-143.tor.primus.ca] has joined #mythtv
20:53<Tack>myth-users?
20:53<nziarek>i think it was before that, so myth-dev (off topic, i suppose :)
20:53<Tack>I'm using bttv 0.7.96. Do you suppose the later bttv drivers might improve quality?
20:54<Tack>I'll scan the archives.
20:54<nziarek>you are asking the wrong guy :-)
20:54<nziarek>good luck
20:54<Tack>Do you know about when it was?
20:55<nziarek>mmmm....before the new year...maybe november
20:56<Tack>Okie.
21:01<rkulagow_>chutt: are you here?
21:03<Tack>Ugh, mp3 audio sounds utterly awful with prologic. :)
21:03<Tack>(To be expected.)
21:12<Tack>Gah, buy.com sucks if you're not American.
21:20<Cloak>it sucks even if you are
21:29<Tack>nziarek: I found the email, but since I don't have the same tuner card as you the option didn't help (in fact it made it worse :))
21:29<nziarek>moegreen: the program finder seems awfully slow, is that due to the theming?
21:29<nziarek>Tack: i thought you had a VE...my bad
21:29<nziarek>didn'tmean to waste your time
21:29<Tack>No biggie.
21:31<moegreen>nziarek: isn't slow here. You might notice it being slow at the beginning because it is executing a bunch of sql statements to fill up the initial data
21:31<nziarek>it just seems slow to scroll through...from a 1GHZ remote frontend...
21:32<nziarek>not a big deal, i just remember the old one, when youdown, it would fly through the letters :)
21:32<moegreen>I'm assuming you've drawn put the transparent background to layer 0 in the background container
21:32<nziarek>when youdown=when you would hold down
21:32<nziarek>i've actually just started on it and have 0 graphics
21:32<nziarek>except the white borders
21:33<moegreen>hmm...well...i don't have my remote repeating keys but I can fly through the list pretty fast
21:33<nziarek>ok...like i said, not a big deal, and it wouldn't surprise me to be my problem :)
21:35<moegreen>nziarek: what resolution are you running at?
21:35<nziarek>1024x768
21:35<nziarek>that could be it
21:35<moegreen>that might also slow it down more than I would notice
21:35<nziarek>on my actual Mythbox only 800x600
21:36<moegreen>make it look pretty then we'll worry about speed :)
21:36<nziarek>deal :)
21:36<moegreen>nziarek: did you finish up the playback then?
21:37<nziarek>for now...i need to sit on it. i wanted to get the finder done and then put out .4
21:37<nziarek>then mythweather, then something better for delete, then a nicer OSD
21:37<nziarek>conflicts can go in there, too :)
21:37<moegreen>my delete recordings is kinda messed up - not sure if I have an older version and maybe because chutt fixed up the usage area not being drawn
21:37<nziarek>is it an iulius?
21:38<moegreen>yeah
21:38<nziarek>'cause on .3 i forgot to even look at it :)
21:38<moegreen>heh, that would be it then
21:38<moegreen>it looks to be using the blue/liquid areas (except for the video)
21:39<moegreen>heh, no problem
21:39<nziarek>i was thinking that the playback and delete shared more XM than they did...when I went to look at the program finder, i realized that they didn't
21:42<moegreen>well...if you were to plan for the usage bar when making the playback - then they would have pretty much the same stuff in the XML file, but ...
21:51-!-TheAsp [asp@CDR13-117.accesscable.net] has joined #mythtv
21:51<TheAsp>hmm
21:53<nziarek>moegreen: right, and once I realized what I did, i just copied over the delete stuff with the iulius playback stuff
21:53<TheAsp>Hmm
21:53<TheAsp>So, a local channel has HDTV in the description of some of the shows on it...
21:53<TheAsp>so some things are recording twice
21:53<TheAsp>as they broadcast them first...
21:55<TheAsp>ohh, more then one channel
21:56<TheAsp>took me a few weeks to figure that one out :P
22:16<Captain_Murdoch>kinda like some put "(live)" after the first showing of a live show's description but not later when they rerun it 3 hours later for west-coast people in the U.S.
