Back to Home / #mythtv / 2007 / 02 / Prev Day | Next Day
#mythtv IRC Logs for 2007-02-23

---Logopened Fri Feb 23 00:00:45 2007
00:08|-|nosun [] has joined #mythtv
00:24|-|JoeyJoeJo [] has joined #mythtv
00:25|-|bwallen_ [] has joined #mythtv
00:27|-|bwallen__ [] has joined #mythtv
00:35|-|bwallen_ [] has quit [Read error: 60 (Operation timed out)]
00:45|-|JoeyJoeJo [] has quit [Read error: 113 (No route to host)]
00:47|-|bwallen__ [] has quit [Read error: 113 (No route to host)]
01:04|-|cattelan [n=cattelan@] has quit ["This computer has gone to sleep"]
01:20|-|okolsi [] has joined #mythtv
01:45|-|splat1 changed nick to splAt1
01:47|-|gnome42 [] has quit [Remote closed the connection]
01:57|-|splAt1 changed nick to splat1
01:57|-|luna6 [] has quit [Read error: 104 (Connection reset by peer)]
01:58|-|luna6 [] has joined #mythtv
02:03|-|aevil [] has joined #mythtv
02:09|-|czth__ [i=dbrobins@nat/microsoft/x-4edc797823a198ca] has joined #mythtv
02:26|-|czth_ [i=dbrobins@nat/microsoft/x-e9529326de4ada1d] has quit [Read error: 110 (Connection timed out)]
02:39|-|mau [] has joined #mythtv
02:39|-|mau [] has left #mythtv ["Leaving"]
02:40|-|splat1 changed nick to splAt1
02:48|-|splAt1 changed nick to splat1
03:19|-|xris [] has quit ["Leaving."]
03:48|-|noddan [] has quit [Read error: 54 (Connection reset by peer)]
03:49|-|noddan [] has joined #mythtv
04:01|-|kormoc [] has quit []
05:07|-|benklop [] has joined #mythtv
05:26|-|Captain_Murdoch [] has quit [Read error: 110 (Connection timed out)]
06:57|-|splat1 changed nick to splAt1
06:58|-|benklop [] has quit [Read error: 104 (Connection reset by peer)]
07:01|-|okolsi [] has left #mythtv []
07:06|-|benklop [] has joined #mythtv
07:19|-|lsobral [n=sobral@] has joined #mythtv
07:22|-|MoRpHeUz [n=morphbr@] has joined #mythtv
07:28|-|cattelan [n=cattelan@] has joined #mythtv
07:29|-|luna6 [] has quit [Remote closed the connection]
07:32|-|cattelan [n=cattelan@] has quit [Client Quit]
07:56|-|benklop [] has quit [Read error: 104 (Connection reset by peer)]
08:08|-|benklop [] has joined #mythtv
08:10|-|Cardoe [n=Cardoe@gentoo/developer/Cardoe] has quit ["Leaving"]
08:30<j-rod>Does VideoOutputXv try to create a Shm Image bigger than the resolution of the video being played back?
08:31<j-rod>I'm dinking around with an XGI Volari video card, and mplayer can play back hdtv video clips using Xv just fine
08:31<j-rod>myth is giving me a big blue block for 1080i stuff
08:32<GreyFoxx>I think myth will try to create and image of 1088 (multiple of 16)
08:32<j-rod>xvinfo says the max xv surface size is 1920x1080
08:32<GreyFoxx>someone mentioned that in the last couple days
08:33<j-rod>ah, that could be the explanation there
08:33|-|cattelan [] has joined #mythtv
08:34|-|Cardoe [n=cardoe@gentoo/developer/Cardoe] has joined #mythtv
08:41<j-rod>GreyFoxx: bingo -- CreateShmImages(32): video_dim: 1920x1088
08:41|-|cattelan [] has quit [Read error: 104 (Connection reset by peer)]
08:42<MoRpHeUz>GreyFoxx: it was you that was in charge of XML+http from the backend side ?
