Back to Home / #mythtv / 2003 / 02 / Prev Day | Next Day
#mythtv IRC Logs for 2003-02-18

00:21-!-paperclip [] has quit ["grits.. they aren't just for breakfast anymore"]
00:22-!-paperclip [] has joined #mythtv
00:25<SoopaFreenode>paperclip, what are grits?
00:27<paperclip>kernel from corn
00:34<SoopaFreenode>does it really take that long to make?
00:35-!-paperclip_ [] has joined #mythtv
00:37<-- paperclip_( has left #mythtv ("grits.. they aren't just for breakfast anymore")
01:39-!-billytwowilly [] has joined #mythtv
01:39<billytwowilly>hey, anyone have any bluetooth devices controlling their mythtv box?
01:43<billytwowilly>bluetooth. like the palm tungsten T
01:43-!-bigguy [] has quit [Read error: 60 (Operation timed out)]
02:03-!-bigguy [] has joined #mythtv
02:04<-- billytwowilly( has left #mythtv ("Client exiting")
02:37-!-SoopaFreenode [] has quit [Read error: 110 (Connection timed out)]
03:23-!-bigguy [] has quit [Read error: 110 (Connection timed out)]
03:41-!-Viddy [] has quit [Remote closed the connection]
04:11-!-paperclip [] has quit ["grits.. they aren't just for breakfast anymore"]
04:52-!-nevertheless [] has joined #mythtv
04:55-!-paperclip [] has joined #mythtv
05:04-!-paperclip [] has quit ["grits.. they aren't just for breakfast anymore"]
05:45-!-nevertheless [] has quit []
05:45-!-PeteCool [] has quit []
05:45-!-mdz [] has quit []
05:45-!-vidar [] has quit []
05:45-!-jfm [] has quit []
05:45-!-yebyen [] has quit []
05:45-!-Justin_ [] has quit []
05:45-!-hurdel [] has quit []
05:45-!-Chutt [] has quit []
05:45-!-poptix [] has quit []
05:46-!-nevertheless [] has joined #mythtv
05:46-!-PeteCool [] has joined #mythtv
05:46-!-mdz [] has joined #mythtv
05:46-!-poptix [] has joined #mythtv
05:46-!-vidar [] has joined #mythtv
05:46-!-hurdel [] has joined #mythtv
05:46-!-jfm [] has joined #mythtv
05:46-!-Justin_ [~justin@H233-225.STATE.RESNET.ALBANY.EDU] has joined #mythtv
05:46-!-yebyen [] has joined #mythtv
05:46-!-Chutt [] has joined #mythtv
05:56-!-Viddy [] has joined #mythtv
06:46-!-bigguy [] has joined #mythtv
06:54-!-PeteCool [] has quit ["Client Exiting"]
09:13-!-Chutt [] has quit [Read error: 60 (Operation timed out)]
09:18-!-Chutt [] has joined #mythtv
09:38<rkulagow_>morning. chutt, have you considered using /etc/mythtv rather than /usr/local/share/mythtv? i know i could use a symlink, but just wondering about that.
09:38<Chutt>stuff the user installs manually should go into /usr/local/
09:39<rkulagow_>fair enough.
09:39<Chutt>and, i suppose it could go into /usr/local/etc/, but that should just be config files
09:39<Chutt>and we're running low on config files these days, what with everything going into the db
09:40<Chutt>i just woke up and started doing email, i'll get to your patch soon
09:41<rkulagow_>NP, do it when you get to it. i'd like to contribute more code-wise (where i can), and just trying to figure out your philosophy and what not.
09:44<rkulagow_>i was thinking more about that sound and volume/mute patch. it'd be better if it were generic, so that mythmusic could take advantage of it too. i haven't started going through libmyth yet; i'm assuming that libmyth will have functions that are common for the various modules?
09:45<Chutt>if you wanted to make a generic volume class that belonged in libmyth, that'd be fine
09:45<rkulagow_>do you happen to know if the audiofd is a global? i've read more on the OSS docs, and i think that rather than opening the mixer device, i can just manipulate the mixer via the fd that's already been opened.
09:46<Chutt>it'd be better to open the mixer
09:48<rkulagow_>ok; should that be configurable, or should i deduce it based on the audiooutputdevice dsp? OSS says you can't assume that DSP1 == MIXER1, but i'm sure i can get it by re-reading the OSS docs again. having the user configure it on the setup screen is "easy", since i already did it in the patch that i posted to the ML last week.
09:49<Chutt>it should be configure it
09:49<Chutt>sorry, not awake =)
09:50<rkulagow_>NP. know the feeling. 2 month old, etc, etc.
09:51<rkulagow_>would you want the volume class stuffed into one of the existing source files in libmyth, or do you want it standalone?
