Back to Home / #mythtv / 2003 / 08 / Prev Day | Next Day
#mythtv IRC Logs for 2003-08-09

00:22-!-hadees [] has quit [Read error: 104 (Connection reset by peer)]
00:24-!-hadees [~gremlin@pool-151-197-34-65.phil.east.verizon.net] has joined #mythtv
00:41-!-bline [~sbeck@h24-84-93-233.vc.shawcable.net] has joined #mythtv
00:52<bline>yay, I won the bid on the touchscreen
00:56<hadees>you going to use a touch screen for mythtv?
00:56<bline>yeah
00:57<bline>Have to attempt to hack in remote frontend to frontend events
00:57<bline>and make qt ignore focus
00:57<hadees>so it isn't going to be dual monitors?
00:58<bline>yeah, it is.. but people want it remote and it will solve my problems as well
00:59<hadees>is it going to be like remote then?
00:59<hadees>just used for controlling mythtv not really displaying music vis or tv or anything
00:59<bline>it's so people can put myth on a notepad and control the tv from that
00:59<bline>yeah
01:00<bline>or a pda maybe
01:00<hadees>ever heard of netremote.org, i know this isn't the best solution but it is what i plan on doing,
01:00<hadees>the reason it isn't the best is because it runs on windows ce
01:01<hadees>how ever it is open source, makes a pronto like remote, the only thing that needs to be done is write a server for linux
01:01<bline>it isn't going to be that hard, we already so something similar with lirc code.
01:01<bline>s/so/do/
01:02<bline>the hard part is going to be making Qt ignore focus
01:03<hadees>my hard part is some said they are working on it, but i have a feeling if i want it done i will have to do it my self from scratch
01:03<hadees>some = someone
01:21-!-rkulagow_ [~mythtv@12-206-148-147.client.attbi.com] has joined #mythtv
01:21-!-chmod_700 [] has quit [Read error: 104 (Connection reset by peer)]
01:30--> danc`(~cheesy@c66-235-37-121.sea2.cablespeed.com) has joined #mythtv
01:37-!-rkulagow [] has quit [Read error: 110 (Connection timed out)]
01:38-!-rkulagow__ [] has quit [Read error: 110 (Connection timed out)]
01:39-!-rkulagow__ [~rkulagow@12-206-148-147.client.attbi.com] has joined #mythtv
01:40<danc`>how does everyone load bttv? mine just stopped loading automatically
01:47<Captain_Murdoch>sometimes let it auto-load sometimes just "/sbin/modprobe bttv" as root
01:47<Captain_Murdoch>with this in my /etc/modules.conf
01:47<Captain_Murdoch>alias char-major-81 bttv
01:47<Captain_Murdoch>options bttv card=10 radio=0
01:47<Captain_Murdoch>options tuner type=2
01:48<danc`>hmm that's what i have too. i think have a larger problem though, nothing from my modules.conf is happening
01:54<bline>check dmesg for errors
01:58-!-hadees [] has quit [Remote closed the connection]
03:46-!-hadees [~gremlin@pool-151-197-34-65.phil.east.verizon.net] has joined #mythtv
03:51<-- danc`has quit ()
03:54-!-choenig [~choenig@pD9E09FBA.dip.t-dialin.net] has joined #mythtv
04:20-!-sfr [~sfr@pD9E60A49.dip.t-dialin.net] has joined #mythtv
05:37-!-bline [] has quit ["Client exiting"]
06:00-!-hadees [] has quit [Remote closed the connection]
06:45-!-sfr [] has quit ["Client exiting"]
07:31-!-sfr [~sfr@p50831B34.dip.t-dialin.net] has joined #mythtv
08:02-!-irish [some1@pc3-oxfd2-6-cust13.oxfd.cable.ntl.com] has joined #mythtv
08:05-!-FryGuy [] has quit [Read error: 104 (Connection reset by peer)]
10:37<-- irishhas quit ()
11:00-!-Rule [~Ruleke@D57773CF.kabel.telenet.be] has joined #mythtv
11:03<Rule>Hello folks. Can anyone explain to me how the mythtv commercial detection (basically) works ?
11:06<Captain_Murdoch>basic code detects blank frames then checks for blank frames that occur within a certain number of frames of each other (ie, about 30 seconds, about 60 seconds, etc.). newer code can also use a preliminary form of scene change detection based upon color histograms to detect segments of video with high rates of scene change. there's an option to use blank frame w/ scene change detection as well.