08:42|-|benklop [] has quit [Read error: 113 (No route to host)]
08:43<GreyFoxx>MoRpHeUz: Actually that's cdev
08:43<GreyFoxx>I just goof around in there, he's the one doing real work on it :)
08:44<MoRpHeUz>GreyFoxx: hhmm, ok..thanks!.maybe I can send a patch to him to implement more functions so more data can be retrieved using xml+http =D...
08:44<GreyFoxx>What sort of data?
08:44|-|Nem^ [] has quit [Read error: 104 (Connection reset by peer)]
08:46<MoRpHeUz>GreyFoxx: the schedule list, maybe insert a new schedule....recording profiles...
08:47<GreyFoxx>I though the schedule list was in there
08:47<MoRpHeUz>couldnt find..
08:47<MoRpHeUz>maybe it is...need to take a deeper look...
08:47<MoRpHeUz>was working on another patch =)
08:48<MoRpHeUz>ticket 3126..
08:49<GreyFoxx>It might not be in there now. I might have just been thinking of the regular program info being available
08:51<MoRpHeUz>still need to think about a secure way of retrieving this stuff....
09:07|-|jgarvey [] has joined #mythtv
09:27|-|CDev [] has quit [Read error: 104 (Connection reset by peer)]
09:31|-|gnome42 [] has joined #mythtv
09:31|-|CDev [] has joined #mythtv
09:34|-|okolsi [] has joined #mythtv
09:44|-|okolsi [] has quit ["Leaving"]
09:46|-|cattelan [] has joined #mythtv
10:35|-|marx1 [] has joined #mythtv
10:41|-|splAt1 changed nick to splat1
11:08|-|splat1 changed nick to splAt1
11:08<sphery>j-rod: I'm pretty sure that bug is fixed in mythtv-vid. See .
11:08<sphery>Now it uses 1920x1080 even if the video reports its size as 1920x1080
11:08<sphery>You might be able to port the change (it's simple) to your version.
11:10<sphery>Just don't know for sure how much else has changed in that area (i.e. whether that simple change can be made to trunk).
11:11[~]sphery wishes _Erroneous was still here...
11:11<sphery>Here's the reason why we don't use seriesid for matches:
11:12<sphery>Easy to find with a search on "seriesid House" (and even easier if you know that Bruce explained it all and search by his e-mail, too)
11:12<sphery>Funny since House is the example he used, too.
11:23|-|okolsi [] has joined #mythtv
11:26|-|marx1 [] has quit [Read error: 110 (Connection timed out)]
11:31|-|xris [] has joined #mythtv
11:33|-|cattelan [] has quit [Read error: 110 (Connection timed out)]
11:36|-|okolsi [] has quit ["Leaving"]
11:37|-|clever changed nick to clever`
11:39|-|beavis [] has joined #mythtv
11:41<xris>stuarta: ping
11:49<j-rod>sphery: ah, good to know
11:50<j-rod>one of the X guys thinks the max Xv texture size can actually be bigger for this card though, it was just hard-coded to 1920x1080, because someone didn't think there'd be need for anything bigger
11:55|-|trentster [] has joined #mythtv
11:57|-|clever` [] has quit ["leaving"]
12:01<trentster>Hey all, I changed my default video player to xine instead of mplayer, all works well except the volume is on 0% when i start a movie, and when i increase the volume on the remote it goes up 1 block increment at a time, and have to press the up volume like 30 times just to get it to half way...
12:01<trentster>any ideas where i change this?
12:02<xris>trentster: you're going to get a better answer if you ask in the proper channel
12:03<trentster>xris, sorry just saw that and have reposted....
12:03|-|splAt1 changed nick to splat1
12:21|-|CDev [] has quit [Read error: 104 (Connection reset by peer)]
12:24|-|xris [] has quit ["Leaving."]
