Back to Home / #mythtv / 2004 / 08 / Prev Day | Next Day
#mythtv IRC Logs for 2004-08-31

---Logopened Tue Aug 31 00:00:23 2004
00:16--- <<-- Rince [] has quit (Read error: 60 (Operation timed out))
00:18--- ---> spyder_ [] has joined #mythtv
00:33--- <<-- spyder__ [] has quit (Read error: 110 (Connection timed out))
00:50* thor_ is looking forward to Chutt's response to that one
00:51dopester evenin thor
00:51thor_ heh
00:52dopester you talkin about postgress boy?
00:53thor_ hibernate framework stuff
00:53thor_ which part of any "harder to use" failed to parse?
00:54dopester wow.. thats a dandy
00:54dopester everytime i re-read it its more and more funn
00:54dopester funny
00:55thor_ "the first step should be a set of support methods to perform queries, updates & deletes in an sql independent way."
00:55thor_ looking forward to that
00:56dopester heh
00:57Chutt naw, i'm done with that
00:58dopester i was looking forward to a reply.. oh well..
00:58thor_ silence is golden
00:59thor_ in other news, I am about to commit a completely non-blocking playlist editor
00:59Chutt sweet.
00:59thor_ doesn't block when new data comes over the wire
00:59thor_ doesn't block when user selects a given playlist to edit
01:00thor_ insanely over engineered
01:00dopester sounds like a honda :)
01:00thor_ that would be "doesn't knock"
01:01Captain_Murdo| Chutt, are you running current CVS on your main setup? or more importantly have you seen any examples of this "black screen" during playback while audio is fine?
01:01Chutt yes, and no
01:02Captain_Murdo| Bruce emailed me and says it's happening to him occasionally and he thinks it's related to the decoderbase stuff I committed. only thing I can think of in there that would cause that is the seekreset code since a positionmap problem shouldnt' cause video not to display.
01:03Chutt nothing in there should affect playback
01:03Captain_Murdo| he did a couple cvs "-D" updates around that timeframe so I was wondering if you had seen anything.
01:03Chutt haven't been using it extensively
01:03Chutt been too busy with work
01:03Captain_Murdo| seekreset could screw it up but not cause a totally black screen.
01:04Chutt right, i meant it shouldn't cause playback to completely break while audio stays fine
01:04Captain_Murdo| he said he'd try to track it down further so I'll see what he comes up with.
01:04Captain_Murdo| just figured I'd ask in here before sending this reply.
01:04thor_ screensaver ?
01:05Captain_Murdo| thanks for responding on that OSD resize post also, I meant to ask you about it yesterday.
01:05dopester how long does it take an analog card to change channels generally now?
01:05Chutt a complete analog card is about a quarter second
01:05Chutt hardware encoder takes a couple seconds
01:05Captain_Murdo| thor_, :) possibly but I figure Bruce wouldn't fall for that.
01:05thor_ :-)
01:07Captain_Murdo| when you do a "-D", is that the local time or the GMT time?
01:08thor_ "The date is interpreted as being in the local timezone, unless a specific timezone is specified."
01:08Captain_Murdo| if it's GMT then his timeframe is the same timeframe of the ffmpeg update.
01:09Captain_Murdo| hmm, man says "local"
01:11--- ---> Rince [] has joined #mythtv
01:17Chutt heh
01:17Chutt so, what else is left?
01:17Chutt that bug, the recording options screen..
01:18thor_ just generic database abstraction and the gmail layer
01:18Chutt oh yeah, i forgot about those
01:20thor_ oh, and native p2p
01:21* Captain_Murd feels kinda stupid cause he just noticed he missed thor's comment earlier which was another quote from the man page. :(
01:21* thor_ begins to worry people may be ignoring him
01:22Captain_Murdo| figure I can get about 1 hour of video per gmail account?
01:22thor_ so, one gmail account per show
01:22thor_ and with native p2p, we could share the accounts
01:22* thor_ will shut up now
01:22Captain_Murdo| yeah, and you can name the accounts something like chanid_starttime
01:22Captain_Murdo| it's all coming together.