22:17<-- decaff(~kevin@clt74-76-015.carolina.rr.com) has left #mythtv ("Client Exiting")
22:17-!-PBlue_away [] has quit [Read error: 111 (Connection refused)]
22:17-!-PBlue_away [~tcpaiqgvn@cs6668127-96.austin.rr.com] has joined #mythtv
22:22-!-PeteCool [~pete@modemcable019.237-201-24.mtl.mc.videotron.ca] has joined #mythtv
22:26<PeteCool>moegreen: this weather condition: http://www.msnbc.com/news/wea_front.asp?pos=&tab=oth&ta=y&czstr=Montreal%2C+Canada&action=&inst=n&accid=CAXX0301 doesn't show a picture in mythweather
22:26<PeteCool>moegreen: light rain with thunder (they forgot about the little ice balls there)
22:28<moegreen>PeteCool: thanks, it's in CVS if you want to update weathertypes.dat and copy it to /usr/local/share/mythtv/mythweather
22:30<PeteCool>moegreen: thanks
22:30<PeteCool>moegreen: that's a cool graphic :)
22:31-!-itsame [~joejoe2@rdu74-170-244.nc.rr.com] has joined #mythtv
22:32<itsame>Anyone know of a problem with Myth locking up
22:32<itsame>basically ..I was up and running
22:32<moegreen>PeteCool: i just updated the weather-ui.xml file to let the current condition line be two lines if it needs to be (like with Light Rain and Thundershowers)
22:32<itsame>now...when I goto the frontend screen...and hit Enter on "WatchTV"
22:33<itsame>it freezes ..and I have to kill/restart the app
22:33<moegreen>itsame: is the backend still running?
22:33<itsame>I can't watch tv no more :(
22:33<itsame>yea..it stays up
22:33<Ndiin>itsame: segfault, or just a lock?
22:33<itsame>just lock
22:33<moegreen>itsame: is there something else trying to use the soundcard?
22:33<itsame>ahh
22:33<itsame>nop
22:33<itsame>I do this right on boot
22:34<itsame>but I do notice I got sound
22:34<itsame>even when myth ain't up
22:34<itsame>if I plug my DSS reciever to the line in of my snd card
22:34<Ndiin>That just means you didn't turn your input line's volume down is all
22:35<itsame>It's annoying
22:35<itsame>I had it up and running
22:35<itsame>:(
22:35<itsame>I think it may have been this app I installed
22:35<itsame>tvtime
22:35<itsame>I backed it out
22:35<itsame>but dunno
22:35<PeteCool>moegreen: great work, once again
22:35-!-PeteCool [] has quit [calvino.freenode.net irc.freenode.net]
22:35-!-Tack [] has quit [calvino.freenode.net irc.freenode.net]
22:35-!-nziarek_ [] has quit [calvino.freenode.net irc.freenode.net]
22:35-!-Captain_Murdoch [] has quit [calvino.freenode.net irc.freenode.net]
22:35-!-sc00p [] has quit [calvino.freenode.net irc.freenode.net]
22:35-!-hurdel [] has quit [calvino.freenode.net irc.freenode.net]
22:35<moegreen>itsame: nah, tvtime won't interfere with mythtv. Does playback work?
22:36<itsame>well..to be honest
22:36<itsame>I never had a chance to record
22:36<itsame>I had live tv
22:36<itsame>up
22:36<itsame>and replay
22:36<itsame>and time shiftin
22:36-!-PeteCool [~pete@modemcable019.237-201-24.mtl.mc.videotron.ca] has joined #mythtv
22:36<itsame>but didn't get to test out the recording before it died
22:37<itsame>I'm thinking I may have to do the nasty
22:37<itsame>and reinstall Mandrake
22:37<itsame>pain in the but thou
22:37<PeteCool>Chutt: I think sometimes the video playback sometimes locks out user input...
22:37<itsame>mythweather still works
22:37<PeteCool>Chutt: with evilwm I can alt-tab to and from each window, eliminating focus problems
22:37<itsame>and the guide works
22:37<PeteCool>Chutt: and still it refused att input
22:38<PeteCool>att == all
22:38<moegreen>itsame: you shouldn't need to reinstall
22:39<itsame>I hope not
22:40<PeteCool>Chutt: it happens very rarely, but it's very blocking for the user
22:41<itsame>I tried to re-run the setup
22:41<itsame>and no go
22:41<itsame>xawtv works
22:41<itsame>tvtime works
22:41<itsame>just no myth
22:41<moegreen>itsame: does the backend say anything when you select Watch TV?