12:25|-|CDev [] has joined #mythtv
12:28|-|cattelan [n=cattelan@] has joined #mythtv
12:46|-|kormoc [] has joined #mythtv
13:10<Chutt> Is there anyone at Linux that I can contact about Linux-based DVRs?
13:10<Chutt>emailed interview questions are funny.
13:12[~]kormoc laughs
13:17|-|trentster [] has quit ["Leaving"]
13:20|-|xris [] has joined #mythtv
13:23|-|cattelan [n=cattelan@] has quit ["This computer has gone to sleep"]
13:24<MoRpHeUz>questions like: "why dont you write more applications for that platform ?" annoy myself hehe ;-)
13:24<MoRpHeUz>maybe the answer should be: "because I have to answer this kind of question for an interview"..=P
13:24|-|cattelan [n=cattelan@] has joined #mythtv
13:26|-|Tbonius [] has quit ["blah"]
13:28<Chutt>the guy called me first, i told him to email me
13:28[~]j-rod notes that he'd like there to be someone you can contact about Fedora-based DVRs...
13:28<Chutt>he had emailed me twice already, i had ignored him those times =)
13:28<Chutt>wasn't rh legal worried about that?
13:28<j-rod>er, contact to buy, that is (I see quite a few ubuntu-based dvr sellers)
13:29<j-rod>but yeah, legal certainly still has some issues
13:29<j-rod>workin' on it though...
13:29<Chutt>cingular shipped my new phone 2nd day fedex instead of next day
13:29<Chutt>called em up, they said 'oops, but you didn't pay for shipping anyway, so...'
13:29<j-rod>I have a Fedora 7-based single-cd installer complete with mythtv packages sitting on one of my dev boxes
13:30<j-rod>old phone dead, needed new one asap?
13:30<Chutt>just wanted it before the weekend.
13:31<CDev>Chutt: same thing happened to the computer I just ordered... Spent extra to get it today, now they say it won't be in until monday :-(
13:32<j-rod>this is why I like living just south of NH
13:32<j-rod>gonna maybe pick up a mac mini this weekend
13:32<j-rod>walk into store, pay exactly what's on the price tag, walk out with item
13:33<j-rod>amazing how close to viable a core solo mini at 1.5GHz is for an hdtv frontend
13:33<j-rod>brought one home from the office, pretty much instantly got my wife's approval to replace the behemoth currently in our living room
13:33<CDev>j-rod: couldn't exactly find the machine I got in a local store. dual quad xeon's running at 2.6GHz.
13:33<j-rod>CDev: I suppose not, for something like that
13:34|-|aevil [] has quit [Read error: 113 (No route to host)]
13:34<j-rod>for such things, we just yell at dell and hp to send up toys :)
13:34<Chutt>i don't want a mini for a frontend in my living room
13:34<Chutt>wouldn't fit in the stack :p
13:34<j-rod>mini will fit nicely on the same shelf as my wii. :)
13:35|-|cattelan [n=cattelan@] has quit ["This computer has gone to sleep"]
13:35<Chutt>my wii's sitting in a bag
13:35<Chutt>has been for over a month and a half now
13:35<j-rod>haven't actually played much myself
13:36|-|S2 [] has joined #mythtv
13:36<MoRpHeUz>j-rod: does mini have tv input ?
13:37<j-rod>well, yes and no... :)
13:37<j-rod>its got firewire, so my cable box will feed into it
13:37<j-rod>and I could certainly point it at my hdhomerun
13:37<j-rod>and its got 4 usb ports, so I could hook up that plextor usb thingy
13:37<j-rod>but all my vid capture happens on the beast downstairs
13:38<MoRpHeUz>does this usb capture cards works well or are they slow ?
