Back to Home / #mythtv / 2007 / 01 / Prev Day | Next Day
#mythtv IRC Logs for 2007-01-25

---Logopened Thu Jan 25 00:00:37 2007
---Daychanged Thu Jan 25 2007
00:00[~]Captain_Murdoch applies xris' indexes to his dev database for kicks, even though they'll be wiped out at 4am when the nightly restore from production happens.
00:00<xris>heh
00:01<xris>odd stuff, too. scheduler query shows as running in about 1 second in the mysql logs, but mythtv says it's upwards of 6.
00:01<Captain_Murdoch>I wonder how many users would see that as a page of people they could email with their "would you take $20 to add feature blah" emails.
00:01<xris>Captain_Murdoch: yeah, that's a problem.
00:02<xris>I'm just sick of getting messages like "hi, I need someone to write my company a cool mythtv plugin, but it can't be GPL."
00:02<xris>or even without that last bit.
00:03<xris>lot of mythtv queries that don't touch index files, but so many of them are for tables that generally have like 2 records in them.
00:03<Captain_Murdoch>:) I've only been contacted by 3-4 different companies that I recall.
00:03<Captain_Murdoch>yeah, I don't see any problems with the ones you posted. those are small tables so there isn't much overhead in maintaining those extra indexes.
00:03<xris>yeah. the two big ones I keep running into don't seem to be fixable by adding indexes.
00:04<Captain_Murdoch>I was curious what that recgroup one was, didn't remember that off the top of my head.
00:04<xris>mythtv finds plenty of possible indexes, but doesn't use them
00:04<xris>pm
00:04<Captain_Murdoch>thanks. I figured it was LiveTV but couldn't recall exactly.
00:05<Captain_Murdoch>probably the expirer's query looking for short LiveTV programs that it can clean up.
00:05<xris>could be
00:05<xris>no clue. I just pull them out of the log
00:06<xris>jobqueue query is the one that shows up the most, but since it's pulling all rows, mysql won't even bother trying an index.
00:06<Captain_Murdoch>if the indexes speed up any at all it is a help because it keeps us from locking certain tables for as long since we're using myisam
00:07<xris>I don't think I fixed anything that takes more than 1/2 second to execute.
00:07<Captain_Murdoch>1/2 can be a lifetime if you're waiting to set a bookmark or something. :)
00:08<xris>true
00:08<Captain_Murdoch>will help with UI responsiveness possibly.
00:09<xris>possibly
00:11<xris>the worst of the queries (aside from scheduler) seem to be for temp tables created by mfdb, and large selects/deletes (which are always going to be slow)
00:16<xris>what I *really* need is something that will log queries that could BENEFIT from an index... I don't need to see things like "SELECT * FROM x" that would never touch an index no matter what.
00:19<Captain_Murdoch>yeah, AI
00:21<Captain_Murdoch>have to rely on grep or egrep
00:29|-|spacecoaster [n=Miranda@c-75-69-133-78.hsd1.nh.comcast.net] has quit [Read error: 131 (Connection reset by peer)]
00:50|-|s34n [n=chatzill@ip-206-159-190-125.mvdsl.com] has joined #mythtv
00:50|-|s34n [n=chatzill@ip-206-159-190-125.mvdsl.com] has left #mythtv []
01:13|-|gnome42 [n=obi@dsl-142-149.aei.ca] has quit [Remote closed the connection]
01:19|-|doc|home [n=doc@skynet.skynet.ie] has joined #mythtv
01:29|-|flatronf701B [n=flatronf@202.75.186.154] has quit [Read error: 113 (No route to host)]
01:57|-|doc|home [n=doc@skynet.skynet.ie] has quit ["Lost terminal"]
02:06|-|tomimo [n=kurre@a80-186-74-125.elisa-laajakaista.fi] has quit [Remote closed the connection]
02:53|-|xris [n=xris@xris.forevermore.net] has quit ["Leaving."]
03:24|-|eskil [n=eskil@adsl-66-120-85-205.dsl.snfc21.pacbell.net] has quit ["Leaving"]
03:32|-|slaine_ [n=glengray@84.203.137.218] has joined #mythtv
03:36|-|stuarta [n=stuart@unaffiliated/stuarta] has joined #mythtv
03:56|-|amccarthy [n=amccarth@cust0841.qld01.dataco.com.au] has joined #mythtv
04:04|-|ernst [n=ernst@elanser.demon.nl] has joined #mythtv
04:11|-|ernst [n=ernst@elanser.demon.nl] has quit ["Weggaan uit"]
04:16<stuarta>janneg: you copied that file off svn.mythtv.org?