11:06<Rule>ic
11:07<Rule>trying to research a bit beforehand if it would work here
11:07<Rule>I think it will
11:07<Captain_Murdoch>also applies a little inteligence as to partial commercials at the beginning and end of videos followed by or preceeded by another commercial.
11:07<Captain_Murdoch>"here" where?
11:07<Rule>belgium
11:08<Captain_Murdoch>oh. :) if you have fixed-length commercials and they are usually separated by blank/blank/dark frames then it may work. you can always try just turning on the scene change detection code but that's nowhere near as accurate as the blank frame code can be if that works.
11:08<Rule>I think it might be seperated by blue frames
11:09<Rule>but I'm not sure
11:09<sfr>Rule fyi, some german users tested it and it worked ok (including me but i only tested it once so far)
11:09<Rule>ok
11:09<Rule>I'm still researching what video tuner to get
11:10<Captain_Murdoch>so you don't have Myth running yet?
11:10<Rule>I have a proof of concept
11:10<Captain_Murdoch>but no tuner?
11:10<Rule>with hardware I already have
11:10<Rule>indeed
11:11<Rule>I think I might buy that hauppage pvr250
11:11<sfr>Rule mythtv went beyond proof-of-concept a looong time ago ;-)
11:11<Rule>for me :)
11:11<Captain_Murdoch>ok. then you can go into the editor in Myth's player and seek to the frame/frames separating a commercial to see what color they are. also check to see how long commercials are to see if they fall between the 30/45/60/90/120 second intervals that are checked by the commercial detection code.
11:11<Rule>need to set it up a bit, etc.. show it to housemates...
11:12<Rule>ok
11:12<Captain_Murdoch>do you have high-speed internet?
11:12<Rule>yes
11:12<Rule>10mbit ;)
11:12<Rule>well theoretically... a download from mirrors at 1 mbyte/s
11:12<Rule>why ?
11:13<Captain_Murdoch>ok. I wrote the commercial skip/detection code. so well if it doesn't work for you then if you wanted sometime you could put up a file for me and I could take a look at it.
11:13<Captain_Murdoch>oops, hit enter too soon. :)
11:13<Rule>cool
11:13<Rule>I now need to figure out the status of the CLE266 drivers
11:13<Rule>especially the mpeg2 decoder
11:14<Rule>(via EPIA M boards)
11:14<Captain_Murdoch>oh well. yeah. I can take a look sometime if it doesn't work out of the box for you. I've got a list of other ways to try to detect commercials but have been spending too much time on other stuff/features recently.
11:14<Rule>anyone here have experience with those ?
11:14<Captain_Murdoch>mpeg2 decoders aren't supported in Myth.
11:14<Rule>well myth can call mplayer, right ?
11:14<Rule>to play DVD
11:15<Captain_Murdoch>for playing .avi and dvd yes.
11:15<Rule>oki
11:15<Captain_Murdoch>for playing anything really but it doesn't use mplayer internally for playback of .nuv files.
11:15<Rule>yes I figured as much
11:15<Rule>does it use its own code ? or a lib ? or what ? :)
11:16<Rule>dont have mpeg2 files now, things to check still .. :/
11:17<Captain_Murdoch>uses included libraries for decoding/encoding and display stuff was written by Chutt (Isaac) I think.
11:19<Rule>any work planned to support decoders ?
11:20<Rule>argh, the usb video capture thing hangs again :/
11:22<Captain_Murdoch>dunno about decoders. problem with most is that we'd have to reencode any non-mpeg2 video as well as encode the OSD to get it to display unless they support a framebuffer overlay. I think the ivtv driver is working on a framebuffer but not sure.
11:22<Rule>I read that
11:22<Rule>somewhere
11:47<Rule>http://www.snowman.net/pipermail/mythtv-users/2003-July/010736.html
11:47<Rule>apparantly there is no overlay, at least not 1 month ago :)
11:55-!-brtb [~brtb@93.18.35.65.cfl.rr.com] has joined #mythtv
11:55<brtb>not sure what i'm doing wrong... "unable to initialize plugin 'mythmusic'"
11:55<Captain_Murdoch>is that a fresh cvs checkout?
11:57<brtb>fresh as in a few days ago
11:57<brtb>something wrong with it?
11:59<Captain_Murdoch>not sure. I just updated my main frontend to current CVS yesterday and can't start mythmusic either anymore. just make sure I applied cvs.sql as well.
11:59<brtb>yah
11:59-!-dopez [~unknown@dopez.xs4all.nl] has joined #mythtv
12:01<brtb>just built a new box so i started form a clean DB... put in the mc.sql for the main, no problems there, put in metadata.sql for the music stuff, and there aren
12:01<brtb>'t any changes in cvs.sql for that
12:02<Captain_Murdoch>not sure. I haven't messed with mythmusic in months but figured I'd update that as well when updating my frontend and backend.