13:39<j-rod>ah, sorry, I don't actually have one of those
13:39<j-rod>it does hardware mpeg4 compression though, from what I understand
13:40<MoRpHeUz>hhmm..ok! just to know =) as soon as I can move to my new home I'm going to set up a mythbox and I'm just doing some research about hardware =)
13:40|-|S2 [] has quit [Remote closed the connection]
13:44<j-rod>sphery: finally poked at that changeset you mentioned... Doesn't appear to be applicable to this case
13:44<j-rod>the video actually is 1920x1080
13:44<Chutt>the video's going to be 1088
13:44<j-rod>the problem is that we're trying to set up a 1920x1088 xv surface
13:44<Chutt>if it's mpeg
13:45<j-rod>and the card reports a max surface size of 1920x1080
13:45<j-rod>so I'm just going to poke at the X driver and see if I can bump that a tiny bit
13:48|-|gnome42 [] has quit [Read error: 110 (Connection timed out)]
13:49|-|gnome42 [] has joined #mythtv
13:52<j-rod>I love modular X
14:00<j-rod>1-line change in a header file for the sis driver, and now 1080i stuff works
14:01|-|marx1 [] has joined #mythtv
14:16<sphery>j-rod: Sorry for the noise. That wasn't even the changeset I was thinking I remembered (which doesn't seem to exist, anyway).
14:17<j-rod>sphery: no prob
14:17<sphery>Guess you still need the 1920x1088 Xv surface for decode.
14:17<sphery>Wonder how MPlayer and xine are doing things differently.
14:17<j-rod>yep, bug filed upstream
14:17<j-rod>mplayer creates a 1920x1080 Xv surface
14:17<sphery>Cool. Better to just change the driver, anyway.
14:18<sphery>Unfortunately, it doesn't help the people using other drivers that, likewise, figured no one would ever need more than 1920x1080.
14:18<j-rod>I'm guessing the fix may be similar for other drivers
14:18<sphery>Yeah. Sounds like it.
14:18<j-rod>istr a very similar bug w/the radeon driver ages ago
14:18<sphery>Some of the Intel drivers have the problem, too.
14:19<j-rod>i810 or intel?
14:19<sphery>I guess Daniel would have to explain why we're using the 1920x1088 surface if it's not necessary (since MPlayer doesn't use it).
14:19<sphery>maybe i810
14:19<j-rod>I need to poke at i810 some more, but intel, at least on a mac mini, does the right thing
14:19<sphery>Don't remember.
14:20<sphery>Are they two drivers for the same chip? (I.e. like nv and nvidia--one open source the other closed?)
14:20<j-rod>i810 is the older driver
14:20<j-rod>intel is the new-and-improved one, fully backed by Intel
14:20<sphery>Both OSS?
14:21<j-rod>i810 uses all sorts of shitty hacks to do mode setting, due to video bios issues
14:21<j-rod>or something like that
14:21<j-rod>but yeah, both oss
14:21<sphery>I think it's i810.
14:21<sphery>So, the intel driver is probably not affected by it.
14:21<j-rod>intel driver has no xvmc support just yet, need to try i810 on the mini so I can try xvmc
14:22<j-rod>I'll poke at the i810 and radeon drivers in a sec...
14:22<jwestfall>i use xvmc on my mac mini
14:22<jwestfall>but have to use a patch to make it choose the 2nd xv port
14:22<j-rod>jwestfall: i810 driver then?
14:22<jwestfall>which can do 2048x2048
14:23<j-rod>I didn't have to change a thing using the intel driver for xorg 7.2
14:23<jwestfall>whatever the one in ubuntu is
14:23<jwestfall>should oss
14:23<j-rod>both are
14:24<jwestfall>i though when I looked at it, xv would report 1080 support, but when myth required that size it would get denied
14:24<jwestfall>and I would get a blue screen
14:24|-|JoeyBorn [] has joined #mythtv
14:24<jwestfall>your xvinfo show 2 ports?