22:41<moegreen>like: Changing from None to WatchingLiveTV
22:42<itsame>is it better to run backend as the user
22:42<itsame>or root
22:42-!-Tack [~tack@h24-70-129-87.su.shawcable.net] has joined #mythtv
22:42-!-nziarek_ [nziarek@pcp010412pcs.unplugged.mu.edu] has joined #mythtv
22:42-!-hurdel [~greg@wnpgmb08dc1-res-104-154.mts.net] has joined #mythtv
22:42-!-sc00p [~oldendic@129.252.132.142] has joined #mythtv
22:42-!-Captain_Murdoch [~buy_more_@ip68-107-147-80.hr.hr.cox.net] has joined #mythtv
22:42<itsame>moegreen: lemme see
22:42<PeteCool>itsame: if your device permissions are alright root or user both work
22:42<moegreen>itsame: you should run it as the user. Check the file permissions on your ringbuffer file to make sure you can write to it
22:43<moegreen>itsame: if you ran it as root originally then as a user - it most likely isn't able to write to that file, which might make it freeze/hang
22:43<itsame>shoot...my backend didn't die yet
22:43<itsame>from the last frontend lockup
22:43<itsame>one sec
22:44<thor_>Ndiin, I got it
22:44<itsame>mand kill-9 won't even kill it
22:45<PeteCool>try kill -SIGHUP
22:45<itsame>[root@localhost root]# kill -SIGHUP 3006 3012
22:45<itsame>mythtv 3006 1 0 22:42 ? 00:00:00 mythbackend
22:45<itsame>mysql 3007 2332 0 22:42 ? 00:00:00 [mysqld]
22:45<itsame>mysql 3008 2332 0 22:42 ? 00:00:00 [mysqld]
22:45<itsame>mysql 3009 2332 0 22:42 ? 00:00:00 [mysqld]
22:45<itsame>mysql 3010 2332 0 22:42 ? 00:00:00 [mysqld]
22:46<itsame>mythtv 3012 3006 0 22:42 ? 00:00:00 [mythbackend] <defunct>
22:46<itsame>stubor
22:46<itsame>nd
22:46<PeteCool>killall mythbackend ?
22:46<itsame>nope
22:47<itsame>man... kill -9 should blow it away..unless Mandrake weird unix
22:47<PeteCool>that doesn't work?
22:47<itsame>nop
22:47<itsame>winblowz....reboot
22:47<moegreen>itsame: did you originally run the backend as root?
22:47<itsame>hehe
22:47<itsame>nah
22:47<itsame>as the "mythtv" user
22:47<itsame>then I logged out
22:47<itsame>and relogged in
22:47<itsame>i'm gonna reboot
22:48<itsame>crazzy!
22:48<itsame>This is a first for Mandrake and me
22:48<itsame>I'm a Slackware man myself
22:48-!-Ripp [~ripp@209.194.224.236] has joined #mythtv
22:49<itsame>All I know..is the few hours I had Myth up and running
22:49<itsame>it's pretty slick
22:49<itsame>made props to the dev's if they in here
22:49<itsame>made =mad
22:49<itsame>hehe
22:51<itsame>ok
22:51-!-NonToxic is now known as zZzToxic
22:51<itsame>it's back
22:52<itsame>WHOLY CRAP
22:52<itsame>it's working now
22:52<itsame>haha
22:52<itsame>i didn't do anything
22:52<itsame>I swear I tried rebooting before
22:53<itsame>oh well...live tv at least working again :0
22:55<Tack>mythvideo really needs some interface loving. :)
23:00<orangey>hey guys.. what's up with the current CVS? I'm getting v4l2 errors
23:00<orangey>and nobody so far has said anything on mythtv-dev from what I can tell
23:01<orangey>channel.cpp:55: `struct v4l2_capability' has no member named `capabilities'
23:01<orangey>that sort of dealie..
23:03<Ndiin>thor_: sweet =)
23:03<moegreen>orangey: are you getting errors or just warnings?
23:04<orangey>orangey: errors
23:04<orangey>oops.. moe: errors
23:04<Ndiin>thor_: Ah, it was continuing after half the decoder was torn down?
23:04<orangey>channel.cpp:109: `V4L2_STD_NTSC_M_JP' undeclared (first use this function)
23:04<orangey>make[2]: *** [channel.o] Error 1
23:04<itsame>is ther anyway to stop a recording once you started...without deleting it
23:04<orangey>it looks like it has to do with video 4 linux 2?
23:05<moegreen>orangey: which distro are you using?