09:51<Chutt>it could be standalone
09:54<Chutt>rkulagow, that duplex patch looks good
09:54<Chutt>want to commit it yourself?
09:54<rkulagow_>what about setting the mixer values to an initial value? in the patch that i had submitted, i had two sliders on the mythfrontend setup; one controlled the master mixer and one controlled the PCM volume. all the code that i had included manipulated the PCM volume. is that still going to be OK? if you remember, i defaulted to "70" for the volume just to make sure that it was _something_ rather than too low.
09:54<rkulagow_>aw, shucks. no, go ahead please.
10:03<Chutt>it's in cvs now
11:12<Chutt>can someone update to current mythmusic cvs and let me know if everything compiles ok?
11:13<rkulagow_>i'm pulling it now.
11:14<rkulagow_>ran configure; don't have OpenGL, FFTW or SDL. Did you want to check any of that?
11:14<Chutt>oh, hmm
11:14<Chutt>yeah, try to compile without all of that
11:15<Chutt>which is good, since I have all 3, so it's more difficult to test =)
11:15<rkulagow_>i think i've got the pre-reqs, so lets see what happens.
11:16<rkulagow_>nope, don't have cdda_interface on my secondary devel box. let me check the other one.
11:19<rkulagow_>looks like not:
11:19<rkulagow_>g++ -c -pipe -Wall -W -O3 -march=pentiumpro -fomit-frame-pointer -funroll-loops -fexpensive-optimizations -finline-functions -D_REENTRANT -D_GNU_SOURCE -DPREFIX=\"/usr/local\" -DQT_NO_DEBUG -DQT_THREAD_SUPPORT -I/usr/local/include -I/usr/lib/qt3/include -I/usr/lib/qt3/mkspecs/default -o mainvisual.o mainvisual.cpp
11:19<rkulagow_>cc1plus: warning: changing search order for system directory "/usr/local/include"
11:19<rkulagow_>cc1plus: warning: as it has already been specified as a non-system directory
11:19<rkulagow_>In file included from mainvisual.cpp:14:
11:19<rkulagow_>goom/mythgoom.h:4:24: mainvisual.h: No such file or directory
11:19<rkulagow_>goom/mythgoom.h:5:20: config.h: No such file or directory
11:19<rkulagow_>make[1]: *** [mainvisual.o] Error 1
11:19<rkulagow_>make[1]: Leaving directory `/home/mythtv/myth-0.8/mythmusic/mythmusic'
11:19<rkulagow_>make: *** [sub-mythmusic] Error 2
11:19<rkulagow_>[mythtv@pvr mythmusic]$
11:22<Chutt>update and try now please?
11:40<rkulagow_>looks like it compiled.
11:42<rkulagow_>now, does it work? that's the question. i was having problems with mythmusic crashing (probably due to that bug on obtaining the meta information from cddb or freedb or whoever). haven't had the cycles to see if any of that was fixed. if memory serves, didn't you say in the ML there was a patch in cddb CVS that fixed that?
11:49<Chutt>in the libcdaudio sourceforge bugzilla stuff, yeah
11:58<rkulagow_>hrmm. tried to follow the link in the message you posted when brent hills posted his patch, SF is saying "invalid artifact ID".
11:58<rkulagow_>libcdaudio appears to be well and truly dead; not much in terms of patches or mailing list...
12:06<rkulagow_>the libcdaudio page points you to musicbrainz, which provides a C++ library for pulling CD data, and appears to be Not Dead.
13:23-!-vidarino [] has joined #mythtv
13:44-!-Viddy [] has quit [Remote closed the connection]
13:45-!-Viddy [] has joined #mythtv
14:22<Chutt>rkulagow, biggest problem with using musicbrainz right now is that the most recent version isn't in any distributions whatsoever
14:23<Chutt>and iirc, the protocol's changed, so earlier versions no longer work
14:30<rkulagow_>chutt, Mandrake 9.1 has libmusicbrainz
14:31<rkulagow_>but that's just one data point.
14:31<rkulagow_>wait, don't know if that's the most recent version
14:32<rkulagow_>it's v1.1 of the library
14:32<Chutt>they're at 2.0
14:33<rkulagow_>right, i just checked the main site and saw that.
14:33<mdz>I don't see a wishlist bug open about it
14:33<Chutt>nothing uses it at all right now
14:33<Chutt>so there's no need for it to be packaged
14:35<rkulagow_>well, there are a few different implementations of cddb; it's too bad that the libcdaudio is no longer maintained.
14:35<Chutt>and i don't think the musicbrainz lib has any concept of local caching of the data
14:36<Chutt>pretty sure he leaves that up to the app writer
14:36<Chutt>that's all built into libcdaudio
14:36<Chutt>the best thing to do would be to strip down libcdaudio so that it's just for cddb lookup, and use an included version for it
14:38<rkulagow_>there's perl CDDB. since we use XMLTV as an external module, could we use perl CDDB instead of trying to modify libcdaudio?