05:01|-|Skiingsean changed nick to Skiingsean_out
05:02|-|adante [n=adante@203-206-83-89.dyn.iinet.net.au] has joined #mythtv
05:43|-|amccarthy [n=amccarth@cust0841.qld01.dataco.com.au] has quit [Remote closed the connection]
05:44|-|amccarthy [n=amccarth@cust0841.qld01.dataco.com.au] has joined #mythtv
05:46|-|purserj [n=purserj@k-sit.com] has joined #mythtv
05:50|-|adante_ [n=adante@203-206-97-160.dyn.iinet.net.au] has joined #mythtv
05:52|-|MrGandalf [n=buechlmr@cpe-72-225-32-214.rochester.res.rr.com] has quit ["work"]
06:06<janneg>stuarta: yes. the error remains with my synced version and ffmpeg HEAD doesn't like the file either
06:06|-|adante [n=adante@203-206-83-89.dyn.iinet.net.au] has quit [Read error: 113 (No route to host)]
06:06|-|adante_ changed nick to adante
06:06<janneg>though it gives a different error "first frame is not a key frame"
06:07<stuarta>okay, i've given you the ticket for now, i'll go remove the file from the server
06:09|-|DrNickRiviera [n=riviera@so-5494-x0.essex.ac.uk] has joined #mythtv
06:15<gbee>janneg: I didn't think there had been any related changes since the version you synced with
06:16<gbee>changes in ffmpeg I mean
06:17<gbee>ahh, nevermind just looked at the commits since - two involve mpegvideo.c
06:21<gbee>7686 would be my guess
06:25<janneg>gbee: no, it's a h264 video stream in mpegts. and there are no related changes. I haven't tried ffmpeg of the synced version yet
06:27<gbee>sorry, I must be thinking of a different error to the one you two were talking about
06:28|-|bruno_ [n=bruno@bigsa-mont-01.big.ulg.ac.be] has joined #mythtv
06:28<gbee>thought you meant the MBAFF error
06:29<janneg>yes, "[h264 @ 0xb72e2168]MBAFF + spatial direct mode is not implemented"
06:29|-|bruno_ [n=bruno@bigsa-mont-01.big.ulg.ac.be] has left #mythtv ["Konversation terminated!"]
06:29<gbee>so that's fixed in ffmpeg, yet still appears with mythtv???
06:30<stuarta>didn't occur for me with the command line ffplay
06:30<gbee>oh well, I should get back to work
06:30<stuarta>not that i could see the video output due to not being in front of the screen
06:31<gbee>stuarta: it occurred for me with an earlier version of ffmpeg, but not with HEAD
06:31<janneg>I got many "first frame is not a key frame" errors
06:31<stuarta>gbee: same.
06:31<stuarta>janneg: same
06:31<gbee>yep, still getting those
06:31<stuarta>which is an improvement :)
06:32<janneg>I'm not so sure.
07:16|-|adante_ [n=adante@203-206-97-160.dyn.iinet.net.au] has joined #mythtv
07:21|-|adante__ [n=adante@203-206-97-160.dyn.iinet.net.au] has joined #mythtv
07:32|-|adante [n=adante@203-206-97-160.dyn.iinet.net.au] has quit [Read error: 110 (Connection timed out)]
07:35|-|amccarthy [n=amccarth@cust0841.qld01.dataco.com.au] has quit ["Konversation terminated!"]