01:23thor_ Muuu Waaaa Haaa Haaa
01:33bobbyz hello! I was curious if anyone knows of any utilities to control the channel via serial port for a motorolla dtc2000 unit
01:34thor_ [21:49:16] <thor_> covered to death, see mailing list archive, also, please note topic
01:38bobbyz k
02:23--- ---> choenig [] has joined #mythtv
02:24--- User: *** Netslayer is now known as _
02:24--- User: *** _ is now known as Netslayer
02:27--- ---> eco [] has joined #mythtv
02:29--- ---> _kch_ [] has joined #mythtv
02:50--- <<-- iver [~iver@] has quit ("Leaving")
02:51--- ---> mchou [] has joined #mythtv
02:53--- ---> Edis0n [] has joined #mythtv
02:53Edis0n moin
03:00--- <<-- bobbyz [] has quit ("Client exiting")
03:26--- <<-- hadees [~hadees@hadees.sustaining.supporter.pdpc] has quit (Remote closed the connection)
03:37--- <<-- eco [] has quit ("Leaving")
03:37--- ---> eco [] has joined #mythtv
03:41--- <<-- mchou [] has quit (Remote closed the connection)
03:53--- ---> KillerBunny [] has joined #mythtv
04:10--- <<-- KillerBunny [] has quit (Read error: 104 (Connection reset by peer))
04:25--- ---> kro [] has joined #mythtv
04:25* kro dances the got lirc working jig
04:29--- <<-- spyder_ [] has quit (Dead socket)
04:30--- ---> spyder_ [~spyder@] has joined #mythtv
04:31--- <<-- Netslayer [] has quit (Remote closed the connection)
04:44--- <<-- choenig [] has quit (Remote closed the connection)
04:52--- <<-- rowley [] has quit ("Leaving")
05:33--- ---> orogor [] has joined #mythtv
05:33orogor hi
05:34orogor anyone there ever looked at msp3400 code ?
05:53--- ---> vedran [] has joined #mythtv
07:10--- User: *** splAt1 is now known as splat1
07:27--- <<-- _shad [~shad@] has quit (Read error: 110 (Connection timed out))
07:35--- Netsplit <-> quits: m0j0, gfiend
07:38--- Netsplit over, joins: gfiend, m0j0
07:44orogor .. no dev there ?
07:44--- <--- kro [] has left #mythtv ("ERC Version 4.0 $Revision: 1.577 $ (IRC client for Emacs)")
07:45--- <<-- Snow-Man [] has quit (SendQ exceeded)
07:45--- ---> Snow-Man [] has joined #mythtv
07:47--- <<-- Snow-Man [] has quit (SendQ exceeded)
07:51--- ---> Snow-Man [] has joined #mythtv
07:53--- <<-- cythraul [] has quit ("Time wasted on IRC: 15 hours 55 minutes 48 seconds")
07:54--- ---> _shad [~shad@] has joined #mythtv
07:56--- Netsplit <-> quits: vedran, onyxsoft, Chutt, jeffpc_, bbeattie, _Nero_, kvandivo, Rince
07:56--- Netsplit <-> quits: [M-M], killefiz
07:57--- Netsplit over, joins: vedran, Rince, bbeattie, jeffpc_, onyxsoft, _Nero_, Chutt, kvandivo, [M-M], killefiz
08:19--- ---> KillerBunny [] has joined #mythtv
08:33--- <<-- vedran [] has quit ("Leaving")
08:34--- <--- Mrgoose3 [] has left #mythtv ()
08:35--- <<-- Edis0n [] has quit (Read error: 110 (Connection timed out))
08:45--- <<-- orogor [] has quit ("Leaving")
08:49--- ---> KillerBun [] has joined #mythtv
08:53--- <<-- _kch_ [] has quit (Remote closed the connection)
08:53--- <<-- KillerBunny [] has quit (Read error: 60 (Operation timed out))
08:54--- ---> _kch_ [] has joined #mythtv
08:55--- <<-- _kch_ [] has quit (Client Quit)
08:58--- ---> _kch_ [] has joined #mythtv
08:59--- ---> __kch__ [] has joined #mythtv
09:00--- <<-- __kch__ [] has quit (Client Quit)
09:02--- ---> orogor [] has joined #mythtv
09:12--- <<-- orogor [] has quit ("Leaving")
09:23gfiend Captain_Murdoch I'm using current CVS on my main box in our living room, I've never seen it. Though I use MPEG@ only
09:24gfiend Chutt want me to make the change to only the "advanced" recording screen?
