Back to Home / #mythtv / 2003 / 05 / Prev Day | Next Day
#mythtv IRC Logs for 2003-05-23

00:38-!-hfb [~hfb@pool0056.cvx30-bradley.dialup.earthlink.net] has joined #mythtv
00:43<poptix>Changing from WatchingRecording to RecordingOnly
00:43<poptix>Illegal instruction
00:43<poptix>fun!
00:58-!-hfb [] has quit ["Client exiting"]
01:05-!-orangey [~orangey@dsl-207-112-60-173.tor.primus.ca] has joined #mythtv
01:05<orangey>hey guys!
01:05<orangey>anybody know what could cause me to be unable to "save position" in a recording?
01:05<orangey>it's properly writing to "bookmark" when I hit the "save position", but it doesn't seem to be playing back
01:05<orangey>(from that spot, that is)
01:05<Captain_Murdoch>maybe no seektable?
01:05<Captain_Murdoch>can you jump around the file quickly with ffwd and rewind? or is it kinda slow compared to normal.
01:06<Captain_Murdoch>actually looks like bookmark will work just real slow if you don't have a seek table so that's probably not it.
01:29<orangey>Captain_Murdoch: indeed there seems no seektable.. how do I get one?
01:29<orangey>the thing is, this movie got recorded today
01:31-!-Drikus_ [~Drikus@cc45940-a.deven1.ov.home.nl] has joined #mythtv
01:45<orangey>AAAHHH! This is freaking infuriating!
01:46<orangey>I'm trying to get through this 2.5 hour movie, and it has taken me more than 3 hours to try to just get to the freaking spot 1.5 hours in!
02:09-!-justin [~justin@ool-18b81fb6.dyn.optonline.net] has joined #mythtv
03:01-!-kiwi_uk [~kiwiuk@pc3-nfds1-6-cust201.nott.cable.ntl.com] has joined #mythtv
04:05-!-orangey [] has quit ["Client Exiting"]
04:09-!-bigguy [] has quit [Read error: 110 (Connection timed out)]
04:23-!-rkulagow_ [] has quit [Read error: 110 (Connection timed out)]
04:24-!-rkulagow_ [~rkulagow@12-207-131-29.client.attbi.com] has joined #mythtv
04:48-!-bline [] has quit ["Client exiting"]
04:53-!-bline [office-7@office.gossamer-threads.com] has joined #mythtv
06:10-!-choenig [~choenig@pD952C194.dip.t-dialin.net] has joined #mythtv
06:34-!-FryGuy [] has quit ["mirc - because a chat client doesn't need a 'real' language - www.mirc.com"]
06:34-!-kiwi_uk [] has quit ["Peace and Protection 4.22"]
06:41-!-kiwi_uk [~kiwiuk@pc3-nfds1-6-cust201.nott.cable.ntl.com] has joined #mythtv
06:42-!-kiwi_uk [] has quit [Remote closed the connection]
06:42-!-kiwi_uk [~kiwiuk@pc3-nfds1-6-cust201.nott.cable.ntl.com] has joined #mythtv
06:43-!-kiwi_uk [] has quit [Client Quit]
07:22-!-kiwi_uk [~kiwiuk@pc3-nfds1-6-cust201.nott.cable.ntl.com] has joined #mythtv
07:28-!-choenig [] has quit [Remote closed the connection]
07:44-!-Captain_Murdoch [] has quit [brunner.freenode.net irc.freenode.net]
07:44-!-Chutt [] has quit [brunner.freenode.net irc.freenode.net]
07:44-!-bline [] has quit [brunner.freenode.net irc.freenode.net]
07:44-!-Edgan [] has quit [brunner.freenode.net irc.freenode.net]
07:44-!-bline [office-7@office.gossamer-threads.com] has joined #mythtv
07:44-!-Chutt [~ijr@dsl093-011-148.cle1.dsl.speakeasy.net] has joined #mythtv
07:44-!-Captain_Murdoch [~buy_more_@ip68-107-147-80.hr.hr.cox.net] has joined #mythtv
07:44-!-Edgan [edgan@24-205-99-99.rno-cres.charterpipeline.net] has joined #mythtv
08:42-!-Morph_ [gareth@3dlci148-2.icsfm.com] has joined #mythtv
08:44-!-Morph [] has quit [Read error: 104 (Connection reset by peer)]
08:49-!-justin [] has quit [Connection timed out]
08:51-!-bigguy [bigman@h97.44.102.166.ip.alltel.net] has joined #mythtv
08:58<rkulagow_>chutt: can you please sync website to CVS?