14:25<sphery>Here's the report I was thinking about:
14:25<j-rod>myth uses 1088 is why you get the blue
14:25<sphery>i810 wouldn't work with 1920x1088.
14:25<j-rod>that's exactly the issue with the sis driver and an xgi volari I just reported upstream. :)
14:25<Chutt>the video is 1088 high.
14:25<j-rod>jwestfall: okay, yeah, i810 driver
14:25<j-rod>I'm using the intel driver on my mini
14:25<Chutt>we render directly to the video memory, to save a copy later on.
14:26<jwestfall>i thought the actual XvShmCreateImage was for 1080 tho?
14:26<sphery>Why is MPlayer able to use 1920x1080 for the Xv surface, rather than the full 1920x1088?
14:26<Chutt>they don't render to the final display surface, obviously.
14:27<jwestfall>maybe they are picking the 2nx xv port which can handle 1088
14:27<j-rod>sphery: cool, radeon driver looks to have been fixed a while back, can do 2048x2048, may have to revisit a radeon some day... :)
14:28[~]sphery thinks j-rod likes a challenge.
14:28<j-rod>indeed. :)
14:28<sphery>The NVIDIA approach works great for me. And, well tested.
14:28<j-rod>'tis what I use and recommend
14:29<j-rod>jwestfall: yep, i810 source has the same shortcoming of the sis driver
14:30<j-rod>hard-coded max image height of 1080 that a 1-line patch can fix
14:31<j-rod>ah, cool
14:32<j-rod>the guy referenced in that ticket is actually the upstream maintainer for i810, so it looks like its already fixed upstream
14:33<jwestfall>i tried that binary back when I was having problems
14:33<jwestfall>and it didnt fix it
14:34<j-rod>ah, hrm
14:34<stuarta>xris: pong!
14:35<j-rod>jwestfall: I'll take a look at it this weekend
14:36<j-rod>that, or just use the intel driver instead. :)
14:36<jams>heh the intel driver is on my list of things to test with the mini
14:37<j-rod>jams: works with the one I've got on my desk
14:37<jams>tv-out works also?
14:37<j-rod>ah, only tried dvi
14:37<j-rod>(well, and vga via adapter)
14:38<jams>would have tried it by now, but got bogged down in the settings code again
14:38<jams>all because I wanted to add a "test X configuration" button
14:38<j-rod>it pushes my 1080p tv via dvi just peachy though, don't have to do anything in xorg.conf
14:39<jams>got the button in, but stopped short of making it gather the right info and test
14:40<jams>also trying to get a TriggeredConfigurationGroup to span across two pages. Not even for sure it's possible though
14:43|-|JoeyBorn [] has quit [Read error: 110 (Connection timed out)]
14:45|-|JoeyBorn [] has joined #mythtv
14:57|-|JoeXBorn [] has joined #mythtv
15:04|-|mark__ [] has joined #mythtv
15:04|-|mark__ [] has quit [Client Quit]
15:04|-|JoeyBorn [] has quit [Read error: 110 (Connection timed out)]
15:19|-|mark__ [] has joined #mythtv
16:25|-|stuarta [n=stuart@unaffiliated/stuarta] has quit ["leaving"]
16:25|-|JoeyBorn [] has joined #mythtv
16:31|-|MoRpHeUz [n=morphbr@] has quit ["Leaving..."]
16:43|-|JoeXBorn [] has quit [Read error: 110 (Connection timed out)]
16:43|-|JoeXBorn [] has joined #mythtv
16:47|-|mark__ [] has quit ["Konversation terminated!"]