12:21<brtb>recompiling with debug
12:34<Captain_Murdoch>I haven't used myth in a lonnnnnggg time. :) I never installed flac and libcdaudio on my new myth frontend so I couldn't run the plugin. I compile everything on my fast Athlon then just run "make install" on the various frontend/backend machines. mythmusic plugin wouldn't run because it was missing flac and cdaudio libraries on that frontend.
12:43<brtb>well slack came with flac, vorbis and paranoia, I installed MAD and then it didn't have any trouble compiling
12:44-!-Rule [] has quit ["dinner"]
12:46-!-robertj [~robertj@c-24-98-40-6.atl.client2.attbi.com] has joined #mythtv
12:47<robertj>what could cause playback of live tv to hitch when recording is fine and there is plenty of cpu power left over?
12:48<paulproteus>disk access?
12:48<robertj>but why would that cause a problem?
12:49<robertj>theres not that much data being pushed pulled
12:49<robertj>I think its a 5400 rpm hd, but still
12:49<robertj>Chutt looked at my hdparm output and said things were normal
12:49<robertj>Im leaning towards drivers, but noone is answering my question on alsa-devel
12:50<robertj>I suppose i really ought to try a 2.6 kernel
12:53<paulproteus>I have an Xbox as a MythMusic + MythWeather system.
12:53<paulproteus>I could give it a wireless keyboard, but is there a nicer solution for a less "intrusive" remote control?
13:04<robertj>paulproteus: rf remote control perhaps interfaced to another computer sending events over tcpip to the xbox?
13:04<paulproteus>I can plug USB in.
13:04<paulproteus>Is there an RF USB remote that's not too expensive? :)
13:05<paulproteus>And/or, would Myth support an RF USB remote?
13:06<sfr>paulproteus if it's supported by lirc, then yes i'd think
13:10<paulproteus>Are there any plans to add song selection to MythMusic?
13:10<paulproteus>I mean, by typing or somehow asking for a single song.
13:11-!-paulproteus is now known as piman-
13:11-!-piman- is now known as paulproteus
13:13<paulproteus>Anyone have any experience with the ATI Remote Wonder?
13:14-!-courtlr [~trillian@66-52-249-103.sttl.dial.netzero.com] has joined #mythtv
13:15<-- courtlr(~trillian@66-52-249-103.sttl.dial.netzero.com) has left #mythtv
13:22<mdz>does the brightness/contrast/hue stuff only work with V4L2?
13:23<sfr>mdz there was bug which prevented it from working with v4l, but it was fixed a long time ago, + the keyboard mapping has changed see keys.txt
13:24<mdz>sfr: it's returning -1
13:24<mdz>I'll fix it to print an error message when it does that
13:24<sfr>mdz mythtv from cvs?
13:24<mdz>sfr: yep
13:25<sfr>mdz, ah, i wondered you asking this kind of question :) didn't update since a week.
13:26<mdz>ah, GetV4L1 field interprets the V4L2 constants and does the right thing
13:36-!-bline [~sbeck@h24-84-93-233.vc.shawcable.net] has joined #mythtv
13:38<mdz>looks like bttv in 2.4.21 is fucked
13:41-!-robertj [] has quit [Remote closed the connection]
13:44-!-robertj [~robertj@c-24-98-40-6.atl.client2.attbi.com] has joined #mythtv
14:05-!-robertj [] has quit ["Client exiting"]
14:05-!-brtb [] has quit [Read error: 104 (Connection reset by peer)]
14:30-!-sc00p [] has quit [Read error: 110 (Connection timed out)]
14:30-!-matt_ [~matt@adsl-20-223-143.mia.bellsouth.net] has joined #mythtv
14:34-!-brtb [~brtb@93.18.35.65.cfl.rr.com] has joined #mythtv
14:36<matt_>i have some questions about how surround sound would work with myth...anyone around?
14:36-!-robertj [~robertj@c-24-98-40-6.atl.client2.attbi.com] has joined #mythtv
14:38-!-mattf [~matt@adsl-20-223-143.mia.bellsouth.net] has joined #mythtv
14:38-!-matt_ [] has quit [Read error: 54 (Connection reset by peer)]
14:38-!-mattf is now known as matt_
14:40-!-cerveza [~none@c66.169.161.70.ts46v-11.otn-e1.ftwrth.tx.charter.com] has joined #mythtv
14:41<cerveza>has anyone out there gotten a voodoo3 3500 card working with mythtv?