14:38<Chutt>rather not.
14:41<mdz>it will be easier for things to start using it if it is packaged
14:41<mdz>I believe in pre-packaging :-)
14:41<mdz>nothing used libflac at the time I packaged it, but now 3 or 4 things do
14:43<Chutt>i get occasional X errors on the console when starting up the gears "vis" plugin
14:43<mdz>if it's already packaged, it's silly for it to lag behind a major release
14:43<mdz>there is too much quality software in Debian that is being neglected by the package maintainer
14:43<Chutt>isn't it orphaned?
14:44<Chutt>or did someone pick it up after bma dropped it when freeamp died
14:44<mdz>it's not orphaned
14:44<mdz>it might be up for adoption
14:45<mdz>nope, not up for adoption either
14:45<mdz>Maintainer for source package musicbrainz is Andreas Rottmann <>.
14:47<Chutt>the package was even updated recently for the c++ transition
14:47<mdz>same guy who maintains zinf
14:48<Chutt>which should be removed, since it's dead.
14:48<mdz>it was just uploaded January 19th
14:48<mdz>when was 2.0 released?
14:49<Chutt>just recently
15:38<hurdel>what are reasonable quality settings, bitrate mainly, for mpeg4 at 480x480?
15:44<mdz>that's pretty subjective, don't you think?
15:45<mdz>why not try a few different settings and see what you think looks good on your display?
16:08<bigguy>man goom is pretty
16:08<Chutt>the code is horrible
16:08<Chutt>it's not very reactive
16:09<Chutt>and it's quite slow =)
16:12<bigguy>well I don't have it running I just found a screenshot with it running
16:22-!-rkulagow [~mythtv@] has joined #mythtv
16:22<rkulagow>moegreen: are you here?
16:25<rkulagow>mythprogfind seg faults when i use page down to get to "60 Minutes" and then hit page up. i've got a backtrace.
16:26<rkulagow>Program received signal SIGSEGV, Segmentation fault.
16:26<rkulagow>[Switching to Thread 16384 (LWP 9466)]
16:26<rkulagow>0x4065a968 in operator==(QString const&, char const*) ()
16:26<rkulagow> from /usr/lib/qt3/lib/
16:26<rkulagow>Current language: auto; currently c
16:26<rkulagow>(gdb) bt
16:26<rkulagow>#0 0x4065a968 in operator==(QString const&, char const*) ()
16:26<rkulagow> from /usr/lib/qt3/lib/
16:26<rkulagow>#1 0xbfffced0 in ?? ()
16:26<rkulagow>#2 0x081256a3 in ProgFinder::pageUp() (this=0xbfffdc40) at progfind.cpp:671
16:26<rkulagow>#3 0x0812d005 in ProgFinder::qt_invoke(int, QUObject*) (this=0xbfffdc40,
16:26<rkulagow> _id=58, _o=0xbfffced0) at moc_progfind.cpp:118
16:34-!-rkulagow [] has quit ["using sirc version 2.211+KSIRC/1.2.4"]
16:41<rkulagow_>chutt: was the "hmm" directed towards me?
16:42<bigguy>the tenebrae2 shots look killer
16:54<hurdel>mdz: about reasonable settings, i thought i used to be running at 3100 bitrate but reinstalled fresh cvs (rebuilt db) and now 3100 does not run smooth, gotta back down to 2200 or so (haven't tried much in between) but seems lower than what i was using
16:55<Chutt>i usually use 3300
16:57<hurdel>Chutt : what kind of hardware you have? (me: p4 1.7 512DDR wintv go)
16:57<Chutt>that's all on the website
16:57<Chutt>hasn't changed in a long time =)
16:57<Chutt>rkulagow, without the trace from #1, it's going to be hard to debug that
16:58<Chutt>since that's where it crashed
16:58<rkulagow_>ok; i've got debug turned on in did i miss something?
16:59<Chutt>not sure, actually
17:06<mdz>rkulagow_: you'll need a debug version of qt to get that
17:07<Chutt>mdz, it's still in mythtv
17:07<Chutt>#2 is a function call
17:07<mdz>oh, #1
17:07<mdz>stack looks borked
17:07<mdz>0xbfffced0 is not in myth's text or libmyth
17:07<mdz>rkulagow_: set a breakpoint at progfind.cpp:671 and see what's going on
17:08<mdz>oh wow, a glibc that might actually make it into testing
17:09<rkulagow_>mdz: ok, let me see if i can figure out how to set the breakpoint. i'll check back if i get totally lost.
17:09<Chutt>so goom's using about 80% cpu on my xp 1800+
17:09<mdz>rkulagow_: "break progfind.cpp:671"
17:09<mdz>it's pretty heavyweight
17:09<rkulagow_>ok, thanks.