07:37|-|adante_ [n=adante@203-206-97-160.dyn.iinet.net.au] has quit [Read error: 110 (Connection timed out)]
07:54|-|MrGandalf [n=mgandalf@cpe-72-225-37-244.rochester.res.rr.com] has joined #mythtv
08:24|-|Cardoe [n=cardoe@gentoo/developer/Cardoe] has joined #mythtv
08:24|-|Cardoe [n=cardoe@gentoo/developer/Cardoe] has quit [Client Quit]
08:25|-|Cardoe [n=cardoe@gentoo/developer/Cardoe] has joined #mythtv
08:30|-|slaine_ [n=glengray@84.203.137.218] has quit []
08:41|-|slaine_ [n=glengray@84.203.137.218] has joined #mythtv
09:18|-|heanol [i=henke@ua-83-227-168-197.cust.bredbandsbolaget.se] has quit [Remote closed the connection]
09:18|-|heanol [i=henke@ua-83-227-168-197.cust.bredbandsbolaget.se] has joined #mythtv
09:53|-|jmk_ [n=jmk@pat.foofus.net] has joined #mythtv
10:21|-|jgarvey [n=jgarvey@cpe-075-177-158-190.nc.res.rr.com] has joined #mythtv
10:28|-|russellb [i=russellb@nat/digium/x-9cd66ad3ae0848e3] has joined #mythtv
10:29|-|slaine_ [n=glengray@84.203.137.218] has quit []
10:47|-|gnome42 [n=obi@dsl-139-129.aei.ca] has joined #mythtv
11:02|-|xris [n=xris@xris.forevermore.net] has joined #mythtv
11:12|-|stuarta [n=stuart@unaffiliated/stuarta] has quit ["leaving"]
11:15|-|MoRpHeUz [n=morphbr@201.23.96.2.rpo.directnet.com.br] has joined #mythtv
11:27<xris>janneg: you did an ffmpeg resync recently, right? did it get committed?
11:29<Chutt>i'm looking it over.
11:30<xris>cool
11:30<xris>found some weirdness with h.264 playback (also happens in xine and even worse in mplayer), but ffplay works fine.
11:30<xris>hoping that the resync will fix things for myth
11:32<janneg>xris: which revision of ffplay worked?
11:34<janneg>xine and mplayer builds are older than the ffmpeg?
11:34<xris>janneg: 7677
11:34<xris>and yeah, xine and mplayer are from livna.. no idea how they do their ffmpeg linking.
11:34<xris>on phone..
11:35<gnome42>janneg: hi! I found the bug in that last patch of yours that confused me so much :)
11:37<janneg>gnome42: it wasn't the thing I saw? missing i++; in the last else?
11:39<xris>oh well.. no work for me at google. they want a full-time coder, and I'm just not interested in that anymore.
11:39<gnome42>janneg: it was that plus antoher one. We weren't hitting the else if (k>0) needed to be if (k>0x01)
11:39<gnome42>janneg: http://pastebin.ca/328267 results
11:40<gnome42>janneg: your last one is labeled 3-byte'ish
11:40<gnome42>janneg: and the last to entries in the list are just to show the dif in 1-byte vs. 3-byte
11:40<gnome42>s/to/two/
11:42|-|MrGandalf [n=mgandalf@cpe-72-225-37-244.rochester.res.rr.com] has quit [Remote closed the connection]
11:47|-|Hamstaman [n=hamsta@p57A4E156.dip.t-dialin.net] has joined #mythtv
12:12|-|Charbal_ changed nick to Charbal
12:26|-|xris [n=xris@xris.forevermore.net] has quit ["Leaving."]
12:28|-|Netsplit calvino.freenode.net <-> irc.freenode.net quits: Agrajag-, jams, packetscan, noddan, Nem^, rn114, lsobral
12:31|-|MrGandalf [n=mgandalf@cpe-72-225-37-244.rochester.res.rr.com] has joined #mythtv
12:33|-|Netsplit over, joins: jams
12:35|-|Nem^ [n=Nem@dslb-084-056-235-013.pools.arcor-ip.net] has joined #mythtv
12:35|-|noddan [n=noddan@noddan-birger.brj.sgsnet.se] has joined #mythtv
12:35|-|lsobral [n=sobral@200.184.118.132] has joined #mythtv
12:35|-|rn114 [n=rn114@82-46-18-118.cable.ubr02.bath.blueyonder.co.uk] has joined #mythtv
12:35|-|packetscan [i=Tsu@srv01.packetscan.com] has joined #mythtv
12:35|-|Agrajag- [n=filip@c211-30-4-5.artrmn1.nsw.optusnet.com.au] has joined #mythtv
12:39|-|DogWater [i=DogWater@dhcp91.cmh.ee.net] has joined #mythtv
12:39|-|FreddyC [n=IceChat7@64.130.105.111] has joined #mythtv
12:39<DogWater>anyone know if it is possible to attach a usb tv tuner to the ps3 and use the ps3 for mythtv? :D
12:39<FreddyC>hey
12:39|-|PointyPumper [i=Pintlezz@OL221-67.fibertel.com.ar] has quit [Read error: 104 (Connection reset by peer)]
12:39<FreddyC>wow. looks like i found one
12:41<FreddyC>i had some trouble with setting up my myth tv on ubuntu 6.06 . i got errors saying that i could not access the db
12:41<FreddyC>i tried running as root and still was denied
12:41<FreddyC>i dont understand why
12:42<gnome42>FreddyC & DogWater: this is the dev channel, try #mythtv-users :)
12:44<FreddyC>thanks
12:44<FreddyC>just saw the pm
12:44<FreddyC>moving there now
12:59|-|MrGandalf [n=mgandalf@cpe-72-225-37-244.rochester.res.rr.com] has quit [Remote closed the connection]
13:01|-|MrGandalf [n=mgandalf@cpe-72-225-37-244.rochester.res.rr.com] has joined #mythtv
13:04|-|FreddyC [n=IceChat7@64.130.105.111] has quit ["Depression is merely anger without enthusiasm"]
13:14<j-rod>janneg: I've got the name and email address of the guy at NXP who is supposedly the software manager for the saa7162 driver...