14:41<cerveza>http://forums.gentoo.org/viewtopic.php?p=456112
14:47-!-choenig [] has quit [Remote closed the connection]
14:50<robertj>when do you need to do between ./configure and make when compiling cvs?
14:51<sfr>robertj nothing
14:51<robertj>make: Makefile: Not a directory
14:52<robertj>dunno, updated from cvs and it fixed that error
14:52<sfr>robertj so it's working now?
14:55<robertj>satisfying deps now
14:57<robertj>making now though
15:00<robertj>between installing apache and compiling cvs, something tells me I might be dropping frames in my recording :)
15:00<robertj>I bet its also a BadIdea(TM) to alter you db schema on the fly as well ;)
15:05-!-brtb [] has quit [leguin.freenode.net irc.freenode.net]
15:05-!-robertj [] has quit [leguin.freenode.net irc.freenode.net]
15:05-!-dopez [] has quit [leguin.freenode.net irc.freenode.net]
15:05-!-sfr [] has quit [leguin.freenode.net irc.freenode.net]
15:05-!-rkulagow_ [] has quit [leguin.freenode.net irc.freenode.net]
15:05-!-mecraw [] has quit [leguin.freenode.net irc.freenode.net]
15:05-!-paulproteus [] has quit [leguin.freenode.net irc.freenode.net]
15:05-!-jfm_ [] has quit [leguin.freenode.net irc.freenode.net]
15:05-!-ca1vin [] has quit [leguin.freenode.net irc.freenode.net]
15:05-!-bigguy [] has quit [leguin.freenode.net irc.freenode.net]
15:05-!-robertj [~robertj@c-24-98-40-6.atl.client2.attbi.com] has joined #mythtv
15:05-!-brtb [~brtb@93.18.35.65.cfl.rr.com] has joined #mythtv
15:05-!-dopez [~unknown@dopez.xs4all.nl] has joined #mythtv
15:05-!-sfr [~sfr@p50831B34.dip.t-dialin.net] has joined #mythtv
15:05-!-rkulagow_ [~mythtv@12-206-148-147.client.attbi.com] has joined #mythtv
15:05-!-mecraw [~mecraw@dnvrdslgw11poolD171.dnvr.uswest.net] has joined #mythtv
15:05-!-paulproteus [~paulprote@roc-66-66-102-254.rochester.rr.com] has joined #mythtv
15:05-!-bigguy [bigman@h4.44.102.166.ip.alltel.net] has joined #mythtv
15:05-!-ca1vin [webtoker@impulse.vortex.org] has joined #mythtv
15:05-!-jfm_ [~jfm@nat-outgoing.dalalu.fr] has joined #mythtv
15:32-!-choenig [~choenig@pD9E09FBA.dip.t-dialin.net] has joined #mythtv
15:32-!-brtb [] has quit [Read error: 104 (Connection reset by peer)]
15:42-!-kja [~ke-aa@ti132110a080-2309.bb.online.no] has joined #mythtv
15:54--> Josh222|(~Josh222|@12-254-144-70.client.attbi.com) has joined #mythtv
15:55<-- Josh222|(~Josh222|@12-254-144-70.client.attbi.com) has left #mythtv
15:57<matt_>hm
15:59-!-mattf [~matt@adsl-20-221-251.mia.bellsouth.net] has joined #mythtv
15:59<mattf><matt_> hm
15:59<mattf><matt_> using aplay /dev/dsp, i get output on all of my 4.1 speakers (left, center, right, rear, and sub)
15:59<mattf><matt_> when watching tv, i only get output on the sub and left
15:59<mattf><matt_> any ideas what's up? i think i remember seeing something on the list about a bug where there was only output on the left speaker
15:59<mattf>(i said that before but i'm pretty sure none of it went through)
16:11-!-courtlr [~trillian@66.42.36.188] has joined #mythtv
16:13<mattf>or just unmuting the line in plays on all the speakers also
16:13<mattf>so i don't know what myth is doing, but it's not right
16:15-!-matt_ [] has quit [Read error: 110 (Connection timed out)]
16:15-!-hadees [~gremlin@pool-151-197-34-65.phil.east.verizon.net] has joined #mythtv
16:16<-- courtlr(~trillian@66.42.36.188) has left #mythtv
16:27<robertj>is mythtv-setup in the mythtv module in CVS still?
16:28<mattf>the actual program?
16:28<mattf>it's there... mythtv/setup/setup
16:29<sfr>robertj i think so, did you check via browse-cvs to find out?