09:24--- User: *** gfiend is now known as GeckoFiend
09:29--- ---> dc_ [] has joined #mythtv
09:49--- ---> marc_in_lux [] has joined #mythtv
10:03--- ---> m0j0__ [~m0j0@] has joined #mythtv
10:06--- ---> gfiend [] has joined #mythtv
10:08--- <<-- marc_in_lux [] has quit ("Leaving")
10:08--- <<-- m0j0 [] has quit (Read error: 110 (Connection timed out))
10:17--- ---> schultmc [] has joined #mythtv
10:24--- <<-- GeckoFiend [] has quit (Read error: 110 (Connection timed out))
10:28--- <<-- KillerBun [] has quit (Read error: 104 (Connection reset by peer))
10:45--- ---> mecraw [~lmarlow@] has joined #mythtv
11:14gfiend In MythWeb "Show total length of recorded shows alongside of disk usage "... Is that the total length or the scheduled length?
11:26--- <<-- riksta [] has quit (Read error: 113 (No route to host))
11:35--- ---> riksta [] has joined #mythtv
12:14--- ---> sfr_ [] has joined #mythtv
12:14--- <<-- stoffel [] has quit (Read error: 110 (Connection timed out))
12:20--- User: *** gfiend is now known as GeckoFiend
12:53GeckoFiend did I miss anything major here? Hoping to make things easier for Isaac.
12:54--- ---> dja [] has joined #mythtv
13:09--- ---> Baylink [] has joined #mythtv
13:11--- ---> jf100 [] has joined #mythtv
13:41--- <<-- nrh1703 [] has quit (Success)
13:44--- ---> nrh1703 [] has joined #mythtv
13:55--- <<-- nrh1703 [] has quit (".<UPP>.")
13:56--- ---> xortch [] has joined #mythtv
13:56--- <--- xortch [] has left #mythtv ()
13:57--- ---> nrh1703 [] has joined #mythtv
14:05Baylink Anyone playing with current CVS?
14:11Baylink I'm trying to complile it, and while older CVS worked fine here, todays is crapping out.
14:22--- User: *** sfr_ is now known as stoffel
14:25--- ---> johnback_ [] has joined #mythtv
14:27--- <--- johnback_ [] has left #mythtv ()
14:27GeckoFiend Baylink he h this is the dev channel most of us use CVS
14:27Baylink Yeah; Stoffel was kind enough to point out a make distclean was in order; problem solved.
14:28--- ---> bjohnson [] has joined #mythtv
14:39Chutt baylink = jay ashworth?
14:43Baylink Yeah, why?
14:43Baylink Ah; that's you. Hi, Isaac.
14:46--- <--- bjohnson [] has left #mythtv ("Leaving")
14:48--- <<-- Baylink [] has quit ("User disconnected")
14:48--- ---> Baylink [] has joined #mythtv
14:52Baylink I lied: distclean didn't fix it: today's CVS is filing make in mythcommflag.
14:53Baylink (failing, even)
14:58--- ---> KillerBunny [] has joined #mythtv
14:58stoffel Baylink, i'll compile latest cvs in a minute
15:03Baylink Did you want to see the error?
15:04stoffel i already did.
15:04thor_ (sigh) what's it doing
15:04stoffel mythtv compiled fine here
15:05stoffel <Baylink> I extracted todays CVS and get a compile/link bug on mythtv: g++ -o mythcommflag main.o -L/usr/lib/ -L/usr/lib/qt3/lib/ -L/usr/X11R6/lib/ -L../../libs/libmyth -L../../libs/libmythtv -L../../libs/libavcodec -L../../libs/libavformat -lmythtv-0.15 -lmythavformat-0.15 -lmythavcodec-0.15 -lmyth-0.15 -lfreetype -lmp3lame -L/usr/X11R6/lib -lXinerama -lXv -lXxf86vm -lGL -lGLU -lqt-mt -lXext -lX11 -lm -lpthread
15:05stoffel <Baylink> main.o(.text+0xb84): In function `FlagCommercials(QSqlDatabase*, QString, QString)':
15:05stoffel <Baylink> : undefined reference to `NuppelVideoPlayer::RebuildSeekTable(bool)'
15:06Baylink It is on the link, on further inspection, which is weird, cause I didn't see a compiler error.