13:14<j-rod>just sent him an email
13:16|-|xris [n=xris@dsl081-161-160.sea1.dsl.speakeasy.net] has joined #mythtv
13:29<janneg>j-rod: good luck
13:30|-|stuarta [n=stuarta@unaffiliated/stuarta] has joined #mythtv
13:41<janneg>stuarta: "first frame is not a key frame" is a bad fix for bad h264 streams
13:41<stuarta>they bodged it?
13:48<janneg>yeah. http://thread.gmane.org/gmane.comp.video.ffmpeg.devel/42627/focus=43183
13:56|-|aevil [n=aevil@i59F551A0.versanet.de] has joined #mythtv
14:05|-|rikstah [n=rick@host81-155-179-213.range81-155.btcentralplus.com] has joined #mythtv
14:09|-|DogWater [i=DogWater@dhcp91.cmh.ee.net] has left #mythtv []
14:21|-|melunko [n=dapper@addr-213-216-194-138.suomi.net] has joined #mythtv
14:21|-|zigovr3 [n=zig@AAubervilliers-153-1-60-250.w81-249.abo.wanadoo.fr] has quit [Read error: 110 (Connection timed out)]
14:22|-|zigovr3 [n=zig@AAubervilliers-153-1-85-237.w86-212.abo.wanadoo.fr] has joined #mythtv
14:25|-|PointyPumper [i=Pintlezz@OL221-67.fibertel.com.ar] has joined #mythtv
14:31|-|stuarta [n=stuarta@unaffiliated/stuarta] has quit ["places to be"]
14:35|-|amccarthy [n=amccarth@cust0841.qld01.dataco.com.au] has joined #mythtv
14:46|-|jbinder [n=jared@ool-44c2f05c.dyn.optonline.net] has joined #mythtv
14:46|-|tonyb2006 [n=tonyb@ool-45712c93.dyn.optonline.net] has joined #mythtv
14:46|-|tonyb2006 [n=tonyb@ool-45712c93.dyn.optonline.net] has left #mythtv []
14:47|-|jbinder [n=jared@ool-44c2f05c.dyn.optonline.net] has left #mythtv []
14:48|-|jbinder [n=jared@ool-44c2f05c.dyn.optonline.net] has joined #mythtv
14:48|-|DrNickRiviera [n=riviera@so-5494-x0.essex.ac.uk] has quit ["Leaving."]
14:48|-|jbinder [n=jared@ool-44c2f05c.dyn.optonline.net] has left #mythtv []
15:04|-|Hambone [n=chatzill@fwiva.spro.net] has joined #mythtv
15:05|-|Hambone [n=chatzill@fwiva.spro.net] has left #mythtv []
15:15|-|slaine_ [n=glengray@89.100.76.74] has joined #mythtv
15:16|-|flatronf701B [n=flatronf@202.75.186.154] has joined #mythtv
15:17|-|Nem^ [n=Nem@dslb-084-056-235-013.pools.arcor-ip.net] has quit [Remote closed the connection]
15:18|-|tzanger [n=tzanger@208.68.91.47] has joined #mythtv
15:19|-|tzanger [n=tzanger@208.68.91.47] has left #mythtv []
15:20|-|eskil [n=eskil@natint3.juniper.net] has joined #mythtv
15:22|-|melunko [n=dapper@addr-213-216-194-138.suomi.net] has quit ["Leaving"]
15:35|-|MrGandalf [n=mgandalf@cpe-72-225-37-244.rochester.res.rr.com] has quit ["Leaving"]
15:38|-|MoRpHeUz [n=morphbr@201.23.96.2.rpo.directnet.com.br] has quit ["Leaving..."]