17:09<mdz>my CPU on my desktop died last week
17:09<Chutt>mdz, the one you were running about 20c to hot? =)
17:10<mdz>no, the other one
17:10<mdz>that was my myth box
17:10<mdz>which is rock solid stable now
17:10<mdz>and cooler
17:10<Chutt>what'd you do to cool it?
17:10<mdz>reinstalled the heatsink with good thermal goop
17:10<mdz>that waxy crap doesn't work right at all
17:10<Chutt>the thermal tape?
17:11<Chutt>mdz, for goom, the newest version is a little more reactive
17:11<Chutt>and i messed with it so that the scope's on screen all the time
17:11<Chutt>so there's at least something moving that's the music =)
17:11<mdz>meaning, like, it switches between animations when there is a strong beat?
17:11<mdz>it reminds me of cthulhu
17:11<Chutt>some color changes are tied to strong beats
17:11<mdz>oh, cthugha
17:12<mdz>whatever it's called
17:12<Chutt>and goom was really nicely laid out
17:12<Chutt>at least from the perspective of integrating it into mythmusic
17:12<Chutt>the code's crap otherwise, but that part was easy
17:13<Chutt>just need to figure out why the vis stuff pauses on song change
17:14<rkulagow_>mdz: mythmusic is almost done ripping, so i'll check in a few minutes.
17:21-!-rkulagow [~mythtv@] has joined #mythtv
17:21<rkulagow>mdz: here's the breakpoint.
17:22<rkulagow>(gdb) break progfind.cpp:671 Breakpoint 2 at 0x8125698: file progfind.cpp, line 671. (gdb) run Starting program: /home/mythtv/mythtv-0.8/MC/programs/mythfrontend/mythfrontend [New Thread 16384 (LWP 9720)] connecting to backend server: localhost:6543 [Switching to Thread 16384 (LWP 9720)] Breakpoint 2, ProgFinder::pageUp() (this=0xbfffdc40) at progfind.cpp:671 671 showSearchList(); (gdb)
17:22<rkulagow>ick. sorry
17:22<rkulagow>let me try pasting again.
17:23<rkulagow>(gdb) break progfind.cpp:671
17:23<rkulagow>Breakpoint 2 at 0x8125698: file progfind.cpp, line 671.
17:23<rkulagow>(gdb) run
17:23<rkulagow>Starting program: /home/mythtv/mythtv-0.8/MC/programs/mythfrontend/mythfrontend
17:23<rkulagow>[New Thread 16384 (LWP 9720)]
17:23<rkulagow>connecting to backend server: localhost:6543
17:23<rkulagow>[Switching to Thread 16384 (LWP 9720)]
17:23<rkulagow>Breakpoint 2, ProgFinder::pageUp() (this=0xbfffdc40) at progfind.cpp:671
17:23<rkulagow>671 showSearchList();
17:23<rkulagow>that's better, paste wise. does this help?
17:23<Chutt>ok, so now you need to step
17:26<rkulagow>ok, so just keep executing "step" at the gdb prompt until it segfaults again?
17:26<Chutt>i think
17:28<rkulagow>ok, segfaulted after about 100 steps.
17:29<rkulagow>i'm going to try pasting it in (last couple of them)
17:30<rkulagow>operator+ (s1=@0xbfffd3a0, s2=0x8215292 " ")
17:30<rkulagow> at /usr/lib/qt3/include/qstring.h:936
17:30<rkulagow>936 QString tmp( s1 );
17:30<rkulagow>(gdb) step
17:30<rkulagow>937 tmp += QString::fromAscii(s2);
17:30<rkulagow>(gdb) step
17:30<rkulagow>~QString (this=0xbfffd310) at /usr/lib/qt3/include/qstring.h:769
17:30<rkulagow>769 if ( d->deref() ) {
17:30<rkulagow>(gdb) step
17:30<rkulagow>QShared::deref() (this=0x83fe778) at /usr/lib/qt3/include/qshared.h:50
17:30<rkulagow>50 bool deref() { return !--count; }
17:30<rkulagow>(gdb) step
17:30<rkulagow>~QString (this=0xbfffd310) at /usr/lib/qt3/include/qstring.h:770
17:30<rkulagow>770 if ( d != shared_null )
17:30<rkulagow>(gdb) step
17:30<rkulagow>771 d->deleteSelf();
17:30-!-rkulagow [] has quit [Excess Flood]
17:30-!-rkulagow [~mythtv@] has joined #mythtv
17:30<Chutt>just paste where it last was in progfind.cpp
17:30<rkulagow>grr. it was about half way done. sorry.