08:58<rkulagow_>(docs)
09:28-!-bline [] has quit ["Client exiting"]
09:41-!-Captain_Murdoch_ [~cpinkham@linoleum.engrs.infi.net] has joined #mythtv
09:44<rkulagow_>morning, captain_murdoch. were you going to commit the --seek enabled mythcommflag?
09:46<Captain_Murdoch_>probably not. working on a better way. :) I'm going to modify the flag commercials function in NuppelVideoPlayer so that it will save the seektable to the DB if it finds a file that doesn't have one internally while flagging. I also have modified the two recorders to save their positionmap/seektable to the DB every so often during recording so even if the backend were to die you'd still have a pretty complete seektable in the DB rathe
09:46<Captain_Murdoch_>aiting to the end of the recording to save it.
09:48<rkulagow_>ah; sounds cool. would you mind resending the patch? i seem to have lost it (conflicts during merge and i wasn't paying attention), and i wanted to see again how you were handling the header. i'd like to add some of the program metadata to the actual .nuv file to help rebuild in case of database failure.
09:49<Captain_Murdoch_>ok, will resend the original email.
09:49<rkulagow_>excellent. thanks.
09:51<Captain_Murdoch_>sent. needs to be applied from the mythcommflag directory.
09:53<rkulagow_>cool.
09:58-!-bline [~sbeck@24.84.93.233] has joined #mythtv
10:13-!-hfb [~hfb@lsanca1-ar2-4-60-014-025.lsanca1.dsl-verizon.net] has joined #mythtv
10:15<Captain_Murdoch_>rkulagow: you have a pvr-250 right? can you tell me how many rows of '6' records there are in the recordedmarkup table for a 1-hour or 1/2 hour recording?
10:15<Chutt>6 is the seektable?
10:15<Chutt>1 every 15 frames
10:17<Captain_Murdoch_>ok, that's what I thought it looked like. about 2 per second.
10:17<Chutt>yup
10:17<Chutt>1 every 12 frames for pal
10:17<Chutt>i think
10:17<Chutt>but, same frequency
10:18<Captain_Murdoch_>I made those mods to the recorders last night to write out the seektable every 15 seconds, so 15 keyframes for non-pvr250 and every 30 gops for pvr-250. also modified the nuppeldecoder to load the map from the db if it's not in the file. also just modified the flagcommercials() function to save the positionmap to the db if it plays through a file and finds haspositionmap == false.
10:19<Captain_Murdoch_>the frequency is 15 and 30, not tied to seconds but tied to positionMap.size()
10:19<Chutt>cool
10:19<Chutt>what's it do if it doesn't know the totalFrames stuff?
10:19<Captain_Murdoch_>is that in the seektable?
10:19<Chutt>it's gotten from the seektable
10:20<Captain_Murdoch_>ok, looking at that code now in nuppeldecoder.cpp
10:21-!-rizzi [] has quit [Read error: 60 (Operation timed out)]
10:21<Captain_Murdoch_>does that apply to mpeg2 also? I didn't see totalframes in anything other than osd, nuppeldecoder and NuppelVideoPlayer
10:22<Chutt>yup
10:22<Chutt>there's a function that sets totalFrames, called by avformatdecoder
10:23<Captain_Murdoch_>ok, I'll check that also. I see the m_parent->SetFileLength
10:24<Captain_Murdoch_>can you cycle through a qmap by setting iterator = map.end() and doing iterator--?
10:25<Chutt>not sure
10:28<Captain_Murdoch_>Chutt: I also modified NuppelVideoRecorder::GetKeyframePosition to check the positionmap like the mpegrecorder does rather than cycling through the seektable vector.
10:28<Captain_Murdoch_>should speed up long jumps a bit.
10:30<tstm>Has anyone of you tried the DVB-function of MythTV?
10:35<Captain_Murdoch_>ok, modified the nuppeldecoder.cpp to get the max keyframe from the positionmap loaded from the db and use that keyframe to calculate totalLength and totalFrames and then call m_parent->SetFileLength. don't need to modify avformatdecoder since that always gets it's map from the db and sets those accordingly.
10:46-!-schultmc [~schultmc@zealot.progeny.com] has joined #mythtv
10:59<Captain_Murdoch_>Chutt: want to take a look at this patch to make sure I covered all the bases?