15:39|-|tzanger [n=tzanger@208.68.91.47] has joined #mythtv
15:39<tzanger>what's the recommended way to build a backend-only system these days?
15:40<tzanger>It's building all kinds of Qt and X apps even with --disable-everything (x11, xv, xrandr, etc.)
15:40|-|mark488 [n=ms@h-66-167-77-99.snvacaid.dynamic.covad.net] has joined #mythtv
15:41<gbee>tzanger: backend requires both and you should be in #mythtv-users
15:41<tzanger>gbee: I am, but since I'm trying to mangle the source, I came here too
15:41|-|mark488 [n=ms@h-66-167-77-99.snvacaid.dynamic.covad.net] has left #mythtv []
15:42<gbee>if your question relates to writing code, here is fine, otherwise -users :)
15:43<tzanger>ok
15:53[~]xris wonders if the topic should be changed to clarify "about the code" instead of "about compiling"
15:56<tzanger>well I wouldn't mind helping provide patches to mythtv-0.20 so --without-frontend actually builds (it seems it's missing -L/usr/X11R6/lib in a few places, so far libmyth and libmythui)
15:57<gbee>--without-frontend is there for packagers only, it's not meant to be used by anyone else
15:57<eskil>xris, I think a bot would be better. Any sentence with a tailing ? get's flagged. If noone writes the person back with 5 mins, tell them to go to the -users. It'll be correct in 90% of the cases...
15:57<eskil>(unfortnuately)
15:59<gbee>did Daniel not backport his removal of --without-frontend from configure --help to -fixes?
15:59<kvandivo>the msg sent upon entry specifically states: Just compiling mythtv isn't considered development.
15:59<janneg>gbee: it's only intended for porting to new archs
16:00<gbee>janneg: that's what I meant, just worded it badly ;)
16:00<tzanger>I guess... seems odd that a backend would need to have X, Qt, etc.
16:00<tzanger>I admit I'm not an expert on Myth though
16:00<gbee>QT is *NOT* just GUI
16:01<tzanger>gbee: I understand that, I do write some minimal qtopia based stuff
16:01<gbee>I'll grant you that X is a tiny bit harder to justify, but it's there for mythtv-setup, which is a backend only app
16:02<tzanger>*nods*
16:02<tzanger>s'alright I'm just being stubborn
16:02<tzanger>common for a taurus like me :-)
16:03<gbee>QT is used throught mythtv for arrays, file access, threading, sockets etc :)
16:03<tzanger>*nods*
16:03<tzanger>beats the shit out of using boost
16:05<gbee>if we seem a little short tempered, it's only because we get these questions a lot
16:05<tzanger>oh I totally understand
16:05<tzanger>I field some support calls myself here at work
16:06<tzanger>generally when I'm up to my eyeballs in development work
16:07<xris>eskil: my point was to prevent people from asking, since it's annoying to respond 500 times/day
16:08<gbee>xris: don't exaggerate
16:08<gbee>450 tops
16:08<xris>lol. ok, I'll give you that. :)
16:10<gbee>problem is that most people don't read the topic at all, but even if it stops one person that's progress
16:10<tzanger>I did :-)
16:11<eskil>xris, my point was trying to be funny, but I'm not always good at that (and I like bots that annoy people for no apparant reason)
16:12[~]briand grins.
16:14<briand>eskil: of course, carrying that just a tad further, if your hypothetical bot were to simply /kick folks off channel upon /join, it'd also be correct 90% of the time.
16:14<eskil>briand, and fire off a "yo mama" joke before kicking them - it's the only way to be sure.
16:16<briand>heheh
16:17<briand>well, i guess my cat is not *quite* the hunter I thought she was... I cleaned up a little puddle of blood on the kitchen counter yesterday afternoon. I figured she had found and dispatched one of those little lizards that get everywhere down here...
16:17<briand>this afternoon, there's another little puddle of blood...
16:18<briand>looking up from the counter, I now understand. I need to teach my girlfriends how to operate a corkscrew, and it shouldn't go completely through the cork.
16:18<briand>my beaujolais noveau is slowly dripping out onto the counter...
16:19<briand>hmm... or, maybe we should just drink the stuff in larger quantities...