17:31<rkulagow>ProgFinder::showSearchList() (this=0xbfffdc40) at progfind.cpp:856
17:31<rkulagow>856 curLabel++;
17:31<rkulagow>(gdb) step
17:31<rkulagow>832 for (int i = (tempSearch - ((showsPerListing - 1) / 2));
17:31<rkulagow>(gdb) step
17:31<rkulagow>860 if (gotInitData[curSearch] > 1)
17:31<rkulagow>(gdb) step
17:31<rkulagow>862 if (update_Timer->isActive() == true)
17:31<rkulagow>(gdb) step
17:31<rkulagow>865 curLabel = 0;
17:31<rkulagow>(gdb) step
17:31<rkulagow>867 for (int i = (int)(tempSearch*showsPerListing); i < (int)((te mpSearch+1)*showsPerListing); i++)
17:31<rkulagow>(gdb) step
17:31<rkulagow>869 if (initData[i] != NULL)
17:31<rkulagow>(gdb) step
17:31<rkulagow>Program received signal SIGSEGV, Segmentation fault.
17:31<rkulagow>0x4065a968 in operator==(QString const&, char const*) ()
17:31<rkulagow> from /usr/lib/qt3/lib/
17:31<rkulagow>(gdb) step
17:31<rkulagow>Single stepping until exit from function _ZeqRK7QStringPKc,
17:31<rkulagow>which has no line number information.
17:31<rkulagow>Cannot find user-level thread for LWP 9720: generic error
17:31<rkulagow>(gdb) step
17:31<rkulagow>Cannot find thread 16384: generic error
17:31<rkulagow>that's it
17:34<rkulagow_>(meaning, that's the end of the paste, as opposed to "ah-ha! that's it! i found the error.)
17:37<Chutt>before the if (initData[i] != NULL) line
17:38-!-OrangeSun [] has joined #mythtv
17:38<OrangeSun>hey guys! : )
17:38<Chutt>add this: cout << i << " " << tempSearch << " " << searchCount << " " << showsPerListing << endl;
17:38<Chutt>and let me know what it prints out immediately before it segfaults
17:38<OrangeSun>I want to look at adding font size/styles into the OSD areas.. Any advice?
17:39<Chutt>not really, that code is fairly simple right now
17:39<OrangeSun>I see that the OSD font stuff is getting pulled from the setup/database.
17:39<Chutt>you'd need to add other font types, and associate em to text sections is all
17:39<Chutt>just the font name
17:39<OrangeSun>can I try to do it all from the qtlook.txt equivalent in the OSD directories?
17:39<Chutt>yeah, i suppose
17:40<OrangeSun>would you like to point me at a file to play with specifically?
17:41<Chutt>just the osd* files in libmythtv/
17:42<rkulagow_>chutt: ok, recompiling
17:50-!-PeteCool [] has joined #mythtv
17:51<OrangeSun>Chutt: Can I also add a colour setting for the fonts? or are you averse?
17:52<Chutt>if you can figure out a way to make the fonts look ok colored, sure
17:53<OrangeSun>Chutt: Well, that's the job of the designers : ) What's the deal with this "COL_WHITE"? Can't I assign colour values in hex?
17:53<Chutt>the font rendering stuff does black and white
17:53<Chutt>and that's all
17:53<OrangeSun>so, how do such conversions occur in the other parts?
17:54<OrangeSun>like the EPG, and such?
17:54<Chutt>those aren't drawing onto video
17:54<OrangeSun>hehe : ) GOod point. OK. well, black and white will have to do for a while : )
17:55<rkulagow>chutt: here's the last few lines of that debug you had me put in:
17:55<rkulagow>44 6 37 7
17:55<rkulagow>45 6 37 7
17:55<rkulagow>46 6 37 7
17:55<rkulagow>47 6 37 7
17:55<rkulagow>48 6 37 7
17:55<rkulagow>-7 -1 37 7
17:55<rkulagow>Segmentation fault
17:55<Chutt>ok, good
17:55<Chutt>i should be able to fix that after dinner
17:55<rkulagow>great. glad i could help.
18:00<PeteCool>anybody here use a kdm session type to autoload myth? I'd need some sort of script, using the contents of my .xinitrc doesn't work at all
18:06-!-rkulagow [] has quit ["using sirc version 2.211+KSIRC/1.2.4"]
18:19<Chutt>rkulagow, think i fixed it
18:21<rkulagow_>chutt: ok
18:29<OrangeSun>Chutt: Quick question: I did the following:
18:30<OrangeSun>OSDTypeText::OSDTypeText(const QString &name, TTFFont *font,
18:30<OrangeSun> const QString &text, QRect displayrect,
18:30<OrangeSun> const QString &fontcolor)
18:30<OrangeSun>I get the following on compile:
18:30<OrangeSun>osdtypes.cpp: In constructor `OSDTypeText::OSDTypeText(const QString&, TTFFont*, const QString&, QRect, const QString&)':
18:30<OrangeSun>osdtypes.cpp:176: invalid conversion from `const char*' to `int'
18:30<OrangeSun>OH! COL_White is #?