10:59<Chutt>sure'
10:59<Chutt>busy atm, though
11:00<Chutt>but send it my way, i'll check it out
11:00<Captain_Murdoch_>ok.
11:10<Captain_Murdoch_>patch sent with a bullet-list of what it's supposed be doing.
11:10<Chutt>ok
11:12-!-kiwi_uk [] has quit [Read error: 104 (Connection reset by peer)]
11:13-!-kiwi_uk [~kiwiuk@pc3-nfds1-6-cust201.nott.cable.ntl.com] has joined #mythtv
11:14<Chutt>hmm
11:14<Chutt>i'm not cure the test in NuppelVideoRecorder is correct
11:14<Chutt>sure
11:14<Chutt>for when it's checking to see if it should write
11:14<Chutt>that won't get the watching recording stuff
11:17<Chutt>and you need to grab the db_lock before calling decoder->SetPositionMap()
11:17<Chutt>other than that, looks good
11:18<Captain_Murdoch_>ok. wasn't sure about the db_lock.
11:19<Captain_Murdoch_>you mean in NVR when it writes every 15 keyframes? or in writeseektable?
11:19<Chutt>in nvr
11:20<Chutt>it should just be checking for curRecording, i think
11:20<Chutt>hmmmm
11:20<Chutt>maybe not
11:21<Captain_Murdoch_>right, there's 2 places it writes in there though. writes inside WriteVideo and then inside WriteSeekTable. ok, so you mean the !livetv
11:21<Chutt>yeah
11:21<Chutt>since if it's gone from livetv->recording
11:21<Chutt>and you're still watching
11:21<Chutt>livetv will still be true
11:21<Chutt>also
11:22<Captain_Murdoch_>well, writeseektable only ever gets called if livetv is false so I figured I shouldn't write the seektable to the db if livetv is true.
11:22<Chutt>nope
11:22<Captain_Murdoch_>except in transitiontoring I just noticed.
11:22<Chutt>see TransitionToRing()
11:22<Captain_Murdoch_>:)
11:23<Chutt>and that brings up something else
11:23<Chutt>the stuff that calls TransitionToRing() in tv_rec.cpp
11:23<Chutt>needs to call nvr->SetRecording(NULL); after that
11:24<Captain_Murdoch_>ah, so I could just key off curRecording != NULL then if that was fixed.
11:24<Chutt>yup
11:24<Captain_Murdoch_>ok. will check/fix that and the db_lock thing.
11:24<Chutt>great
11:25<Captain_Murdoch_>should RecorderBase::SetRecording set curRecording to NULL if pginfo is passed in as NULL
11:26<Chutt>yup
11:26<Captain_Murdoch_>seems like you'd want to delete curRecording if non-null and then set to null or a new programinfo
11:26<Chutt>does it new it?
11:27<Captain_Murdoch_>yeah, does curRecording = new ProgramInfo(*pginfo);
11:27<Chutt>ah, then it should delete it, sure
11:27<Captain_Murdoch_>ok. fixing that as well.
11:28<Captain_Murdoch_>so... if (curRecording) { delete curRecording; curRecording = NULL } then does the original stuff checking pginfo and creating a new curRecording if necessary.
11:28<Captain_Murdoch_>not in that format of course. :)
11:29<Chutt>right
11:29<Chutt>that's good
11:34-!-Morph_ is now known as Morph
11:50<Captain_Murdoch_>finished the fixes a 10-15 minutes ago. I'll commit to cvs sometime after lunch. then rkulagow can re-run mythcommflag on all his broken no-seektable files. :)
11:50<rkulagow_>yay!
11:50<Captain_Murdoch_>mythcommflag will automatically fix them by writing the seektable to the DB. should work for pvr-250 and non-pvr-250 files.
11:51<rkulagow_>coolio.
12:53<yebyen>fucking hell :)
12:53<yebyen>i just updated cvs about an hour ago
12:53<yebyen>and there are fixes going in for commercial skipping now?
12:54-!-rizzi [~root@roc-24-93-16-37.rochester.rr.com] has joined #mythtv
12:55<yebyen>i need to pick up a huge drive
12:56<Captain_Murdoch_>yebyen: no, they're for seektables.
12:56<Captain_Murdoch_>just that commercial skipping benefits since if you don't have a seektable then skipping is slow.