16:47|-|JoeyBorn [] has quit [Read error: 110 (Connection timed out)]
16:51|-|JoeyBorn [] has joined #mythtv
16:52|-|lsobral [n=sobral@] has quit ["Leaving"]
16:58|-|Cardoe [n=cardoe@gentoo/developer/Cardoe] has quit ["Leaving"]
17:09|-|JoeXBorn [] has quit [Read error: 110 (Connection timed out)]
17:09|-|jgarvey [] has quit ["Leaving"]
17:14|-|jmk [] has quit ["Leaving"]
17:19|-|beavis [] has quit ["Verlassend"]
17:34|-|purserj [] has quit [Read error: 104 (Connection reset by peer)]
17:34|-|purserj [] has joined #mythtv
17:38<jams>JumpConfigurationWizard has some interesting possibilities
17:45|-|plink212 [] has joined #mythtv
17:51|-|plink212 [] has left #mythtv ["h"]
18:02|-|Cardoe [n=Cardoe@gentoo/developer/Cardoe] has joined #mythtv
18:11|-|cattelan [] has joined #mythtv
18:32|-|MoRpHeUz [] has joined #mythtv
18:42|-|nvzn [] has quit [Read error: 145 (Connection timed out)]
18:51|-|MrGandalf [] has joined #mythtv
19:04|-|sperciba [n=sperciba@] has joined #mythtv
19:04|-|sperciba [n=sperciba@] has left #mythtv []
19:09|-|jams [] has quit [Read error: 104 (Connection reset by peer)]
19:11|-|LLyric [] has quit [Remote closed the connection]
19:11|-|czth [i=dbrobins@nat/microsoft/x-976622c61f9eeba3] has joined #mythtv
19:20|-|czth [i=dbrobins@nat/microsoft/x-976622c61f9eeba3] has quit ["Leaving"]
19:26|-|jams [] has joined #mythtv
19:27|-|czth__ [i=dbrobins@nat/microsoft/x-4edc797823a198ca] has quit [Read error: 110 (Connection timed out)]
19:29|-|cattelan [] has quit ["This computer has gone to sleep"]
19:38|-|crashdummyMCH [] has joined #mythtv
19:41<crashdummyMCH>New to HDTV. Have a DVICO FusionHDTV5 RT Gold and am able to get terrestrial signals, but when connected to Cable am not able to scan for any channels. Am I doing something wrong or does anyone know if this card only works with the antenna?
19:41<briand>wrong channel
19:42<crashdummyMCH>briand sorry for thanks for pointing that out.
19:46|-|crashdummyMCH [] has left #mythtv []
19:59<janneg>yeah, I found finally the cause for seeking errors in my h264 HD samples
20:08<janneg>mpegts parser was resetting the pts
20:08<briand>so it's fixed now?
20:09<janneg>not properly
20:15<janneg>and I need a faster processor :( I don't think that the ffmpeg guys can make the h264 decoder 25% faster
20:26|-|cattelan [] has joined #mythtv
20:28|-|MoRpHeUz [] has quit ["Leaving..."]
20:39<janneg>oh, make that 15%-20%. a release build is faster
20:39<xris>oh, no debug symbols, etc.
20:40[~]xris needs to track down why livetv is so slow, but watching a live recording is not.
20:40<xris>(hoping someone fixed it since my last compile)
20:41<janneg>I don't think so.
20:43<xris>might have just been the with-proc-opts compile flag last time, too.
20:43<janneg>nice. some parts play now at 25fps
20:43<xris>I'll know when I get home from work and install the new version.
20:43<xris>time to go home. later.
20:43|-|xris [] has quit [""]
20:44|-|kormoc [] has quit ["Done with work, yay! ( )"]
20:44|-|marx1 [] has quit [Read error: 110 (Connection timed out)]
21:03|-|Captain_Murdoch [] has joined #mythtv
21:09|-|kormoc [] has joined #mythtv
21:23|-|cattelan [] has quit ["This computer has gone to sleep"]
21:42|-|eskil [] has joined #mythtv
21:52|-|JoeyBorn [] has quit [Read error: 110 (Connection timed out)]
22:05|-|orgy [] has joined #mythtv
22:05|-|orgy [] has left #mythtv ["Leaving"]
22:06|-|MoRpHeUz [] has joined #mythtv
22:13|-|JoeyBorn [] has joined #mythtv
22:32|-|wrench [] has joined #mythtv
22:33|-|wrench [] has left #mythtv []
22:34|-|xris [] has joined #mythtv
22:37<briand>re-hi, xris
22:38<briand>i see has a couple rpms on their site: linux files for lightscribe support on 2.6.x kernels on rpm distros, and a Linux API/SDK :)
22:38<briand>maybe i'll look at what it might take to add a bit of magic to mytharchive...