18:30<OrangeSun>what number is it?
18:31<OrangeSun>got it : )
18:48<OrangeSun>where does "Getfont" get the fonts from?
18:51-!-bigguy [] has quit [Read error: 54 (Connection reset by peer)]
18:58<OrangeSun>Chutt: Can different Font TYPES be done? or is that back to the same problem dealie?
19:00<Chutt>yes, but not with the current code
19:02<OrangeSun>Chutt: OK, but it's not the same restriction as the font color thing?
19:02<OrangeSun>Chutt: I've so far done the following:
19:03<OrangeSun>1) Divided the OSD "categories of text" into 3 (other than time/chan/etc): a) Title b) subtitle, and c) description.
19:03<OrangeSun>each so far has a colour setting.
19:03<OrangeSun>but i want to add dropshadow (yes/no) settings and font size/face settings.
19:07<Chutt>really, the osd theme stuff needs reworked for what you want to do
19:07<OrangeSun>how do you mean?
19:07<OrangeSun>the osd.txt file?
19:08<OrangeSun>well, I'm adding quite a few settings..
19:08<OrangeSun>for example:
19:08<OrangeSun>str InfoTitleFontType=Tahoma
19:08<OrangeSun>str InfoTitleFontSize=30
19:08<OrangeSun>str InfoTitleFontColor=1
19:08<OrangeSun>that's for the title font stuff..
19:08<OrangeSun>and then something like that for each of the other two categories.
19:09-!-nevertheless [] has quit [Read error: 54 (Connection reset by peer)]
19:09<OrangeSun>does that sound kind of like how you would want it done?
19:10<Chutt>it's really be better to rework the whole thing so that it's in xml
19:10<OrangeSun>the osd?
19:10<OrangeSun>in xml? I can't compute : )
19:10<Chutt>the theme description for it, yes
19:11<OrangeSun>so, end result: Do you want me to continue doing this (My limited mental resouces mean I *can't* do the XML bit), or stop and leave it for somebody else?
19:11<Chutt>i'd kind of rather you left it to me
19:12<OrangeSun>heheh : ) Done and done.
19:22<PeteCool>I'm using this script (;search_string=X%20start;guest=151483&t=search_engine#50037) but I'm getting "connection to :0 denied", any ideas why this might happen?
19:24<PeteCool>I don't see why this won't work, nothing unusual in there
19:29-!-poptix [] has quit [Read error: 110 (Connection timed out)]
19:35-!-poptix [] has joined #mythtv
19:39<moegreen>PeteCool: you'll get a connection denied if you are attempting to connect to a display that is running as a different user
19:40<PeteCool>I'm having a hard time make X + myth start up at boot, I'm making some progress... when I'm done I'll probably write it down/distribute the files
19:40<PeteCool>it's such a ******* pain
19:41<moegreen>what window manager?
19:41<PeteCool>it worked fine a week ago, but seems like gentoo raised the "security" of their startup system
19:41<PeteCool>I'm using oroborus, but it doesn't have anything to do with it...
19:42<PeteCool>I tried using kdm+autologon, but it just rejected my script without explanations
19:42<PeteCool>now I'm only getting some errors about missing files (which I DID delete :(
19:44<PeteCool>I copied theme files to the wrong dir, deleted some files which I thought were leftovers from that
19:50<-- OrangeSunhas quit ()
19:58<PeteCool>Chutt: which vis runs faster (more fps) for you?
20:02<Justin_>PeteCool: what are you using kdm for?
20:02<Justin_>just add su user -c startx to your init scripts
20:09<PeteCool>Justin_: won't it ask for a password?
20:09<Justin_>why would it?
20:11<PeteCool>Justin_: unless the scripts are run as root, su can't switch you password-less-ly to any other user but yourself
20:12<PeteCool>Justin_: I won't be using kdm anyway, it doesn't work well and takes >15 seconds to load (very weird)
20:12<Justin_>if only those system startup scripts were run by root...
20:29<PeteCool>Chutt: I can't seem to find the mail about the asm compile error in mythmusic, was there a resolution for that problem?
20:30<PeteCool>goom/zoom_filter_mmx.c:62:33: warning: multi-line string literals are deprecated
20:33<mdz>PeteCool: there is no assembler in mythmusic, the problem is with your compiler
20:34<mdz>your assembler doesn't like the output of your compiler
20:34<Chutt>mdz, there's asm in the goom vis plugin
20:35<PeteCool>Chutt: what might be causing that? wrong nasm version?
20:35<Chutt>well, what's the error you're getting?