15:06thor_ like it's getting old myth libs from the wrong place
15:07thor_ you got a sticky -D from a previous cvs co/update ?
15:08Baylink I'm not far enough up on CVS to be sure how to tell. Should I do a clean checkout on everything and try again?
15:08thor_ well, if you've never used -D, that's not the problem
15:10Baylink This is the build script I'm written; the cvs command has varied only from checkout to update:
15:11thor_ ... I'm just buidling it here
15:12Baylink Cool; thanks.
15:13Baylink One possibility: my MythTV and Mythweb were updates at beginning of Today, the others were new checkouts. If I understand CVS well enough, that's not *supposed* to matter, but I thought I'd mention it.
15:15Chutt get rid of the old mythtv install in /usr/lib.
15:15thor_ ah, yupp, cvs over an installed package, 'fer example
15:16Baylink I shouldn't have one; my prefix has always been /usr/local... no it hasn't. I inadvertantly followed someone's instructions to change that, then changed it back. Damn. Thanks.
15:16Baylink Do they *all* start with libmyth?
15:18Chutt mostly
15:18Chutt there's other stuff to remove, though
15:19Baylink I'll do an ls -lrt and look at the times. // And, finally, some of the modules don't appear to *have* makefiles, even as stubs. Is that likely to change?
15:19Chutt be, /usr/lib/mythtv/, /usr/share/mythtv, /usr/bin/myth*, /usr/lib/libmyth*
15:19Chutt nope
15:19Chutt just have to follow the instructions to generate the makefiles.
15:20Chutt no sense in including generated files in cvs/source packages, as they're distro dependent
15:20Baylink Got it. Yeah.
15:20Baylink I'll fold that back into my script.
15:20Chutt shouldn't have to
15:20Baylink Thanks, guys; I think I'm on track now.
15:21Chutt since a cvs update won't blow it away, and you shouldn't be doing clean checkouts
15:21Baylink Oh. I see your point. But I gotta do it manually the first time.
15:21thor_ or check and print a warning
15:22Baylink Bartons are fast; no sense not just doing it, is there?
15:22thor_ ah, sorry ... I'm 2 subjects behind
15:23Baylink I picked mythvideo as an example, and it has two .pro files, and no INSTALL; is there a general rule about what to qmake?
15:24thor_ if there's a configure, run it first
15:24thor_ then qmake ; make
15:25thor_ but the myth ./configures are not m4 auto-sillyness, they're just ways for the user to choose what sub-parts to build
15:26thor_ and in each case, the README explains what to do, and ./configure --help lists options
15:31Baylink Got it. I'm just trying to script the whole build. And this is cvs newbie question, I realize, but: is there any way to get CVS to give me a token I can hand it back to get the same stuff on a later checkout.
15:34--- ---> xortch [] has joined #mythtv
15:34--- <--- xortch [] has left #mythtv ()
15:39thor_ you could make note of the date
15:40--- ---> hadees [] has joined #mythtv
15:43--- ---> Captain_Murdoch_ [] has joined #mythtv
15:46--- ---> Dibblah1 [] has joined #mythtv
15:46--- <--- Dibblah1 [] has left #mythtv ()
15:47--- ---> Dibblah1 [] has joined #mythtv
15:47Dibblah1 Anyone having any issues compiling from CVS? Just tried a clean checkout...
15:48Dibblah1 main.o(.text+0x2c21): In function `internal_media_init()':
15:48Dibblah1 : undefined reference to `MythMainWindow::RegisterMediaPlugin(Qstring const&, Qstring const&, int .....
15:48stoffel remove old libs in /usr/lib ;)
15:48Dibblah1 Doh.
15:48Dibblah1 Is that going to catch me out every time?
15:48stoffel Dibblah1, hey, just guessing ;)
15:50Dibblah1 export PATH=/usr/lib/ccache:$PATH
15:50Dibblah1 Doh. And now I'm pasting into the wrong window.
15:50Dibblah1 Time to go to bed.
15:51--- <--- jf100 [] has left #mythtv ()
15:54spyder_ chutt, what is your opinion on the videooutbase ? Could it be used to avoid having duplicate code in the different output plugins ?
16:01Chutt that's what it's there for
16:01Chutt there's not _that_ much duplicated code, though
16:03spyder_ I've implemented a black border detection filter, that sets extra attributes in the frame so that afterwards I can detect whether I need to do a MoveResize.