16:29<mattf>http://cvs.mythtv.org/cgi-bin/viewcvs.cgi/mythtv/setup/
16:30<robertj>i installed mythtv and its running but I dont see the resulting binary around
16:30<robertj>does it have to be built seperately?
16:30<sfr>robertj the setup binary?
16:30<mattf>make install doesn't install setup
16:30<mattf>you have to run it from the cvs dir
16:31<mattf>what you can do is
16:31<mattf>cp $CVSROOT/mythtv/setup/setup /usr/local/bin/mythtv-setup
16:31<robertj>I think thats alright...
16:32<robertj>I think I should just copy my old config to the cvs dir
16:33* robertjwonders how well mythtv-setup jives over X
16:33<robertj>anser: not well
16:34<sfr>robertj just fine
16:34<robertj>sfr: im not getting anything but the background
16:34<robertj>there we go
16:34<sfr>robertj you're actually talking about running setup via a remote X connection?
16:35<robertj>yeah
16:35<robertj>actually I need to change a setting through the normal frontend
16:35<sfr>robertj i run it sometimes via ssh -X backend and it works ok.
16:35<robertj>its just slow, thats all
16:36<robertj>im on 802.11b anyway
16:36<sfr>robertj could be. 100Mbs ethernet here
16:44-!-sc00p [~sc00p@adsl-35-144-251.cae.bellsouth.net] has joined #mythtv
16:46<mattf>does anyone know what myth actually does with sound, starting with the line in and ending with line out?
16:50<sfr>mattf not really, imho it will record from your soundcards capture input, sync the video stream to it and play it back, but i guess you already knew that ;)
16:52<mattf>yea
16:54<Chutt>it plays back exactly what it records.
16:55<robertj>Chutt: is libflac-dev from sid no good for building mythmusic?
16:55<Chutt>no, it should be fine.
16:56<mattf>i just don't get why i only get output on the left speaker
16:56<Chutt>mattf, probably because it's only recording on the left channel for some reason.
16:59<robertj>what about libmad0?
16:59<robertj>maddecoder.cpp: In member function `virtual Metadata*
16:59<robertj> MadDecoder::getMetadata(QSqlDatabase*)':
16:59<robertj>maddecoder.cpp:600: error: `assert' undeclared (first use this function)
16:59<mattf>how does dolby encoded surround sound work? are all the other channels (right, center, rear, etc.) all encoded into the right channel?
16:59<Chutt>robertj, update your checkout.
16:59<robertj>oky
16:59<Chutt>mattf, dolby surround encodes all sound into the left and right channels.
17:00<Chutt>the 'center' channel is any audio that's the same on both channels
17:00<robertj>there we go, thanks
17:00<Chutt>and the 'rear' is anything out of phase
17:00<Chutt>essentially
17:01<hadees>what is the average file size created for 1hr of video recorded from an hardware mpeg 2 board
17:01<Chutt>it's anything you tell it to be.
17:01<paulproteus>4KB ?
17:01<robertj>42kb
17:01<Chutt>well, within reason =)
17:02<mattf>hmm
17:02<hadees>i know i can set it up to anything, how ever i don't have one yet so i was wondering what the consensus was
17:02<sfr>Chutt, how difficult would it be to select either the left/right channel to be recorded and played back on both channels (this is for movies with two languages being transmitted, one per channel)
17:03-!-anduinw [~awithers@polias.anduin.com] has joined #mythtv
17:03<Chutt>sfr, shouldn't be _terribly_ difficult, but i'd make that a playback option
17:03<Chutt>record both, etc. hardest part would be selecting which channel to play
17:04<hadees>how large would the files be for an 1hr of TV quality
17:05<mdz>robertj: libflac-dev from testing/unstable is the only version which will work, in fact
17:05<mdz>the 1.0.2 in woody is too old
17:05<mdz>(and incompatible)
17:05<Chutt>mdz, do you ever communicate with flac upstream?
17:05<Chutt>having an 'assert.h' is fucking stupid.
17:05<mdz>Chutt: yeah, flac has been pretty dead upstream for a while now though
17:06<mdz>he's at LWE right now
17:06<robertj>I'm getting pretty mad at this ghetto driver
17:06<hadees>i am really trying to fine out how large a HD i should buy
17:06<anduinw>hadees - how long is a cubit?