12:56<yebyen>oh
12:57<yebyen>Captain_Murdoch_: done anything major to commercial skipping since about 20 days ago?
12:57<Captain_Murdoch_>and the way to fix a file without a seektable will be to just run it through the commercial flagging program again.
12:58<Captain_Murdoch_>I think the scene detection stuff is newer than that. it helps some if you use it in conjunction with blank-frame detection. It works OK on some shows by itself but not on others.
12:58<yebyen>so, something I need to do beyond the default settings?
12:59<Captain_Murdoch_>no, just go into setup and change to "blank frame + scene change detection" for the skip method. do you have more than one backend?
13:00<yebyen>only one
13:03<yebyen>hmmm
13:03-!-choenig [~choenig@pD952C194.dip.t-dialin.net] has joined #mythtv
13:03<yebyen>black frames before the simpsons were incredibly slow
13:03<yebyen>that what your fix is for?
13:07<Captain_Murdoch_>not sure what you mean by that.
13:07<yebyen>let me see if it happens later in the video before I complain...
13:08<Captain_Murdoch_>this just trys to detect scene changes and if it uses that information to try to fill in gaps that look like commercials but don't fit the blank-frame detection for exact-length.
13:08<yebyen>oh, heh
13:08<Captain_Murdoch_>you might be missing the seektable if skipping is slow and ffwd and rewind are slow.
13:09<Captain_Murdoch_>so if a commercial starts 25 seconds into the video but the first 25 seconds have a high rate of scene changes then the code flags that as a commercial also.
13:09<yebyen>does "Automatically skip commercials" actually skip commercials automatically? :)
13:09<yebyen>or do I still have to tag them
13:09<poptix>it does, if you have a sufficient buffer
13:09<Captain_Murdoch_>it skips them when they come so you don't have to hit 'Z'
13:10<Captain_Murdoch_>I don't use it but figured some people would like it.
13:10<yebyen>hm
13:10<Captain_Murdoch_>maybe I don't trust my own code enough yet. :)
13:11<yebyen>i don't either, at least not as of 20 days ago
13:11<yebyen>i still had the occasional tv show with a billion commercial borders in it
13:11<Captain_Murdoch_>that was a bug and should be fixed now.
13:12<Captain_Murdoch_>a problem in a loop somewhere from what I remember.
13:12<yebyen>oh, heh
13:12<yebyen>cool
13:12<Captain_Murdoch_>I found a show that had multiple commercial ends marked for the same break and tracked that down so I think it's probably what you saw.
13:18-!-vektor [~bbiggs@nat-pc.cs.Dal.Ca] has joined #mythtv
13:19<vektor>Chutt: my pvr-250 shipped to my house (but i'm away on vacation so i can't play with it yet)
13:19<vektor>Chutt: but damn, the driver looks _rough_
13:19<bline>ivtv?
13:20<vektor>Yeah.
13:20<bline>it's still alpha iirc
13:20<vektor>Some of this uses four-space tabs, other parts use eight-space.
13:20<bline>it works for me though
13:20<vektor>Their utils are pretty, erm, silly in how they work.
13:20<bline>oh heh, yeah I noticed that too
13:20<bline>some use 1 space tabs
13:20<vektor>yeah.
13:21<Chutt>vektor, going to install mythtv finally? :p
13:21<vektor>Chutt: Yeah maybe. ;)
13:21<bline>they havn't gotten their utils strait on what things are going to do what
13:21<vektor>Chutt: Have you got yours to cap at 12Mb/sec yet?
13:22<Chutt>the default bitrate is 8Mb, with a max vbr rate of 16
13:22<vektor>Hrm.
13:22<Chutt>8 looks pretty good, though
13:22<vektor>Yeah, I know.
13:22<Chutt>hell, 4 looks pretty good
13:22<vektor>But I still want to push it to its max.
13:23<vektor>Like, my first application is transcoding my VHS tapes.
13:23<vektor>So, you can see my bias.
13:23<Chutt>ah
13:23<vektor>PVR is secondary. :)
13:24<Captain_Murdoch_>that's the reason I picked up my 250 late last year. 400 tapes takes way too much space but I could put 400 CD's in a space 8 inches wide on the bookshelf.
13:24<vektor>I wonder if they'll take cleanup patches.
13:25<Captain_Murdoch_>cds/dvds that is.
13:25<bline>the project leader is out of town for a couple of weeks
13:25<vektor>bline: Ok.