16:04spyder_ This code would need to be executed each time the filters are called
16:04--- ---> Chang [] has joined #mythtv
16:04Chutt every frame?
16:04spyder_ Since I use heuristics, yes
16:05spyder_ You don't know in advance which frame might have another aspect ratio
16:05Chutt right, but I'd think checking every half second would be almost as good, and much less of a performance hit
16:05spyder_ There are even channels that switch between programs/commercials
16:05stoffel now that sounds like a way to detect commercials
16:06spyder_ I don't know the code to detect commercials
16:07stoffel neither do i
16:08spyder_ Hmmm, I only report an aspect ratio change after 30 frames with the same detected ratio, so I guess it would be possible to redirect this to checking for a change every 30 secs
16:09kvandivo and once a change is detected you could look at other frames around it for robustness
16:09Chutt 30 frames or 30 seconds?
16:09spyder_ 30 frames
16:09spyder_ sorry
16:09Chutt that's cool.
16:09Chutt but, to answer your original question, that can probably live in the base class
16:10Chutt though, only the xv output class (and similar) can use it
16:10Chutt no xvmc, viaslice, ivtv, etc
16:10Chutt so, a function in the base class that can be called from videoout_xv's ProcessFrame() or whatnot
16:10spyder_ I was thinking of creating a ApplyFilters method in the videooutbase, in analogy to the DisplayOSD
16:11Chutt geckofiend, that list looks good, btw.
16:11Chutt thanks =)
16:13spyder_ is it ok if I adapt the xv output plugin and then submit my patch to the mailinglist so you can have a look ?
16:13Chutt of course
16:13spyder_ allrighty :-)
16:14--- <<-- KillerBunny [] has quit ("Leaving")
16:18--- ---> gfiend [] has joined #mythtv
16:18Chutt gfiend, hey
16:32spyder_ patch is on it's way... the problem will be calibration : I'm not quite sure that my values will work for others
16:33--- <<-- GeckoFiend [] has quit (Read error: 110 (Connection timed out))
16:34Baylink qmake and the lib cleanup did it. I'm golden; thanks, guys.
16:37--- <--- Dibblah1 [] has left #mythtv ()
16:51--- <<-- nrh1703 [] has quit (Connection reset by peer)
16:52--- ---> nrh1703 [] has joined #mythtv
17:04--- <<-- Chang [] has quit (Remote closed the connection)
17:15--- <<-- _Nero_ [~Nero__@mdobossy.student.Princeton.EDU] has quit ()
17:31--- ---> broug [] has joined #mythtv
17:31broug sorry was looking for some help, I'll move over to the users group
17:42--- ---> Netslayer [] has joined #mythtv
17:42--- <<-- schultmc [] has quit ("Client exiting")
17:45--- ---> AridWork [] has joined #mythtv
17:50--- <<-- Captain_Murdoch_ [] has quit ()
17:54--- <<-- Baylink [] has quit ("User disconnected")
17:58--- ---> Baylink [] has joined #mythtv
17:59--- <<-- Baylink [] has quit (Remote closed the connection)
18:04--- ---> Aridwrk [] has joined #mythtv
18:06gfiend Chutt heya
18:07--- User: *** gfiend is now known as GeckoFiend
18:14--- ---> choenig [] has joined #mythtv
18:19--- ---> lisadaveh [] has joined #mythtv
18:22--- <--- broug [] has left #mythtv ()
18:23--- <<-- choenig [] has quit (Remote closed the connection)
18:23--- <<-- AridWork [] has quit (Read error: 110 (Connection timed out))
18:25--- <<-- zibbe [] has quit (Connection timed out)
18:40--- <<-- Netslayer [] has quit (Remote closed the connection)
18:47--- <<-- bline [] has quit (Remote closed the connection)
18:51--- ---> tmk [] has joined #mythtv
18:51tmk you here chutt?
18:54--- ---> spyder__ [] has joined #mythtv
18:56--- ---> bline [] has joined #mythtv
18:56--- ---> _Nero_ [~Nero__@mdobossy.student.Princeton.EDU] has joined #mythtv
18:56tmk sup bline
19:02--- ---> zibbe [] has joined #mythtv
19:09--- <<-- spyder_ [~spyder@] has quit (Read error: 110 (Connection timed out))
19:20--- ---> cmorgan [] has joined #mythtv
19:30--- ---> mchou [] has joined #mythtv
19:31--- ---> lisadaveh_ [] has joined #mythtv
19:31mchou has the database structure changed between 0.15.1 and current cvs?