16:20|-|Nem^ [n=Nem@dslb-084-056-235-013.pools.arcor-ip.net] has joined #mythtv
16:21|-|slaine_ [n=glengray@89.100.76.74] has quit []
16:25|-|aevil [n=aevil@i59F551A0.versanet.de] has quit ["Verlassend"]
16:52|-|Cardoe [n=cardoe@gentoo/developer/Cardoe] has quit ["Leaving"]
16:53|-|adante [n=adante@203-206-83-89.dyn.iinet.net.au] has joined #mythtv
16:57|-|_nero_ [n=nero@unaffiliated/nero] has joined #mythtv
17:08|-|adante__ [n=adante@203-206-97-160.dyn.iinet.net.au] has quit [Read error: 110 (Connection timed out)]
17:10|-|jgarvey [n=jgarvey@cpe-075-177-158-190.nc.res.rr.com] has quit ["Leaving"]
17:16|-|adante [n=adante@203-206-83-89.dyn.iinet.net.au] has quit ["Lost terminal"]
17:18|-|adante [n=adante@203-206-83-89.dyn.iinet.net.au] has joined #mythtv
17:22|-|jmk_ [n=jmk@pat.foofus.net] has quit ["Leaving"]
17:46|-|DrNickRiviera [n=riviera@so-5494-x0.essex.ac.uk] has joined #mythtv
17:48|-|LLyric [n=simon@d220-238-228-208.dsl.vic.optusnet.com.au] has joined #mythtv
18:16|-|logost [n=aolaol@ip68-107-165-193.lu.dl.cox.net] has joined #mythtv
18:19|-|Wiedi [n=wiedi@p549C775E.dip.t-dialin.net] has joined #mythtv
18:29|-|rikstah [n=rick@host81-155-179-213.range81-155.btcentralplus.com] has left #mythtv ["Leaving"]
18:54|-|Wiedi [n=wiedi@p549C775E.dip.t-dialin.net] has quit ["Leaving"]
18:57|-|monkeyBox [n=bdavis@cpe-76-185-4-217.tx.res.rr.com] has joined #mythtv
18:57<monkeyBox>Does anyone here have any experience w/ the pcHDTV card & QAM?
18:58<monkeyBox>I'm not sure how to tell for sure if my cable company broadcasts QAM (time warner in dallas-area, TX)
18:58<Chutt>topic.
18:58<monkeyBox>oops sorry.
18:59|-|monkeyBox [n=bdavis@cpe-76-185-4-217.tx.res.rr.com] has left #mythtv ["Leaving"]
19:13|-|MrGandalf [n=buechlmr@cpe-72-225-32-214.rochester.res.rr.com] has joined #mythtv
19:14<MrGandalf>Hmm, looks I just lost 4 HD channels to h264 all at once.. time to upgrade to a faster CPU I think - buts how fast I need I don't know.
19:18<briand>it looks like xris has been playing around with the mysqlar application a bit.. :)
19:19<xris>briand: yeah. tried to thank you for that pointer, but you haven't been around. :) so thanks.
19:19<briand>you're quite welcome. :)
19:20<xris>haven't actually used it for mythtv, but it's been helpful for finding new tweaking methods.
19:20<xris>I think I want to rewrite it, though.. for one, it doesn't stay connected to mysql5.
19:20<briand>...and, you can thank either a bunch of hackers and/or comcast for my not being around.
19:21<briand>the hackers for finding a DoS attack for my ancient Surfboard SB4100, then comcast for dragging their feet getting me a solution (the first replacement they sent was dead out of the box!)
19:22<briand>hmm. i wonder why it doesn't like mysql5... isn't it simply logging into the db and grabbing stats periodically? It shouldn't require a persistent login to the db...
19:24<janneg>MrGandalf: athlon64 2.6ghz would be enough for my h264 channels
19:24|-|Anduin [n=awithers@adsl-69-110-40-109.dsl.pltn13.pacbell.net] has quit ["Leaving."]
19:25<MrGandalf>janneg: what res HD?
19:25<xris>briand: there seems to be change in how mysql5 handles activity timeouts.. Beirdo's bot has the same issue. it gets hung up on, but doesn't realize that it has, so it never tries to reconnect.
19:25<xris>I think it stays logged in, thgouh
19:25<janneg>at least after I optimized mythtv a little bit
19:26<Beirdo>I swear, I will hunt that bug down one of these days :)
19:26<MrGandalf>and these are interlaced h264 I believe.. dunno if the new ffmpeg will decode that
19:27<briand>xris: ah.. my failing memory, i guess. I just looked, and (sure enough) it does stay logged in... only the graphing function is called via cron periodically.