23:05<orangey>mandrake 9.1
23:05<orangey>and I am installing a virgin copy on a system with mandrake 9.0
23:05<orangey>all in all, 2 x 9.1 and 1 x 9.0 have the same problem
23:06<moegreen>heh, from what I understand Mandrake has included a very old version of V4L2 in their distribution
23:06<moegreen>which is the source of your problems right now
23:06<orangey>moegreen: where can I get the newer version?
23:06<thor_>Ndiin, yes but now there's problem .... hang on, better fix coming
23:08<orangey>moegreen: is v4l2 a part of the kernel?
23:08<moegreen>orangey: not sure what the best solution is. You could check to see if mandrake has an update
23:08<moegreen>orangey: there is a patch for the 2.4 kernels and I think it is part of 2.5
23:09<orangey>moegreen: tres strange.
23:09<nziarek>moegreen: i think the help_info_1 and help_info_2 containers for the finder are using the area from the program_info container
23:09<nziarek>instead of their own
23:09<orangey>moegreen: are you saying I'm going to have to recompile my kernel?
23:09-!-Edgan [] has quit ["Client exiting"]
23:10<moegreen>orangey: not necessarily - you could try and edit the videodev.h file to comment out the videodev2.h file which it is including
23:10<moegreen>orangey: just keep a copy of the original in case you break it :)
23:11<orangey>moegreen: this is myth-side?
23:11<moegreen>nziarek: you think they should be seperate? The reason they are like that is because there is always going to be available space there (because there isn't any program info unless you are in the third list)
23:12<nziarek>no, together is fine, but you have an area tag in both the help_info containers...i was using them and not seeing anything...
23:12<nziarek>am i using them correctly?
23:12<moegreen>orangey: well...try adding this to your channel.cpp (near the top) #define HAVE_V4L2 0
23:13<moegreen>orangey: you'll get a redefinition error, but it should stop it from trying to use the newer v4l2 stuff
23:13<orangey>moegreen: which line? when I just tried it right after the includes, it didn't work out
23:14<moegreen>orangey: what did it say?
23:14<orangey>exact same error abou undeclared and whatnot
23:14<orangey>you want me to paste it?
23:14-!-rkulagow_ [] has quit [Read error: 54 (Connection reset by peer)]
23:14<moegreen>msg it to me (just a line or two, right?)
23:14-!-rkulagow_ [~rkulagow@12-207-131-29.client.attbi.com] has joined #mythtv
23:15<moegreen>nziarek: there is an area tag in each, because only one is shown at a time
23:15<nziarek>hmmm....ok, well, it looks fine now, so whatever I am doing is fine :-)
23:16<moegreen>nziarek: actually - make sure they are on the drawlayer="6"
23:16<nziarek>draworder?
23:17<moegreen>errr, yeah :)
23:17<nziarek>just making sure - not being too much of a smart a$$
23:19<itsame>Is it better to plug into the Line IN sound on the TV card or Line in on your sound card...I'm getting this bad echoing problem now
23:20<rkulagow_>plug what? speakers?
23:20<itsame>oh
23:20<itsame>sorry..my external tuner
23:20<itsame>like I"m using th s-video in
23:20<itsame>I got a winTV Dbx 401
23:21<moegreen>itsame: you need to mute your line in volume and set it to capture
23:21<Chutt>orangey, still here?
23:21<rkulagow_>what /dev/dsp are you using - the one to your soundcard, or the one to your tv card (if it's creating a /dev/dsp device)
23:21<itsame>ahh I have /dev/dsp
23:21<Chutt>orangey, you've got a file on your system -- /usr/include/linux/videodev2.h
23:21<moegreen>Chutt: i got him to comment out a few lines in the videodev.h file - it seemed to have worked
23:21<Chutt>orangey, could you email that to me?
23:22<Chutt>moegreen, yeah, i want to add another ifdef so other people using that version of mandrake or whatever it is don't have the same problem
23:22<moegreen>Chutt: he just said he'll be right back, he's doing that on his other machines :)
23:22<rkulagow_>itsame: so assuming that your /dev/dsp is on your soundcard, use that.
23:22<Chutt>itsame, your hang was (most likely) something sitting on /dev/dsp
23:22<Chutt>like artsd or esd or some other program
23:22<itsame>yea
23:22<rkulagow_>itsame: did you read the HOWTO?