22:42|-|marx1 [] has joined #mythtv
22:53|-|thequux_ [] has joined #mythtv
22:53|-|thequux_ [] has left #mythtv []
22:54<kormoc>niceness. Lightscribe is really nifty
22:56<MoRpHeUz>Captain_Murdoch: is there any way to make a job inside JobQueue be started as soon as it's put in there ?
22:57<Captain_Murdoch>no, :) was just mentioning that in -users, but it's better on-topic in here.
22:57<xris>easier to listen in, too. :)
22:57<MoRpHeUz>for sure
22:57<MoRpHeUz>yes, that would be great..
22:57<MoRpHeUz>does JobQueue already has this "priority" feature ?
22:58|-|marx1 [] has quit [" - Your Online Ham Radio Elmer!"]
22:58<briand>kormoc: agreed... looks neat. now, since I've swapped out the 18x DVD drive in the myth box for the lightscribe drive, excuse me for a bit while I swap out the 32x CD drive in this machine with the 18x DVD drive recovered from the myth box...
22:58[~]kormoc waves
22:58|-|briand [] has quit ["Leaving"]
23:00<xris>MoRpHeUz: from what I just saw in the other channel, that would be a "no"
23:01<MoRpHeUz>xris: yes hehe, my fault =P
23:01<xris>priority schedule would be nice, though... with something along the lines of "asap"
23:01<MoRpHeUz>yes ;-)
23:01<MoRpHeUz>problems with my english reading at 02:00 am hehe
23:02<xris>I can understand
23:03<Captain_Murdoch>The JobQueue doesn't have priorities right now, that's something on my wish/todo list but it's a ways down it right now.
23:03<xris>there should be some other/better way to handle inline transcodes, anyway.
23:04<Captain_Murdoch>yeah, I'd like to have a priority be able to override the job queue processing window.
23:04<Captain_Murdoch>that was to an ealier comment. :) yeah, I still haven't thought about hows the best way to do those type of transcodes.
23:05<xris>priority would help for things like on-demand transcodes, but wouldn't help much for livetv
23:06<MoRpHeUz>yes...a better inline transcode would be the solution for livetv...but the easiest way to give priorities to jobs should be adding one more field on jobqueue table and then sort it by priority level when starting jobs..
23:08<MoRpHeUz>dont you think ?
23:08<Captain_Murdoch>MoRpHeUz: that's the kind of thing I had in mind, yes.
23:08<xris>MoRpHeUz: would also need the poller to go more often, or add a trigger command to the backend to force it to poll
23:08<MoRpHeUz>xris: yes! that's what I thought about this afternoon...
23:08<Captain_Murdoch>and the GUI would be changed from saying "Do/Don't run Commercial Flagging for this recording" to something like "Run Commercial Flagging at priority 9 for this recording"
23:09<xris>Captain_Murdoch: that'd be cool
23:09<MoRpHeUz>hhmm..good idea...
23:09<Captain_Murdoch>xris: that's in my notes also. I was planning on sending an event to wake the JobQueue up when we queue an item.
23:09<xris>Captain_Murdoch: sounds like you and I are usually on the same page
23:10<MoRpHeUz>I can help you with this code...not sure if I'll be able to do that on the weekend but on Monday I'll start working on that! (recode my patch and this priority stuff)
23:10[~]xris really needs to just finish this channel icon lookup stuff
23:11<Captain_Murdoch>MoRpHeUz: anything you can do on it will be help, it is a ways down on my TODO list right now.