13:27<Captain_Murdoch_>from freshmeat announcement "Zile is Lossy Emacs 1.6.2". why would I want a "lossy" text editor. I'm sure my compiler will love trying to figure that out. :)
13:27<Chutt>the cpu fan on my mini-itx box just got a lot louder
13:27<Chutt>hrmph
13:28<Chutt>i think the power supply fan's louder than it was a few days ago as well
13:29<Chutt>moegreen, if you need anyone to test the mythvideo stuff, i'd be glad to
13:31<Captain_Murdoch_>slashdot just posted a story an hour ago about replaytv dropping commercial advance feature so mythtv.org is going to get hit again this afternoon. :)
13:32<Captain_Murdoch_>saw a link off of freshmeat with the title 'ReplayTV May Drop "Commercial Advance"'
13:33<Captain_Murdoch_>this must be their weekly story about pvr's
13:35<bline>what do you think the chances are of getting myth to work on a laptop with the usb WinTV card?
13:35<poptix>nill.
13:35<bline>well there is a driver for it..
13:36<poptix>the mpeg one?
13:36<bline>http://pvrusb.sourceforge.net/
13:36<poptix>yeah, it hardly works, and iirc, you can't change channels
13:36<bline>nice
13:37<bline>a friend from work was looking into this
13:38-!-choenig [] has quit [Remote closed the connection]
13:49<rkulagow_>chutt: mailed you a patch to prevent the double/triple counting of NFS shared diskspace between the master and slave encoders.
13:52<Chutt>looks fine to me
13:53<Chutt>go ahead and commit it whenever you want =)
13:53<rkulagow_>great, thanks.
13:54<rkulagow_>is there even a controlled exit routine to the backends? i did the unlink in main.cpp (and i see one there for the pidfile), but how do you tell a mythbackend to cleanly shutdown anyway?
13:56<Captain_Murdoch_>just committed the seektable changes.
13:57-!-Edgan [] has quit [Read error: 110 (Connection timed out)]
14:11<rkulagow_>geez, PVRs sure are a popular topic these days. hadn't heard of sage tv before though.
14:13-!-Edgan [edgan@66-224-116-98.atgi.net] has joined #mythtv
14:16<Chutt>sagetv apparently works fairly well
14:16<Chutt>but damn, it looks worse than my original ui =)
14:24<rkulagow>hrmm. they're charging $60 for the software and say that they've licensed zap2it / TMS data, and _don't_ have a monthly subscription. is TMS data free now? maybe i'm missing something.
14:24<yebyen>hm
14:24<yebyen>licensed zap2it?
14:24<yebyen>hmmm
14:24-!-rkulagow_ [] has quit [Read error: 104 (Connection reset by peer)]
14:25<rkulagow>in their forum, one of the developers (only?) was talking about how SageTV EPG is better than scrapers because they license the data. they've plastered "no subscription fee" all over the forum and website.
14:40-!-rkulagow_ [~rkulagow@12-207-131-29.client.attbi.com] has joined #mythtv
14:48-!-gsfgf [~gsfgf@user-11fa1dm.dsl.mindspring.com] has joined #mythtv
14:51<gsfgf>i'mhaving sound trouble. If i have line-in unmuted, the sound won't stop when i pause. If i mute it i can't get sny sound period. I am usig alsa
14:53<yebyen>gsfgf: you've got your channels mixed up somehow
14:53<yebyen>gsfgf: it's in the FAQ
14:54<Chutt>make sure the capture or igain volume's turned up and stuff
14:54<Chutt>etc etc
14:57<gsfgf>all the faq said was to mute line-in
14:57<gsfgf>where do i set igain?
14:57<Chutt>alsamixer
14:58<Chutt>i believe it's all covered in the howto
14:59<gsfgf>thre's no igain in alsamixer. Or is that all the IEC958 * stuff
15:00<Chutt>there's no capture slider?
15:00<gsfgf>that's at max
15:00<gsfgf>it says capture on top and line on the bottm, right?
15:00<Chutt>fiddle with things
15:06<gsfgf>hmmm... now when i open myth, i don't get a signel. Just a green screen with some static on top and static over the audio. tvtime works though, so it's not the signel
15:24-!-Drikus_ [] has quit [Read error: 104 (Connection reset by peer)]
15:24-!-Drikus__ [~Drikus@cc45940-a.deven1.ov.home.nl] has joined #mythtv
16:50-!-justin [~justin@ool-182ed584.dyn.optonline.net] has joined #mythtv
16:51-!-kiwi_uk [] has quit ["Peace and Protection 4.22"]
16:56<rkulagow_>chutt: please sync the website to CVS docs.