19:28|-|Anduin [n=awithers@adsl-69-110-40-109.dsl.pltn13.pacbell.net] has joined #mythtv
19:29<xris>yup
19:29<xris>Beirdo: that would be REALLY nice. :)
19:30<xris>briand: I think in this case, mysqlard will probably stay connected (since it actually creates activity every 60 seconds), but doesn't reconnect if the db resets, and I've been restarting the db to mess with ram cache values, etc.
19:30<janneg>MrGandalf: 1920x1088p at 24mbps
19:31<MrGandalf>hmm
19:31<Chutt>janneg, see my email?
19:31<MrGandalf>thanks janneg for the pointer
19:32<janneg>and current ffmpeg won't play it. h264 got buggy after 7671
19:32<janneg>Chutt: not yet
19:32<briand>xris: gotcha... and i remember reading somewhere along the way that mysqlar can deal with a db that goes away and comes back, but it would have a problem if the program were originally started before the mysql daemon...
19:32<briand>xris: so, he must have *some* watchdog code in there already... it just doesn't work with mysql5 yet. :)
19:33<xris>yeah. same with Beirdo's code. apparently something changed in the API and it just doesn't detect that the connection has gone away.
19:33<MrGandalf>janneg: It's always been extremely buggy for me
19:33<janneg>Chutt: ok, I'll fix that and commit then. probably tomorrow
19:34|-|Caliban_ changed nick to Caliban
19:34<MrGandalf>so it seems there's no free implementation of an h264 codec which works well
19:36<xris>MrGandalf: the absolute latest version of ffmpeg's + x264 seems ok, but only it and apple's stuff can play back the streams without weird artifacts (my latest annoyance)....
19:36<janneg>MrGandalf: my proc estimation is based on overclocking and mplayer. I got with 2.5ghz 65-95% cpu usage
19:37<MrGandalf>wondering if Intel copes better than AMD..
19:39<janneg>playback with mythtv was watchable but had dropped frames
19:40|-|logost [n=aolaol@ip68-107-165-193.lu.dl.cox.net] has quit []
19:40<MrGandalf>janneg: single core, right?
19:40<MrGandalf>I think a dual core 2.6 would be worse
19:41<janneg>Intel core 2 duo is not much better
19:42<janneg>MrGandalf: no dual core, single core would be worse since the kernel and graphic driver use according to oprofile 10%
19:44<MrGandalf>so you're talking an X2 5200+
19:48<janneg>opteron 185 (sockel939), but please wait and see how the optimizations go. single core 2.8ghz should also be enough
19:50<MrGandalf>hmm, socket 939.. would work in my current MB
19:51|-|DrNickRiviera [n=riviera@so-5494-x0.essex.ac.uk] has quit ["Leaving."]
19:51|-|DrNickRiviera [n=riviera@so-5494-x0.essex.ac.uk] has joined #mythtv
19:57|-|DrNickRiviera [n=riviera@so-5494-x0.essex.ac.uk] has quit ["Leaving."]
19:58|-|DrNickRiviera [n=riviera@so-5494-x0.essex.ac.uk] has joined #mythtv
20:06<MrGandalf>janneg: I think I already asked this, but does the newer ffmpeg support interlaced streams?
20:09<Chutt>the current stuff does.
20:11<MrGandalf>just in time for me then, thanks chutt
20:12<xris>woot. found a query I can rewrite to use an index. heh.
20:16<xris>damn, doesn't help
20:17<xris>it would speed up the query in general, though.
20:19<eskil>Anyone happen to have any experience with QSocketDevice and doing non-blocking connects ?
20:27<xris>someone want to double check that I didn't horribly break something with: http://svn.mythtv.org/trac/changeset/12642
20:34|-|kvandivo [n=kvandivo@ppp-70-225-162-235.dsl.chmpil.ameritech.net] has quit [Read error: 104 (Connection reset by peer)]
20:50<briand>xris: i would, but my system is busy recording for the next 4 hours
20:50<xris>briand: heh.
20:51<xris>should be fine. mainly wanted someone to make sure the query looked valid.
20:51<xris>it should be, I just reversed things.
20:51[~]briand nods.
20:52<xris>briand: btw: http://webdev.siliconmechanics.com/slow_log.pl.txt
20:52<briand>i haven't (yet) started digging into the code to look at the queries yet, like i spoke about last week... but there's certainly room for some performance improvement...
20:52<xris>I put in a ticket the other day with a bunch of index changes.
20:52<briand>yeah, I saw that when I was catching up on my mail... that's what sparked my earlier comment re: mysqlar
20:54<briand>nice script.. short & to the point. :) i'll add it to my toolbox. thanks!