17:06<robertj>and alsa has to have some of the worst support of any major linux subsystem
17:06<Chutt>someone who went to lwe should let me know if VIA really did demo myth
17:06<mdz>Chutt: yeah, that should really be called something else
17:07<mdz>though, it is FLAC/assert.h, so it shouldn't conflict with <assert.h>
17:07<Chutt>right
17:07<Chutt>but i can't just blindly include /usr/include/FLAC to deal with redhat's brokenness
17:07<Chutt>because of that
17:07<hadees>anduirw: A measure of length, being the distance from the elbow to
17:07<hadees> the extremity of the middle finger.
17:08<anduinw>hadees - exactly, it varies. I don't think you'll ever buy too large a drive.
17:08<mdz>Chutt: get my message about the ia64 experiment? got playback more or less working, except that the OSS driver for that card sucks
17:08<Chutt>ah
17:08<Chutt>cool
17:09<robertj>is there any place a driver might store its options besides in modules.conf?
17:09-!-paulproteus [] has quit ["Client exiting"]
17:09<mdz>it didn't have DSP_CAP_REALTIME
17:09<Chutt>anduinw, i'm merging in your patches now, btw
17:09<robertj>It's like the defualt sampling rate has been permenantly changed
17:09<robertj>even across reboots, and I just edited the moduels.conf so the update-moduels should have fixed that...
17:09<anduinw>chutt - even the ones you didn't like? I showed up here to try to justify them btw :)
17:10<mdz>Chutt: unfortunately, if we wanted to support it for real, we'd have to break file format compatibility
17:10<hadees>anduinw: well how big is your mythtv drive
17:10<Chutt>the extra functions?
17:10<Chutt>i'm just removing em
17:10<Chutt>mdz, heh, for 64bit safeness?
17:10<mdz>Chutt: alignment
17:10<mdz>extendeddata has 64-bit elements that aren't aligned on a 64-bit boundary
17:10<anduinw>hadees - the master has a 120 GB, the slave has a 160 GB.
17:11<mdz>if there had been an even number of ints before the first long long, everything would have been OK
17:11<Chutt>aah
17:11<mdz>but it's odd
17:11<Chutt>can't you specify the structure as packed?
17:11<mdz>maybe
17:11<Chutt>that should fix it
17:11<mdz>but the CPU might require that alignment
17:11<Chutt>it's slower to read in, of course
17:11<Chutt>hmm
17:11<anduinw>Chutt - they're not extra, they allow it to function based entirely on the thread state rather than some explicit error condition
17:11<mdz>well, I guess you could just read them in one at a time
17:12<mdz>instead of reading the whole struct as one blob
17:12<hadees>anduinw: and how many hours does your frontend usally hold? and do you use mpeg 2 or mpeg 4
17:12<mdz>that would fix it too
17:13<Chutt>mdz, i _think_ putting an attribute(__packed__) or however you format it should work
17:13<mdz>could be
17:13<mdz>will that automatically fix up unaligned memory access?
17:13<anduinw>hadees - mpeg 2 (I have several pvr 250s). How many hours is a function of the bit rate I use though. I've tried a few and have pretty much settled on 4mb VBR (1 hour = about 2 GB)
17:14<mdz>so that when you go to access the 64-bit values, it would need to do it as 2 32-bit loads or something instead of one
17:14-!-paulproteus [~paulprote@roc-66-66-102-254.rochester.rr.com] has joined #mythtv
17:14<Chutt>hm
17:14<Chutt>i dunno
17:14<hadees>anduinw: thank you
17:15<anduinw>hadees - I live to serve.
17:15<Chutt>anduinw, but setting that variable could easily be done elsewhere, so i don't see the need for the specific functions, especially as they're only used in one place each
17:17<anduinw>Chutt - I didn't use an explicit variable because I made it a laziness catch. Not honoring external expectations was how the problem started. By making the check whether the thread was still alive rather than some error state I was attempting to ensure that when someone adds another error check to the decoder thread they don't inadvertently cause another unending loop.
17:18<anduinw>Chutt - As for the one place argument, well yeah, but only because I didn't fix other places :)
17:19<anduinw>Chutt - still I can see your objection, as I said I just came to plead for the life of my changes.
17:20<Chutt>heh
17:20<Chutt>they're going in
17:20<Chutt>just slightly changed :p
17:21-!-mecraw [] has quit [Read error: 104 (Connection reset by peer)]
17:21<mdz>Chutt: I think I'm going to modify README to strongly recommend Qt 3.1
17:21<Chutt>thanks
17:21<bline>morning
17:21<Chutt>i'm planning on doing so in the release announcement as well
17:21-!-mecraw [~mecraw@dnvrdslgw11poolD171.dnvr.uswest.net] has joined #mythtv
17:22<anduinw>Chutt - I've mentioned this before and didn't get a response. How sold is everyone on QSocket?