23:22<itsame>I'm using artsd
23:22<itsame>yea
23:23<itsame>thats how I got this far ;)
23:23<rkulagow_>then keep reading, because your sound issue is addressed in the HOWTO.
23:23<Chutt>you generally can't have artsd running at the same time
23:23<itsame>I'm looking at it right now...and the email
23:23<itsame>I saw redhat stuff
23:23<Chutt>unless you're using one of the few cards that do multiple playback streams in hardware
23:23-!-hfb [~hbarrett@pool0658.cvx11-bradley.dialup.earthlink.net] has joined #mythtv
23:23<Cloak>Chutt: you get the 250 wintv yet?
23:24<Chutt>cloak, err, read the mailing list.
23:25<Cloak>oh, good point :)
23:26<rkulagow_>chutt: setup in mythfrontend still has the mouse pointer show up; don't see it in any other screens / modules. is this an issue?
23:26<Chutt>it's intentional =)
23:26<orangey>Chutt: i'm here.
23:26<orangey>Chutt: Moegreen walked me through it.
23:26<itsame>rkulagow_: I actually read that part
23:26<itsame>and for Mandrak 9.1
23:26<Chutt>orangey, i still want you to send me that file
23:26<orangey>Chutt: Email coming
23:26<itsame>it don't have a red box under my Linx in
23:27<Morph>hmm..is MythTV known to not work well with the low-latency patches? :)
23:27<orangey>unless you would rather a /dcc
23:27<Chutt>morph, works better without?
23:27<Chutt>orangey, no, email.
23:27<itsame>I'm trying to look for stuff ON my box now...a Shuttle SK41g
23:27<Tack>Whoever hacks on mythmusic here, I have a request: when selecting music, allow the enter key to check the box.
23:27<orangey>Chutt: sent
23:27<Chutt>tack, it's something i've been meaning to do, but it's slightly non-trivial
23:28<Morph>Chutt: I just built a kernel with a couple low-latency related patches and mythtv is hanging when trying to watch live tv and erroring out with a floating point error when watching recorded tv.
23:28<rkulagow_>itsame: section 18.11 doesn't mention mandrake at all. are you sure you read the HOWTO?
23:28<Tack>Chutt: Okay, as long as it's on your mind. :)
23:28<Chutt>i've never used the low latency stuff, so, dunno
23:28<itsame>7.1
23:28<itsame>I wuz looking at 7.1
23:29-!-Edgan [edgan@24-205-202-237.rno-cres.charterpipeline.net] has joined #mythtv
23:29<itsame>rkulagow_: I'm gonna try this xawtv test it talks about
23:29<itsame>I didn't lookin the troubleshooting part
23:29<itsame>sorry bout that
23:29<itsame>thnx
23:31<nziarek>moegreen: the "Align=center" does that align horizontally, vertically or both?
23:32<moegreen>nziarek: that does horizontal, I just added an "allcenter" which does both
23:32<moegreen>well, I committed that a few hours ago
23:33-!-rcaskey [~rcaskey@adsl-33-65-87.asm.bellsouth.net] has joined #mythtv
23:33<Morph>weird.
23:34<Morph>rebooted the box, now it works.
23:34<nziarek>by horisontal we mean that I I I the middle I is centered horizontally and the middle line in an E is lined up vertically, right?
23:36<itsame>Anyone in here have a Shuttle SK41g ?
23:37* rcaskeywishes there were lirc modules for 2.4.20-1
23:37* rcaskeyreally doesn't want to give up running his stock kernel :{
23:39<nyquiljer>rcaskey: you shouldnt have to. just compile lirc, and do 'setserial /dev/ttySx uart none' before loading the module that lirc will build
23:39<rkulagow_>chutt: "read the error message" - that's pretty terse. :)
23:39<rcaskey>nyquiljer: im confused...
23:40<nyquiljer>rcaskey: i think you will need the source for 2.4.20 though
23:40<mdz>just the headers
23:40<rcaskey>well yeah, that's why I want binaries
23:40<Chutt>bah
23:40<rcaskey>I e-mailed the lirc maintainer and he said he would do it, but I dont think he plans to
23:40<mdz>Chutt: my frontend is up to 100M virtual
23:40<nyquiljer>rcaskey: you dont even have to compile the kernel, just get the headers
23:40<Chutt>mdz, ok
23:40<Chutt>track it down, now =)
23:40<rcaskey>nyquiljer: it's the principle of the thing though ;)
23:40<Chutt>does it go up while playing?