19:33Chutt tmk, yo
19:33tmk heya
19:33Chutt mchou, of course it has
19:33Chutt gfiend, that list looks nice, thanks for taking the time to do that =)
19:33Chutt tmk, what's up? =)
19:33tmk i've been having some issues with my backends crashing.. was wondering if you prefer backtraces from cvs or release versions
19:34Chutt from cvs
19:34tmk k
19:34Chutt current cvs, too.. =)
19:34tmk yea
19:34--- <<-- lisadaveh_ [] has quit (Client Quit)
19:34tmk you know how i feel about compiling myth
19:35Chutt yeah
19:35tmk it's even worse lately.. i have 128mb of ram, and it takes like 3 hours to compile due to some swapping issue on one file
19:35Chutt actually, this a multiple backend problem?
19:35tmk yea
19:35Chutt same thing as you mentioned a long time ago?
19:35Chutt crashes after a disconnect?
19:36tmk probably
19:36tmk err no
19:36tmk when i watch something in-progress
19:36tmk on the remote-backend
19:36tmk the master usually dies
19:36mchou tmk, what happens? I'm also experienceing multiple backend problems......
19:36Chutt heh
19:36Chutt well, give current cvs a try
19:37Chutt i've fixed a few long-standing issues with socket disconnects
19:37mchou Do you get a QStringList errot of some kind?
19:37Chutt though, i might have missed the slave backend case
19:37mchou error*
19:37tmk i'll try current before i whine
19:37tmk now that my wife uses it i have to be more careful
19:37tmk :)
19:37josephk_ you don't happen to be be pausing around the time the recording finishes do you?...hehe
19:38tmk i had an issue with quality settings though, while i'm ranting
19:38mchou I think wtching live tv on slave backend causes master to crash, but I don't really know where the problem is.....
19:38tmk full-size recordings would be 480x480 with a black bar on the right-side filling in the rest of the 720
19:38tmk i think it's probably ivtv's fault though
19:38cmorgan most of my crashes are when watching a video and having something recording at the same time
19:38tmk yea
19:39tmk sounds like my problem
19:39mchou cmorgan: yeah, that too :)
19:39cmorgan it usually just goes to a blank screen
19:39cmorgan tv loses signal etc
19:39josephk_ more than 2 cards?
19:39cmorgan i've just got a single card
19:39tmk i have 2 in my master and one in my slave
19:40mchou tmk, what version are you using?
19:40tmk old cvs
19:40mchou ok, from when?
19:40tmk 0.15.20040528-1
19:42Chutt that's not a version number :p
19:42cmorgan hah
19:42tmk [tmk@localhost utils]$ mythfrontend --version
19:42tmk 0.15.20040528-1
19:42tmk <shrug>
19:42Chutt tmk, did you like that thread on ivtv-dev where the guy was trying to tell me about how DMA worked on the harddrive? =)
19:42Chutt yeah, it's reported as a version number, but it's really just for binary compatability with libmyth
19:42tmk Chutt: i saw that
19:42tmk i stayed otu of it
19:42tmk :)
19:43cmorgan when was this?
19:43tmk last few days
19:43* cmorgan is looking in the archives...
19:43tmk on ivtv-dev list
19:43tmk i alwasy read mail from a few people
19:43tmk chutt is one of them
19:43tmk Chutt: where should i look for a real version #
19:44Chutt there isn't one :(
19:44tmk heh. sounds like old ivtv :P
19:44Chutt i really should have something auto-incremement each cvs checkin
19:44Chutt but i'm too lazy to write the script to do that
19:44tmk well it at least ties the cvs down to a date
19:45mchou can't you tell from the timestamps on d/l cvs files?
19:45Chutt tml, please try to get a backtrace of your crash, though, i should be able to fix it easily now, if i know where it's getting through
19:45tmk k
19:45tmk i'll upgrade and with luck the problem will be gone
19:45Chutt it's a silly qsocket not being threadsafe issue
19:45* cmorgan can't find which one of the amusing threads it is... ;-) oh well
19:46dc_ can anyone recommend a decent backup method for linux?