16:56<Chutt>sure
16:56<rkulagow_>thanks
16:58<Chutt>done
16:58<rkulagow_>excellent.
17:08-!-Edgan [] has quit [brunner.freenode.net irc.freenode.net]
17:08-!-paperclip [] has quit [brunner.freenode.net irc.freenode.net]
17:09-!-Edgan [edgan@66-224-116-98.atgi.net] has joined #mythtv
17:09-!-paperclip [~joe@ip68-11-30-158.no.no.cox.net] has joined #mythtv
17:19<Captain_Murdoch_>rkulagow: have you tried running mythcommflag against any of your broken files to see if it created a seektable for them in the DB?
17:20<rkulagow_>not yet. can though. hold please.
17:33-!-justin [] has quit [Read error: 110 (Connection timed out)]
17:38-!-bline is now known as AgentSmith
17:38-!-AgentSmith is now known as bline
17:42-!-justin [~justin@ool-182ed584.dyn.optonline.net] has joined #mythtv
17:47<Captain_Murdoch_>rkulagow: I'm heading home now but I have irc up at the house so if you reply with a good/no-good status I'll see it when I get home.
17:47<-- Captain_Murdoch_has quit ()
17:48<rkulagow_>captain_murdoch: i found a broken file, running it now. the "breaks" is showing in the 1000's, so i'm assuming it's actually a frame count.
17:53<rkulagow_>captain murdoch: looks like it didn't work. jump ahead 10 minutes takes about 1-2 minutes; with a seektable it's pretty instant.
18:00-!-Drikus__ [] has quit ["leaving"]
18:12-!-justin [] has quit [Read error: 110 (Connection timed out)]
18:28-!-bline [] has quit ["Client exiting"]
18:29-!-schultmc [] has quit ["Client exiting"]
18:56-!-inman [] has quit ["Client exiting"]
19:18<Captain_Murdoch>rkulagow: it should have inserted a bunch of 6 records if it was pvr-250 file or 7 records if non-pvr250. did it put either of those into the recordedmarkup table?
19:19<Captain_Murdoch>yeah, when you run flagging on a file with broken seektable it will show frame count in the column that's counting up. then when done it will show the number of commercial breaks found.
19:26-!-hfb [] has quit ["Client exiting"]
19:27-!-shad [] has quit ["My damn controlling terminal disappeared!"]
19:28<rkulagow_>at dinner; weird, then. plenty of "7" records, but jumping 10 minutes still takes it much longer than it should. i'll investigate more after dinner.
19:37<Captain_Murdoch>ok. thanks. no hurry. I'm messing with that machine I gave the link to the other day. I got mine today and am going to see if I can get it up and running as my main frontend tonight. :)
19:38<bigguy>what machine is that?
19:47<Captain_Murdoch>the small dell celeron 466
19:47<Captain_Murdoch>3.6" tall, 12" wide, 14" deep
19:47<Captain_Murdoch>weight 1200Lbs
19:48<Captain_Murdoch>:) this thing has a tank for a body and a little plastic cover to make it look decent.
19:48<bigguy>heh
19:48<Captain_Murdoch>even has an intrusion detection switch on the cover so the computer can tell you on bootup whether anybody has been inside your case.
19:49<bigguy>we had a ton of those in the lab
19:49<bigguy>they sucked as windows boxen
19:50<Captain_Murdoch>disappointed that the audio isn't onboard though. the picture showed it onboard but with one of those disclaimers that says "may not look exactly as pictured" it's using one of the 2 slots for audio. should be ok though cause I was thinking of putting the pvr250 in here.
19:50<bigguy>I wish you luck
19:50<bigguy>I hated dealing with those boxen in the lab
19:51<bigguy>I much prefered the in house built pc's
19:51<bigguy>we had less problems with the in house ones ;)
19:51<Captain_Murdoch>yeah, me too, but I figured I'd try one for a frontend. if it doesn't work well then I'll make it my wife's pc. :)
19:53<bigguy>anyway dinner time
19:53<bigguy>good luck
19:58-!-Sg4 [~Zpimpa@adsl-18-68-25.sdf.bellsouth.net] has joined #mythtv
20:52<mdz_>hmm, myth 0.8 doesn't build with gcc 3.3
20:52<mdz_>I assume this is resolved in cvs?