20:36<PeteCool>gcc -c -pipe -Wall -W -O3 -march=pentium3 -fomit-frame-pointer -fexpensive-optimizations -finline-functions -D_REENTRANT -D_GNU_SOURCE -DPREFIX=\"/usr/local\" -DQT_NO_DEBUG -DQT_THREAD_SUPPORT -I/usr/local/include -I/usr/qt/3/include -I/usr/X11R6/include -I/usr/X11R6/include -I/usr/qt/3/mkspecs/linux-g++ -o zoom_filter_mmx.o goom/zoom_filter_mmx.c
20:36<PeteCool>goom/zoom_filter_mmx.c:62:33: warning: multi-line string literals are deprecated
20:36<PeteCool>goom/zoom_filter_mmx.c: In function `zoom_filter_mmx':
20:36<PeteCool>goom/zoom_filter_mmx.c:62: impossible register constraint in `asm'
20:36<PeteCool>goom/zoom_filter_mmx.c:62: impossible register constraint in `asm'
20:36<PeteCool>goom/zoom_filter_mmx.c:131: confused by earlier errors, bailing out
20:36<PeteCool>make[1]: *** [zoom_filter_mmx.o] Error 1
20:36<Chutt>try not using pentium3 as your arch
20:41<Chutt>"It certainly has put an end to mythtv's use as a digital vcr for me."
20:41<Chutt>well, i think i shall just go cry now
20:41<mdz>jesus, merle is never going to get myth running
20:41<Chutt>i believe he's doing multiple installs
20:41<Chutt>or is forwarding questions from other people
20:41<mdz>he screws up every one then
20:42<mdz>I get this message that says ERROR opening file what could the problem be??
20:42<mdz>maybe I compiled it wrong
20:42<Chutt>i bet it's the kernel
20:42<PeteCool>nope, it's in glibc :)
20:44<Chutt>damnit, i'm having trouble not responding to that melodramatic guy on -users
20:46<PeteCool>Chutt: -march=i586 worked, thanks :)
20:46<Chutt>there's really no reason for excessive compile opts in mythmusic, so =)
20:47<PeteCool>woohoo, I finally have auto-starting myth back!
20:47<Chutt>mdz, is still working perfectly
20:47<Chutt>i really wonder why they don't switch back to it
20:49<mdz>yeah, I've been running 0.5.6 with the patch to use tvlistings forever now
20:49<mdz>works better than the current stuff
20:49<Chutt>proper end times, no missing channels
20:49<Chutt>no hole at midnight
21:02-!-cgray [] has joined #mythtv
21:03<cgray>hey, i was gone for a while and now mythfilldb is segfaulting... any ideas why?
21:03-!-bigguy [] has joined #mythtv
21:15-!-JonnyRo [] has joined #mythtv
21:23-!-cgray [] has quit [Read error: 60 (Operation timed out)]
21:31<JonnyRo>hows it going moegreen?
21:31<PeteCool>Chutt: I still see a mouse pointer in the vis mode in mythmusic
21:33<PeteCool>Chutt: and goom gives a blank screen and crashes :(
21:37-!-cgray [] has joined #mythtv
21:39<PeteCool>would a g400 run the viz faster than a gf2mx400?
21:43<Chutt>it's pure cpu speed, barely touches the video card
21:45<PeteCool>Damn, new CPU + Mobo is much more expensive than a used g400 :(
21:45<PeteCool>I need a job... for the first time ever
21:46<PeteCool>Those Logitech Z-680's I want aren't gonna pay themselves :(
21:48<PeteCool>Chutt: any ideas why goom won't run? Too slow CPU maybe? Or a bug?
22:16<Chutt>try editing the two new settings in mythmusic-settings.txt
22:25-!-cgray [] has quit [Read error: 60 (Operation timed out)]
22:34-!-Chutt [] has quit ["Client Exiting"]
22:48-!-Chutt [] has joined #mythtv
22:56<PeteCool>Chutt: isn't it possible to use hardware-accel OpenGL for the vis plugins? they certainly don't seem to do so
23:06-!-FredFunk [] has joined #mythtv
23:07<FredFunk>Anyone herE?
23:19-!-aw [] has joined #mythtv
23:20<aw>anybody else having trouble getting audio playback to work w/ latest cvs?
23:21<aw>or know how to fix it?
23:21<Chutt>read the mailing list.
23:22<aw>I did, didn't see a solution except "something must be wrong w/ your box..."
23:22<Chutt>oh well
23:24<-- rkulagow_(~rkulagow@ has left #mythtv
23:25<Chutt>there was a small thread about it today
23:25<Chutt>started by Larry Matter, i believe.