19:47tmk my wife's out of town this weekend.. i'll probably upgrade then
19:47tmk soccer time
19:47tmk see you folks later
19:47mchou tmk: back up your db before upgrade :)
19:47tmk oh?
19:48tmk myth's good about conversions in my experience
19:48mchou just in case you need to revert :)
19:48tmk there is no revert
19:48tmk it takes too long to compile
19:49Chutt =)
19:49mchou heh.
19:49--- ---> gentoouser [] has joined #mythtv
19:49--- <<-- tmk [] has quit ("Client exiting")
19:57--- <<-- mecraw [~lmarlow@] has quit (Read error: 104 (Connection reset by peer))
19:57--- <<-- Ashlar [] has quit (Read error: 110 (Connection timed out))
19:58--- ---> mecraw_ [~lmarlow@] has joined #mythtv
20:10GeckoFiend bleh this whole Compuserve contract has me so nervous I can't even write code. I *really* want this contract and I have to wait till Friday for the first (phone) interview.
20:14GeckoFiend So it's cool with you if I tell them I designed all of Myth right Chutt? ;)
20:15Chutt hehe
20:24--- <<-- dc_ [] has quit ("sleep")
20:28jams GeckoFiend, finding a box that's "shipping quality" is proving to be difficult
20:29--- ---> Netslayer [] has joined #mythtv
20:32--- <<-- mecraw_ [~lmarlow@] has quit ("Trillian (")
20:54GeckoFiend 99% of that is drivers, and unfortunatly you can't get solid video and capture it seems, least not in my experiance
21:01GeckoFiend hrm current CVS doesn't seem to listen when you say "no I want to watch TV"
21:02jams heh i meant a cardboard box for the 250 :)
21:05--- <<-- lisadaveh [] has quit (Remote closed the connection)
21:05* Captain_Murd thinks "that's quite a bit of stuff in WhatsNew"
21:10--- <<-- m0j0_ [] has quit (Read error: 110 (Connection timed out))
21:11--- ---> m0j0 [] has joined #mythtv
21:20--- ---> lisadaveh [] has joined #mythtv
21:36--- <<-- lisadaveh [] has quit (Remote closed the connection)
21:57--- Netsplit <-> quits: cmorgan, poptix, spyder__
21:58--- Netsplit over, joins: spyder__, cmorgan
22:02--- ---> poptix [] has joined #mythtv
22:18--- ---> uierh4 [] has joined #mythtv
22:18--- <--- uierh4 [] has left #mythtv ()
22:23--- <<-- _shad [~shad@] has quit (Read error: 104 (Connection reset by peer))
22:23gentoouser thx to whoever fixed the black screen, I have another prob in cvs, it happens with nvidiaglx, if I open a recording,close,open again mythfrontend crashes, if I do the same with livetv it doesnt, I think its in the exit of watchrecording&prerecorded not releasing/resetting the video card port? is the livetv & watchrecord exit code different that would cause it to happen in one& not the other
22:34--- <--- gentoouser [] has left #mythtv ("Ooh, look, a shiny object...")
22:41--- ---> toplight3 [] has joined #mythtv
22:44toplight3 i realize this isn't support, but i think this is a bug, not a support question
22:44toplight3 mythbackend seems to segfault out when recording 2 shows at once; not all the time, however
22:45toplight3 2 bttv cards, p1.8 running 2.6.7 kernel - also happened with 2.4.20
22:45toplight3 when it does happen, myth records just a few seconds of both program and then dies - as it dies, mythbackend.log gets this:
22:45toplight3 2004-08-31 19:00:28 Changing from None to RecordingOnly ... strange error flushing buffer ... VIDIOCSYNC: Input/output error
22:45toplight3 im sure this isn't intended behaviour ;-) ... what can i do to further help in the diagnosis of the problem?
22:46thor_ without a backtrace, it is generally very hard to guess what the problem is
22:47thor_ if I had to guess, I would hazard a driver problem (VIDIOCSYNC), but that would be just a guess
22:48toplight3 i should recompile with debug support and run under gdb then?