20:53<mdz_>yep
20:54<mdz_>error: call of overloaded `llabs(long long int)' is
20:54<mdz_> ambiguous
20:54<mdz_>/usr/include/stdlib.h:746: error: candidates are: long long int llabs(long long
20:54<mdz_> int)
20:54<mdz_>/usr/include/c++/3.3/cstdlib:152: error: long long int
20:54<mdz_> __gnu_cxx::llabs(long long int)
20:54<mdz_>looks like a new gift from g++-3.3
20:54<Chutt>i think i got rid of that in cvs
20:54<mdz_>yeah, by just not calling llabs
20:54<mdz_>I'm curious to know how one is supposed to resolve the ambiguity though
20:54<Chutt>yup =)
20:55<mdz_>since llabs isn't in a namespace
20:55<Chutt>right
20:55<mdz_>::llabs maybe?
20:55<Chutt>might be just including cstdlib instead of stdlib.h
20:55<mdz_>yeah, ::llabs works
20:56<mdz_>once I get these fixed packages uploaded, I plan to finally take a stab at CVS again
20:58<mdz_>I have an RMA number for my CPU
20:58<mdz_>but I think I need to melt the thermal compound on the heatsink before I send it back
20:58<mdz_>otherwise they will get suspicious
21:00-!-Sg4 [] has quit [Read error: 60 (Operation timed out)]
21:14<Chutt>heh
21:16-!-jrh [] has quit [Read error: 110 (Connection timed out)]
21:18<rkulagow>captain murdoch: disregard the previous report of the seektable fix not working; i think it's a PEBCAK. i'll let you know once i have a chance to retest.
21:18<Captain_Murdoch>ok, thanks.
21:21<mdz_>heh, I'm building 0.8 for woody inside UML
21:21<mdz_>and I think it's about as fast as native
21:21<mdz_>because it uses g++ 2.95
21:21<mdz_>while the host uses 3.3
21:26<rkulagow_>damn. mythfrontend is hung after hitting exit while watching a recording. 99% CPU, which it never hits while watching a recording. not in gdb because the frontend doesn't have the CPU to run debug code with MPEG2. trying to attach using strace to the pid isn't doing anything.
21:35<rkulagow_>hrmm. there's a nice 19'd mythbackend still running on the master (which is running in gdb); the "hung" frontend is still using 99% CPU. chutt: is there anything that can be done from the backend to try to see what may have happened on the frontend?
21:37<Captain_Murdoch>with the commercial flagging running niced and sleeping every frame for a little, it takes about as long to flag as it does to record on my Athlon 700 so even on a faster machine you can expect the flagging to stick around for a while.
21:38<rkulagow_>hrmm. what happens if the program is deleted before flagging is done? i'm assuming that it was flagging JAG, which was the only program recorded in the last few hours. M-I-L said it was a repeat, so i deleted it (the program was fully recorded when I deleted). a few interactions later with the frontend, the frontend hung.
21:39<rkulagow_>(although it's still burning 99% CPU)
21:39<Captain_Murdoch>shouldn't matter. my wife does that all the time. the flagging still has the file open so it's not really deleted. when flagging finishes it checks to see if the record is still in the recorded table before writing the blank frame and skip info to the db.
21:40<Captain_Murdoch>any files with 0 byte lengths? it hung the other night on my wife with a 0-byte file.
21:44<rkulagow_>nope, no 0 length files. it looks like it stuck somewhere after coming back into the "Select a recording" screen; the background is there, but nothing is filled in. i think i need a faster frontend so i can run it in gdb.
21:49<rkulagow_>175GB drive for $109 after rebate at compusa. that's crazy cheap.
21:52<Chutt>captain_murdoch, craig longman didn't understand your email about the seektable stuff
21:53<Captain_Murdoch>oh well. I posted my patch in case he wants that. :) I thought I was pretty clear the first time that it wasn't necessary with current cvs.
21:54<rkulagow_>even wackier combo's at anandtech get you a 120GB drive + CD-RW + 200 blank CDs + 100 jewel cases for $19.99
21:56<rkulagow_>chutt: is there anything you'd like to try to deduce from the gdb'd backend on what could be causing the frontend to loop with 99% CPU?