23:26* awlooks again
23:28<Chutt>that's the only audio issue there is currently
23:30<aw>hmm, that doesn't sound like what I'm seeing
23:31<Chutt>give it a try, though
23:31<Chutt>petecool, goom should work now in CVS
23:31<Chutt>petecool, the crappy mmx code was broken
23:31<Chutt>now it's not
23:31<mdz>this problem where the playback box shows up empty is happening even more often than before, now
23:32<mdz>eventually it will irritate me enough to debug it
23:32<Chutt>current cvs works perfectly for me
23:32<Chutt>both on my main box, and using my p3-550 as a remote frontend
23:32<mdz>sometimes it has even happened twice in a row
23:32<mdz>acts like a race
23:32<Chutt>it probably is
23:33<Chutt>backend/frontend getting out of sync
23:33-!-rkulagow_ [~rkulagow@] has joined #mythtv
23:33<mdz>I think bruce markey posted something about this problem, and backend/frontend getting out of sync
23:33<aw>Chutt: the change you're suggesting I try is reverting NuppelVideoPlayer back a rev?
23:33<Chutt>aw, yes
23:34<aw>k, will do, tnx
23:34<Chutt>revert the changes that that link shows
23:34<aw>yep, 1.125 vs 1.126
23:35<mdz>is all the frontend/backend communication synchronous?
23:35<Chutt>i wasn't sure off the top of my head if 1.126 was the most current or not
23:35<Chutt>mdz, it should be, yes
23:35<mdz>is there more than one connection to the backend ever, or do the threads share one?
23:35<Chutt>subthreads create their own connections if needed
23:36<Chutt>i think it may be a problem with the backend sending data to the frontend
23:36<Chutt>if it is, then i should be able to get around it by opening up another connection
23:38<aw>Chutt: yep, that fixes it, tnx!
23:38<Chutt>it does?
23:38<Chutt>can you un-revert it for a sec?
23:39<Chutt>and test something for me?
23:39<Chutt>lemme know when you're ready
23:41<Chutt>line 340 in NuppelVideoPlayer.cpp
23:41<Chutt>cout << info.bytes << " " << audio_buffer_unused << endl;
23:41<mdz>building mythtv is a lot slower after my involuntary CPU downgrade
23:42<mdz>is printing out each sent/received stringlist in SendReceiveStringList a reasonable way to see what's going on?
23:42<Chutt>kinda similar to bruce's debugging?
23:42<mdz>yeah I think so
23:42<mdz>it's not going to be a ridiculous amount of output, is it?
23:42<Chutt>but it'd be nice to know what the payload is, instead of just the size of it
23:43<mdz>I was just going to do something like this
23:43<mdz> cout << "SendReceiveStringList: send: " << strlist.join(" ") << endl;
23:43<mdz>next time I push new code onto my myth box
23:43<aw>Chutt: 81920 69120
23:43<mdz>if I can fix that, and somehow get it to buffer smoothly during playback, I'll be in great shape
23:44<Chutt>i think adding a smallish input ringbuffer for playback would work
23:44<Chutt>just pre-read stuff into a block of ram
23:44<mdz>I'm sure it's a lot worse for me than for you
23:44<Chutt>for buffering stuff
23:44<mdz>but it gets pretty choked up if I skip around too fast during playback
23:45<mdz>much better if I pause it and then skip around and then unpause
23:45<Chutt>mdz, does current cvs help that any?
23:45<mdz>how current? I updated it late last week or so
23:45<mdz>I think it was after I heard about some audio buffering stuff
23:45<Chutt>past couple days
23:45<mdz>I'll update it again tomorrow or so and see
23:46<Chutt>aw, what samplerate did you record at?
23:48<aw>I get the same values for livetv at 44.1 though
23:48<Chutt>you shouldn't =)
23:49<aw>and 48k...
23:49<Chutt>you should get 81920 64280 for 44.1
23:49<Chutt>but, hmm
23:49<Chutt>can you try and change that / 10 to a / 5
23:49<Chutt>and see if things run ok with that?
23:50<Chutt>or if they're still broken
23:50<aw>line 337?
23:52<aw>better... got a bit, then a bunch of "Delaying to next trigger", then things cleared up
23:52<aw>81920 56320
23:53<Chutt>i suppose i'll have to make that configurable
23:53<Chutt>defaulting to off
23:53<Chutt>oh well
23:53<aw>np, lemme know if you need more testing, thank you
23:54<Chutt>you can fix it without modifying all the other changes by just adding a 'audio_buffer_unused = 0;' a couple lines down
23:54<aw>fwiw, this is just a generic via builtin audio w/ alsa 0.9rc7
23:55<Chutt>normally, audio stops pretty fast when you pause it, right?
23:56<aw>yeah, there's sometimes a weird sync problem when changing channels
23:56<aw>the video switch before the audio
23:56<Chutt>the guy that made this change said it took almost a second for audio to stop when he paused things
23:57<Chutt>and it speeds up seek response time a little as well
23:57<aw>I also have an annoying lag in my audio that I wish I knew how to get rid of
23:57<Chutt>but apparently it doesn't work for everyone :(