17:22<Chutt>i'd like to replace it
17:22<Chutt>just too lazy
17:22<Chutt>and it does have some nice features
17:22<Chutt>but it'd be nice not to have it need the event loop
17:23<Chutt>every time it goes to use the network
17:23<anduinw>Chutt - exactly, and no more close events firing from any thread and crashing everything!
17:23<Chutt>heh, yup.
17:23<mdz>Chutt: hey, __attribute__((packed)) seems to work fine
17:23<mdz>you have to specify it on each element of the struct to be packed, though
17:23<Chutt>i think it may be __attribute__((__packed__))
17:23<Chutt>at the end of the struct
17:24<mdz>in gcc 3.3 at least, it's __attribute__((packed))
17:24<Chutt>hrm
17:24<mdz>according to the docs, and it seems to have worked when applied to the last 3 leements of struct extendeddata
17:25<Chutt>anduinw, so you want to write a little replacement class? =)
17:26<Chutt>QSocketDevice may be a decent starting point, i believe it's a rather low level socket wrapper
17:26<anduinw>it used to be that only the windows port of egcs allowed easy entire struct packing attributes (so they could support the MSVC #pragma pack(push, 1) sort of stuff), normal gcc forced you to set the attribute on every member. _maybe_ this has changed.
17:27<Chutt>ah
17:27<anduinw>Chutt - A replacement class, and a slight architectural change. Make the threads more responsible for the lifetime of sockets... somewhat minor.
17:27<Chutt>anduinw, that's cool. i'd be happy with that
17:28<Chutt>as for your lirc patch, all i did was remove the namespace (since they're not used anywhere else, really), and move the ifdef around customEvent inside the class
17:28<Chutt>err, inside the function
17:28<Chutt>so customEvent's always there
17:29<anduinw>Chutt - good to hear, I'll take a crack at it then.
17:30<bline>wasn't someone talking about makeing mythfrontend recieve events from other mythfrontends?
17:30<Chutt>external sources
17:30<Chutt>not necessarily other mythfrontends
17:30<bline>did anyone ever do any code?
17:31<Chutt>no, that was thor
17:31<anduinw>Chutt - yeah sorry about the namespace, I just didn't want to throw the enum global after I decided someday someone might want to send a different sort of lirc event. The ifdef move was probably a good one, make sure you get the one I stuck in the header as well :)
17:31<Chutt>he got distracted by the dvd stuff
17:31<bline>ok, cause I won the bid on the touch screen
17:33<Chutt>oh, i made that enum a constant, too
17:36<anduinw>Chutt - Just because of the warning?
17:36<Chutt>naw, just because i don't like single element enums
17:36<anduinw>ah
17:39<Chutt>sfr, aside from all the tabs, your patch looks fine to me
17:41<sfr>Chutt, sorry for that, i'll have to check how to convert tabs to spaces in vim, any reason why you don't like them?
17:41<anduinw>set expandtab, :retab
17:42<sfr>ok, i will try to remember it.
17:42<mdz>.vimrc
17:42<anduinw>i do a set ts=4 sw=4 expandtab whenever I'm playing with myth source
17:42<mdz>you can add a magic comment to the end of each file to do that automatically
17:42<mdz>dunno how Chutt feels about those
17:44<anduinw>yeah that last thing was what I worried about, can also do local .vimrc but you need to sprinkle them in various dirs
17:44<Chutt>eh
17:44<Chutt>i don't like em, but.. =)
17:44<Chutt>sfr, mainly, i don't like mixing tabs and spaces in the same file
17:45<sfr>i just corrected the obvious (to me) places
17:46<Chutt>want to resend it to me off list?
17:47<sfr>np
17:47<Chutt>thanks =)
17:52<sfr>Chutt btw, i'm not experiencing these 'unknown command ??UERY_RECODER' segfaults anymore. it might have been related to some package in sid
17:53<Chutt>sfr, i rewrote that code
17:53<sfr>Chutt i must have missed the related cvs commit
17:54<Chutt>i don't think i mentioned that error message specifically
17:54<Chutt>but i changed how that all worked
17:54<Chutt>wasn't sure if it'd fix that or not, but it seemed like a good idea
17:54<sfr>so what 'area' was the bug in?
17:54<Chutt>if you do see it again, let me know please
17:55<Chutt>i don't know, i never got any more info than what you provided
17:56<robertj>Chutt: mythmusic is playing things at maby 1/2 speed, and they sound fine in xmms
17:56<robertj>any ideas?