23:12<xris>Captain_Murdoch: would probably also have to add a field for "simultaneous ok"
23:13<xris>for jobs that shouldn't block
23:13<MoRpHeUz>xris: when you say "shouldn't block" you mean jobs that need to be run at the time they are requested to, right ?
23:14<xris>MoRpHeUz: I mean jobs that should allow other jobs to happen at the same time.
23:14<xris>right now, if you set up a commflag job, it prevents other commflag jobs from starting before it finishes.
23:14<Captain_Murdoch>yeah, jobs that don't count against the max jobs counter.
23:15<MoRpHeUz>ok, now I understood...
23:15<xris>and then there's the ability to renice jobs, etc.
23:15<MoRpHeUz>well, took notes from everything on my TODO list I'm going to sleep...good night! see you! thanks for the help!
23:16<Captain_Murdoch>that gets a bit harder since it mythcommflag actually does it's own nicing since it also has internal sleeps to help it slow down even more.
23:16<MoRpHeUz>any problem just email me!
23:16<MoRpHeUz>bye bye
23:16|-|MoRpHeUz [] has quit ["Leaving..."]
23:17<xris>Captain_Murdoch: well, the more i thought about *my* needs for it, it actually doesn't matter much.
23:17<xris>except for livetv stuff
23:17<xris>and then you wouldn't need renice, just a good nice level to start with.
23:17[~]xris wishes this compile would finish so he could test livetv before Psych
23:18|-|briand [] has joined #mythtv
23:19<Captain_Murdoch>renicing would be handy because you could renice jobs if a recording started, etc..
23:20<kormoc>Captain_Murdoch, any preticular reason the commflag job slows itself down? My curiosity is peeked.
23:20<Captain_Murdoch>not necessarily because of the cpu usage even, because a full-throttle commflag can hammer the disk much more than a normal recording or playback.
23:20<Captain_Murdoch>kormoc: it only slows itself down if you tell it to in the JobQueue CPU usage setting.
23:21<Captain_Murdoch>if you set that to High, mythcommflag will run full-bore.
23:21<kormoc>Ahh, fair 'nuff.
23:21[~]kormoc nods
23:21<Captain_Murdoch>default is Low I think because it's safest for people.
23:21<Captain_Murdoch>Low sleeps and nices, Medium just nices, High does neither.
23:21[~]xris wonders if that has anything to do why his livetv is slow
23:21<kormoc>Makes sense
23:22<Captain_Murdoch>well, realtime flagging always runs at ~realtime speed, it sleeps as necessary to stay a short ways behind live.
23:22<xris>well, my case is weird. livetv is slow/choppy, but if I press R and go watch from the recordings page, things are fine.
23:23<xris>it may be compile flags, though. or fixed in svn.
23:23<xris>ok, compile finally finished
23:30<xris>yup, still choppy.
23:30<xris>it's weird, too. cpu usage stays really low.. like 7% for backend+frontend
23:30<xris>system is swapping, but not THAT much.
23:30<kormoc>any io wait?
23:31<xris>how do I check?
23:34<xris>you mean load avg?
23:34<kormoc>nah, under the cpu %'s, there's one called wa
23:34<Captain_Murdoch>does your 'top' have an iowait cpu stat?
23:34<Captain_Murdoch>mine has user, system, nice, iowait, idle.
23:34<xris>wa, gotcha.
23:34<xris>hang on.
23:37<kormoc>mine is us (user), sy (system), ni (nice), id (idle), wa (iowait), hi (?), si (?), st (?)
23:39<xris>nope, nothing. even rebooted to clear out the swap. 84% idle, and less than 1% wait
23:51|-|eskil [] has quit ["Leaving"]
---Logclosed Sat Feb 24 00:01:01 2007