22:48thor_ that would be good
22:48toplight3 alright, will do - do i post the trace to the mailing list? seems rather long to post one there
22:49thor_ yup
22:49toplight3 alright, thy will be done sir
22:49thor_ instructions:
22:50toplight3 thanks, i see those... just wish i could make it hurry up and crash
22:50thor_ heh
22:54--- <<-- cmorgan [] has quit (Remote closed the connection)
23:05--- <<-- nrh1703 [] has quit (Read error: 104 (Connection reset by peer))
23:06--- ---> nrh1703 [] has joined #mythtv
23:06splat1 nn all
23:07--- User: *** splat1 is now known as splAt1
23:08toplight3 will the debug-enabled code run slower? my p4/1.8 is pretty maxed doing 2 bttv encodes at the same time
23:09--- ---> jf100 [] has joined #mythtv
23:13--- ---> lisadaveh [] has joined #mythtv
23:14--- ---> eco_ [] has joined #mythtv
23:15--- <<-- eco [] has quit (Read error: 104 (Connection reset by peer))
23:17--- <--- jf100 [] has left #mythtv ()
23:20--- ---> _shad [] has joined #mythtv
23:28toplight3 im getting aLOT of this:
23:28toplight3 Program received signal SIG32, Real-time event 32.
23:29toplight3 i had to add "handle SIG32 nostop" to get the program to run correctly
23:34toplight3 yeah okay, here we go
23:34toplight3 Cannot remove breakpoints because program is no longer writable.
23:34toplight3 It might be running in another process.
23:34toplight3 Further execution is probably impossible.
23:34toplight3 0x413402e1 in sigpending () from /lib/
23:34toplight3 (gdb) thread apply all bt full
23:34toplight3 .... now what
23:38--- ---> spyder_ [] has joined #mythtv
23:41Chutt means your gdb is broken.
23:45Chutt captain_murdoch, unitialized var in decoderbase.cpp - fixed in cvs
23:45Captain_Murdo| ok :(
23:45Chutt easy to miss =)
23:46Chutt current_aspect isn't set in nuppeldecoder, as it's always 4:3
23:46Chutt but it's _always_ set in avformatdecoder
23:46Captain_Murdo| the current_* ones or which was causing it? I don't think I removed them from the original *decoder.cpp files.
23:46Captain_Murdo| ahh
23:46* Captain_Murd goes back and looks at the original diff to see how he missed those.
23:47Chutt they're never directly initialized
23:47Chutt in the constructor, at least
23:47Captain_Murdo| I was just looking at the aspect code since Bruce said it started working when he changed aspect ratio.
23:47Chutt yeah, that's what i was doing =)
23:47Chutt didn't see how it could be causing that, unless the value was completely wrong
23:47Chutt and, well, it was =)
23:48toplight3 great, now i need to: kwenda@blaster: gdb gdb
23:48toplight3 ;-)
23:50Captain_Murdo| ahh, so the m_parent->SetVideoParams() call in DecoderBase::SyncPositionMap was setting an undefined aspect ratio since it was never initialized.
23:50Chutt exactly
23:50Chutt the old code just used the default value
23:50Chutt of 1.3333
23:51Chutt that's the only value that needed initialized, but i stuck the others in for good measure
23:51Captain_Murdo| weird that I didn't run into that since I tested on a few nuppel files. I'm pretty sure I did at least. I have 2-3 around that I keep for testing.
23:52Chutt it ran fine for me, too
23:52Chutt <shrug>
23:52Chutt anyway, it's fixed =)
23:52Chutt well, should be at least
23:53Captain_Murdo| setvideoparams checks to see if it's positive or else it doesn't use the value, that must be why it didn't fail for us. it was negative and ignored.
23:53Captain_Murdo| ditto on glad it's found and fixed.
23:55--- <<-- spyder__ [] has quit (Read error: 110 (Connection timed out))
23:55Captain_Murdo| I was about to make a joke yesterday about rewriting all our sql queries in xml and somebody suggested it today. :)
23:55Chutt yeah, that was hilarious
23:57--- ---> Viddy [] has joined #mythtv
23:57toplight3 thanks Chutt, i got a new gdb and all the SIG32 madness stopped... now to just make mythbackend crash again...
23:58Chutt that 'VIDIOCSYNC: Input/output error' indicates that it's probably a driver issue
23:58Chutt or myth isn't handling the driver choking properly
---Logclosed Wed Sep 01 00:00:27 2004