23:41<Chutt>anyone tried those mythweb conflict patches
23:41<nyquiljer>rcaskey: i guess i dont get it. you didnt want to compile a kernel, you stil dont have to
23:41<rcaskey>make-kpkg buildpackage modules would get the job done, but what's the fun in that
23:41<nyquiljer>rcaskey: thats even too damn much work. just 'apt=get install kernel-source-2.4.20' or whatever, then untar lirc and compile lirc. thats it
23:42<rcaskey>nyquiljer: nahh, its easier with buildpackage
23:42<rcaskey>i'v done it before
23:43<rcaskey>it just makes me irritated that there is no procedure for autobuilding binary modules
23:43<nyquiljer>lirc builds its OWN module
23:44<rcaskey>nyquiljer: i'll stick to the debian blessed way of buildinng modules
23:44<nyquiljer>all you need is kernel source. i dont see how it could possibly GET any easier
23:44<itsame>Is Kmix just a GUI fro alsamixer ?
23:46<rcaskey>nyquiljer: apt-get install kernel-header-blah && apt-get install lirc-module-source-blah && cd /usr/src/linux && make-kpkg buildpackage modules && dpkg -i /usr/src/lirc-modules-blah.deb && apt-get install lirc-x
23:46<nyquiljer>rcaskey: thats entirely too much work
23:46<TheAsp>rcaskey: you can use any kernel source with make-kpkg
23:46<rcaskey>but the advantage of that is it integrates with the packaging system
23:46<TheAsp>nyquiljer: it is soooooo not.
23:46<rcaskey>TheAsp: yeah, but I am just saying debian should be autobuilding the modules anyway for all the archs
23:47<TheAsp>who uses stock kernels?
23:47<TheAsp>:P
23:47<rcaskey>long time Debian users know thou shalt have no other packages besides me
23:47<nyquiljer>ARE there long time debian users?
23:48<Chutt>i use stock kernels on 2 of my machines
23:48<nyquiljer>i thought they all got fed up with the packaging system and switched to something else
23:48<rcaskey>anything of any complexity at all that gets built usually is well accompanied by alternatives work
23:48<hurdel>i'm trying to compile cvs on RH9 and i'm getting this error: "themedmenu.cpp:755: Internal compiler error in make_label_edge, at cfgbuild.c"
23:48-!-Ripp [] has quit ["Client exiting"]
23:48<rcaskey>.debs whip RPM
23:49<rcaskey>if you run a production server where everything you need is packaged, then it's tops
23:49<-- mdz(~mdz@216-15-124-77.c3-0.smr-ubr3.sbo-smr.ma.cable.rcn.com) has left #mythtv ("Client exiting")
23:49-!-mdz [~mdz@216-15-124-77.c3-0.smr-ubr3.sbo-smr.ma.cable.rcn.com] has joined #mythtv
23:49<nyquiljer>rcaskey: .tar.bz2 is the best
23:49* mdzkicks xchat hard
23:49<mdz>Chutt: seems to go up after playback cycles
23:50<Chutt>after it cycles?
23:50<mdz>90624 at idle -> 123796 during playback -> 108708 at idle
23:50<Chutt>hmm
23:50<hurdel>am I missing something that is causing the Internal compiler error?
23:50* rcaskeyneeds to get around to transcoding these dvd rips to divx
23:50<-- mdz(~mdz@216-15-124-77.c3-0.smr-ubr3.sbo-smr.ma.cable.rcn.com) has left #mythtv ("Client exiting")
23:51<Chutt>hurdel, mythtv can't cause that :p
23:51-!-mdz [~mdz@216-15-124-77.c3-0.smr-ubr3.sbo-smr.ma.cable.rcn.com] has joined #mythtv
23:51<Chutt>^W again? :p
23:51<mdz>yes
23:51<mdz>repeatedly
23:51<rcaskey>Chutt: btw, I have to say that using my hardware transcoder does look mighty fine, and some videophiles concur ;)
23:51<hurdel>Chutt : i know, but can any of the pre-reqs?
23:51<Chutt>hurdel, no.
23:52<hurdel>well, wth then? fresh install of rh9, is it that botched?
23:52<Chutt>bad hardware is the most likely thing.