20:54<xris>we have a LOT of queries that could benefit from mysql5 stuff like subqueries, but we obviously can't require that.
20:54<briand>not yet, anyway.. :/
20:54<xris>briand: I plan to keep improving that script.. maybe make a web interface so you can browse the actual queries instead of just the extracted hash, etc.
20:54<briand>InnoDB might help while still in mysql4, but that'd probably be a pita to re-code
20:55<xris>inno is slower than myisam
20:55<xris>just more robust
20:55<briand>i thought it would allow for some limited transactional queries as well...
20:56<briand>so we wouldn't get the full table lock blocking the recordedseek updates during recording
20:56<xris>not just limited, but yes.
20:57<xris>I think we're still in the "supporting 3.23" world, though
20:57<briand>really? hmm.
20:58<xris>kormoc is going to rewrite the scheduler query with subqueries, though. he apparently got a 10x speed improvement from that, and if it really IS that much, we can hopefully convince Chutt to at least put in a version-conditional query for that one.
20:58<briand>are folks running that old a version of mysql, yet still running current mythtv?
20:59<xris>I think there are, yes.
20:59[~]briand shakes his head.
21:00<xris>lots of people running php 3, too.
21:00<briand>that's worse than these clowns that come by the boards every day, thinking that *their* ATI graphics card is, somehow, "different" and they won't have any problems running Myth on it.
21:00<xris>and I *know* I'm going to piss some of them off when I make mythweb require php5 for .21
21:00<xris>or rather .22
21:01<briand>what's the big roadblock stopping folks from upgrading to php5 in the first place?
21:01<xris>laziness, probably. and dislike of touching a stable system
21:02<briand>well, if they want true stability, they'd be well-served to get off of using php3, I should think
21:02<briand>not that it was unstable, per se.. but php4 seems significantly more secure...
21:03<xris>yeah
21:03<xris>and php5 is decent.
21:04<briand>in any case, you get a little more leeway in that mythweb is an optional plugin... so it wouldn't be a 'show-stopper' for someone wanting to run current mythtv and out-of-date php
21:04<briand>not as broad-ranging a change as, say, upgrading to using mysql5, anyway.
21:04<xris>yes
21:04<xris>anyway, I'm headed home. later.
21:04|-|xris [n=xris@dsl081-161-160.sea1.dsl.speakeasy.net] has left #mythtv []
21:04<briand>take care..
21:07|-|eskil [n=eskil@natint3.juniper.net] has quit ["Leaving"]
21:24|-|isses [n=isses@ppp91-252.lns1.mel3.internode.on.net] has joined #mythtv
21:24|-|isses [n=isses@ppp91-252.lns1.mel3.internode.on.net] has left #mythtv []
21:37|-|DrNickRiviera [n=riviera@so-5494-x0.essex.ac.uk] has quit [Remote closed the connection]
21:50|-|gr8nash [n=andy@70.90.130.61] has quit [Read error: 110 (Connection timed out)]
22:01|-|xris [n=xris@xris.forevermore.net] has joined #mythtv
22:12|-|sphery [n=mdean@user-0c6sj2g.cable.mindspring.com] has joined #mythtv
22:36|-|Nem^1 [n=Nem@dslb-084-056-253-012.pools.arcor-ip.net] has joined #mythtv
22:54|-|Nem^ [n=Nem@dslb-084-056-235-013.pools.arcor-ip.net] has quit [Read error: 110 (Connection timed out)]
22:54|-|Nem^1 changed nick to Nem^
23:00|-|_Vile [n=vile@bc182112.bendcable.com] has joined #mythtv
23:00<_Vile>curious why my video stream is off by about 500ms from my audio?
23:03<xris>_Vile: /topic
23:05<_Vile>ok
23:14<_Vile>blah
23:14|-|_Vile [n=vile@bc182112.bendcable.com] has left #mythtv ["Leaving"]
23:26|-|luna6 [n=luna6@ip68-14-110-43.no.no.cox.net] has joined #mythtv
23:32|-|packetscan changed nick to PacketScan
23:40|-|siXy [n=james@88-96-183-185.dsl.zen.co.uk] has joined #mythtv
23:42<siXy>Chutt are you online?
23:44|-|eskil [n=eskil@adsl-66-120-85-205.dsl.snfc21.pacbell.net] has joined #mythtv
---Logclosed Fri Jan 26 00:00:43 2007