21:57<Chutt>nope
21:57<rkulagow_>ok, i'll just rest and hope it doesn't happen again. thanks.
21:58<Chutt>hopefully it wonly
21:58<Chutt>wont
22:29-!-rcaskey_ [~rcaskey@c-24-98-40-6.atl.client2.attbi.com] has joined #mythtv
22:30<rcaskey_>is MythMkMovie going to become a standard part of myth?
22:31<-- rkulagow_(~rkulagow@12-207-131-29.client.attbi.com) has left #mythtv
22:32-!-rkulagow_ [~rkulagow@12-207-131-29.client.attbi.com] has joined #mythtv
22:40-!-jrh [asdf@ip68-5-230-73.oc.oc.cox.net] has joined #mythtv
22:41<mdz_>MythMkMovie is that big script wrapped around mencoder, right?
22:41<mdz_>if so, I doubt it
22:41<rcaskey_>oh it uses mencoder?
22:41<rcaskey_>I assume that puts it in licensing limbo then
22:42<mdz_>there was a question mark after that bit
22:42<mdz_>but that doesn't have anything to do with licensing as far as I know
22:42<rcaskey_>oh, what would the issue be then?
22:42<mdz_>that it's a big script wrapped around mencoder
22:42<rcaskey_>what's wrong with that?
22:42-!-Edgan [] has quit [Read error: 110 (Connection timed out)]
22:43<mdz_>quite a number of things
22:44<mdz_>not the least of which that mencoder doesn't actually understand mythtv streams quite right
22:44<rcaskey_>oh
22:44* rcaskey_awaits the coming of gstreamer
22:45<Captain_Murdoch>rkulagow: got that box up and running already as a frontend. just need to put lirc on it for my remote and It's set. Using just 20-25% cpu for playing back mpeg4 @ 352x480 @ 3500(scaled whatever that comes out to)
23:06* Captain_Murdochwonders how many hours per day the sourceforge site is "down for maintenance"
23:06<mdz_>rkulagow_: how many rebates are involved in that $19.99 deal?
23:06<mdz_>Captain_Murdoch: slightly over 24
23:06<mdz_>if there's any time left after 'maintenance', it stays broken in more subtle ways
23:06<Captain_Murdoch>mdz: and just when I was going to download the latest ivtv driver so I could set it up.
23:07<mdz_>Captain_Murdoch: I have a galeon smartbookmark
23:07<mdz_>where I can enter a sourceforge project name
23:07<mdz_>and it goes to a nearby sourceforge mirror and gives me a directory index
23:07<mdz_>that is about the extent to which I deal with SF
23:07<mdz_>http://telia.dl.sourceforge.net/sourceforge/%s/
23:07<mdz_>avoid the whole mess
23:08<rkulagow_>captain
23:08<rkulagow_>murdoch: http://forums.anandtech.com/messageview.cfm?catid=40&threadid=1053712
23:08<Captain_Murdoch>I forgot I had cvs checked out from a month or two ago. I was just able to do a "cvs update" on that.
23:08<rkulagow_>sorry; meant mdz.
23:09<Captain_Murdoch>:) realized that after I clicked the link.
23:10<mdz_># Western Digital 120GB Hard Drive (Item #s 20148555,20346127) @ $159.99 - $20IR - $50 MFR-MIR - $30 OM-MIR (#38) = $59.99 AR
23:10<mdz_>nice
23:10<Captain_Murdoch>I have one of those 120GB drives sitting right here. picked it up a few weeks ago when Circuit city ran a bad ad with 3 rebates listed. end price $10.
23:11<Captain_Murdoch>I made printouts of all 3 rebates and talked with the store about it and they said they'd honor it.
23:19<mdz_>is that $10 at the register or $10 after rebates by mail?
23:20<Captain_Murdoch>after rebates
23:21<Captain_Murdoch>they messed up on their website. they sold out of the drives at every store around here before noon that day I think.
23:21<Captain_Murdoch>then later that day we went back to the website and they had not only removed the link to the 3rd rebate but they had changed the text on the 3rd rebate page in case anybody had bookmarked it.
23:32<-- rkulagow_(~rkulagow@12-207-131-29.client.attbi.com) has left #mythtv
23:43-!-justin [~justin@ool-18b81fb6.dyn.optonline.net] has joined #mythtv
23:52-!-rcaskey_ [] has quit [Remote closed the connection]