17:56<Chutt>nope
17:56<sfr>but i see the (remote) mythfrontend stopping to accept any keyboard input, have to CTRL-C it and restart.
17:57<Chutt>which output driver are you using in xmms?
17:57<Chutt>sfr, ctl-c it in gdb sometime, and send me a bt of where it's hung up
17:57<Chutt>after recompiling for debugging, of course
17:58<robertj>ahh,, found it. myth music has it's own dsp config
17:58<sfr>Chutt when it happens it still plays tv fine, just the keyboard input isn't processed.
17:58<Chutt>huh
17:59<Chutt>sure it isn't losing focus somehow?
17:59<Chutt>i'd still need to see a backtrace if at all possible
17:59<sfr>i don't think, i'm using windowmaker
17:59<Chutt>robertj, yeah, that code hasn't been merged yet
18:00<sfr>Chutt how can i get the backtrace without hitting CTRL-C then? or is that ok then.
18:00<Chutt>switch to the gdb terminal
18:00<Chutt>and hit ctl-c in there -- it'll break it and allow you to get a backtrace of all the threads
18:01<sfr>ok, will try that some time.
18:12-!-paulproteus [] has quit [Remote closed the connection]
18:15-!-paulproteus [~paulprote@roc-66-66-102-254.rochester.rr.com] has joined #mythtv
18:16-!-choenig [] has quit [Read error: 104 (Connection reset by peer)]
18:18-!-choenig [~choenig@pD9E09FBA.dip.t-dialin.net] has joined #mythtv
18:19-!-lunk [~lunk@user92.net549.nc.sprint-hsd.net] has joined #mythtv
18:21<-- mattfhas quit ()
18:24-!-robertj [] has quit ["Client exiting"]
18:24-!-hadees [] has quit [Remote closed the connection]
18:39-!-kja [] has quit [Remote closed the connection]
18:40-!-kja [~ke-aa@ti132110a080-2309.bb.online.no] has joined #mythtv
18:46<kja>how do you feel about making channel changes 'snappier'?
18:47<anduinw>quick channel changes, bah, who'd want that? (sorry someone had to say it)
18:50<kja>it just annoys me more and more, feels like it takes a half an hour to change channels ;)
18:52<kja>i've done some peeking, and it seems to me that mythtv might require some light reconstructive sugury to make it happen
19:01-!-brtb [~brtb@93.18.35.65.cfl.rr.com] has joined #mythtv
19:02<brtb>music works now, no idea why
19:11-!-hfb [] has quit ["Client exiting"]
19:26-!-kja [] has quit [Remote closed the connection]
19:27-!-kja [~ke-aa@ti132110a080-2309.bb.online.no] has joined #mythtv
19:31-!-thor_ [1000@193.251.157.230] has joined #mythtv
19:36-!-kja [] has quit ["zzzzzZZZZZZ"]
19:37-!-thor_ [] has quit ["using sirc version 2.211+KSIRC/1.2.1"]
19:38<-- sfr(~sfr@p50831B34.dip.t-dialin.net) has left #mythtv ("Client exiting")
19:41-!-choenig [] has quit [Read error: 54 (Connection reset by peer)]
19:50-!-nitrogen [~nitrogen@cpe-66-87-226-247.ut.sprintbbd.net] has joined #mythtv
20:21-!-lunk [] has quit ["burn baby burn"]
20:21-!-brtb [] has quit [Read error: 54 (Connection reset by peer)]
20:50-!-FryGuy [~fryguy@12-246-43-18.client.attbi.com] has joined #mythtv
21:18-!-dopez [] has quit [Read error: 110 (Connection timed out)]
21:20-!-dopez [~unknown@dopez.xs4all.nl] has joined #mythtv
21:23-!-mecraw [] has quit [Read error: 60 (Operation timed out)]
21:29<cerveza>can anyone help me get this voodoo3 3500 tv card working with mythtv? it works fine with xawtv
21:31-!-Edgan [] has quit ["Client exiting"]
22:13<mikegrb>I love you
22:13<mikegrb>oops
22:13<mikegrb>wrong window, was supposed to go to my wife
22:24-!-Edgan [edgan@proton.cygnusx-1.org] has joined #mythtv
22:29<bline>Chutt: you around?
23:27<bline>if you see this in scroll back, I added a MythApplication class that has a x11EventFilter in it to filter out events that would unfocus the app. With this class I can irc in on screen and use my remote on myth in the other.
23:28<bline>Not sure if you want a patch for this, it is rather a hack.