23:52<mdz>Chutt: looks like there are some threads hanging around
23:52<Chutt>mdz, ok
23:52<mdz>7 threads at idle
23:52<Chutt>i'm recompiling for debugging now
23:52<Chutt>if you can figure out which threads, that'd be keen =)
23:52<mdz>an additional 5 during playback, then back to 7
23:52<-- mdz(~mdz@216-15-124-77.c3-0.smr-ubr3.sbo-smr.ma.cable.rcn.com) has left #mythtv ("Client exiting")
23:53-!-mdz [~mdz@216-15-124-77.c3-0.smr-ubr3.sbo-smr.ma.cable.rcn.com] has joined #mythtv
23:53<mdz>I'm going to kill someone
23:53<mdz>the xchat maintainer most likely
23:53<mdz>or find a new client
23:53<rcaskey>Xchat is great
23:53<rcaskey>I wish 2.x was in unstable already though
23:53<Chutt>be like me and use a dedicated machine for irc
23:53<mdz>rcaskey: er, it is
23:53<Chutt>that uses an assold checkout of unstable
23:53<rcaskey>mdz: really?
23:53<mdz>rcaskey: it's what I'm running right now
23:53<-- mdz(~mdz@216-15-124-77.c3-0.smr-ubr3.sbo-smr.ma.cable.rcn.com) has left #mythtv ("Client exiting")
23:53-!-mdz [~mdz@216-15-124-77.c3-0.smr-ubr3.sbo-smr.ma.cable.rcn.com] has joined #mythtv
23:53<itsame>Has anyone heard of the Shuttle SK41g audio working with Myth...I'm reading some bad audio stories about the VIA card
23:53<mdz>ad that is my problem
23:54<rcaskey>hmm, brb then ;)
23:54-!-rcaskey [] has quit ["Client Exiting"]
23:54<mdz>no other gtk2 programs have this problem
23:54<Chutt>mdz, anyway, you said 7 -> 12 -> 7
23:54<mdz>yep
23:54-!-rcaskey [~rcaskey@adsl-33-65-87.asm.bellsouth.net] has joined #mythtv
23:54<mdz>and I can't find wheer in the xchat source it binds it
23:54<rcaskey>your right
23:54<Chutt>curious about the 7, but..
23:54<rcaskey>must have just went in ;)
23:54<rcaskey>xchat2 is purdy
23:55<mdz>5 for playback sounds about right
23:55<Morph>itsame: only issue I have with the VIA board is the CPU isnt beefy enough to decode the video. VIA needs to release a driver for the decoder on their M boards :)
23:55<mdz>but I'm pretty sure there were less than 7 at idle before
23:55<Chutt>right
23:55<mdz>what do you have?
23:55<Chutt>2 at idle
23:55<nyquiljer>itsame: http://www.gossamer-threads.com/perl/mailarc/gforum.cgi?post=56507;search_string=Shuttle%20SK41g;guest=343689&t=search_engine#56507
23:55<Chutt>hrmm
23:55<Chutt>one at idle
23:56<Chutt>do you have the playback video preview on?
23:56<mdz>enabled, yes
23:56<mdz>displayed, no
23:56* rcaskeyis not sure if Gnome could benefit from an Apple style top-menu or not
23:56<Chutt>how's that?
23:56<mdz>all threads are 0.0% CPU
23:57<mdz>as in, it's turned on, but the playback screen is not displayed
23:57<mdz>so it's not previewing anything right now
23:57<PBlue_away>Can anyone tell me where the code is that mythbackend uses to determine if there is enough space to record?
23:57<Chutt>right
23:57<Chutt>ok
23:57<Chutt>pblue, it doesn't.
23:57<PBlue_away>ok...so what happens when you run out of space
23:57<PBlue_away>?
23:58<Chutt>things break
23:58<rcaskey>mdz: what do you think about changing the mythtv packages to accept any video users instead of having a seperate group for myth
23:58<PeteCool>exactly
23:58<moegreen>anyone notice if the 'down arrow' is still being shown when it shouldn't be on the playback and delete screens?
23:58<PBlue_away>...I guess that explains why I couldn't find it tho :)
23:58<PeteCool>you don't want to run out of space... 15 megs recordings suck
23:59<rcaskey>mdz: reason being is that Debian is going to have a special policy to make it easy for desktop users to add users to video, sound, etc
23:59<mdz>rcaskey: that _might_ be ok
23:59<mdz>access to the capture device and access to mythtv data should be pretty close in terms of access control
23:59<mdz>but I need to get to bed