Back to Home / #mythtv / 2003 / 05 / Prev Day | Next Day
#mythtv IRC Logs for 2003-05-12

00:00<foom>but pretty much the only thing I'm not quite sure about that I changed.
00:00<Chutt>if it's a problem, it can always be added back
00:00<foom>yeah
00:02<foom>who's the primary maintainer of mythmusic right now?
00:02<Chutt>thor, or me
00:02<foom>ok
00:03<foom>wasn't sure whether it was one of the more seperetely modules and if so I'd want to talk to the person doing it.
00:03<Chutt>for the audio stuff, that can go through me
00:04<foom>k
00:04<Chutt>you won't be touching the playlist/ui code, which is what thor's working on
00:04<foom>right.
00:04<bline>is someone adding volume control to mythmusic?
00:05<Chutt>not directly
00:05<Chutt>foom, also, if you're going to mod mythmusic, the audio output stuff should probably live in libmyth
00:05<foom>ok, i can move that over
00:15-!-hfb [] has quit ["Huh! Wasn't me!"]
00:15<foom>wow, the mythmusic audio code is really hairy
00:16<foom>there's so many layers of abstraction i'm having a really hard time following it
00:16<Chutt>heh
00:16<Chutt>yeah :(
00:17<Chutt>mainly, audiooutput.cpp, starting at line 70 or so
00:19<foom>i'm trying to decode "Recycler" since it seems to depend on that
00:20<Chutt>that's just internal buffering
00:22<foom>can that be removed and replaced with the buffering scheme I used in nuppelvideoplayer that I already copied?
00:22<Chutt>probably not easily
00:22<Chutt>heh
00:22<Chutt>i dunno =)
00:28-!-bline [] has quit ["Client exiting"]
00:30<foom>entirely too much locking and inter-thread communication spread all over the code for my tastes.
00:36-!-poptix [] has quit [Read error: 104 (Connection reset by peer)]
00:36-!-poptix [poptix@precious.net] has joined #mythtv
00:43<Chutt>hrmph
00:43<Chutt>wish the qt sql stuff was threadsae
00:46<Chutt>safe
00:46<foom>it's not threadsafe even if you have one open connection per thread?
00:46<Chutt>it's threadsafe if you have separate connections per thread
00:46<foom>ok, not too bad then. :)
00:46<Chutt>it's just a pain sometimes
00:46<foom>hey, i'm a little C++-fresh, why is it that people write constructors like "Output::Output(unsigned int sz) : r(sz), bufsize(sz) { }" to initialize the r and bufsize variables, instead of simply putting r = sz; bufsize = sz; inside the body
00:46<Chutt>they're silly
00:46<Chutt>i don't like that, either
00:46<foom>ok, cause that just looks really *wrong* to me. :P
00:46<Chutt>it's like, you're assigning a variable, not calling a function
00:46<foom>i never knew you could do that. :P
00:46<Chutt>i think there's a technical difference
00:46<Chutt>maybe ordering, i dunno
00:46<Chutt>might be faster or something, too
00:47<Chutt>but, it's silly
00:47<foom>C++ is entirely too concerned with faster. if i had my way everything would be virtual always. :)
00:48-!-digital [~digital@h24-66-54-32.wp.shawcable.net] has joined #mythtv
00:48<digital>hello
00:48<digital>anyone awake?
00:49<foom>yep
00:49<digital>foom, do you know if there are any pre-packaged mandrake mythtv binaries?
00:50<foom>mail archives says this: http://atrpms.physik.fu-berlin.de/dist/rh80/mythtv/
00:51<digital>solid
00:51<digital>thanks
00:51<foom>btw i found that by googling for "mandrake mythtv"
00:52<digital>do i still need to build the other stuff (thexmltv and lame?
00:52<foom>I would assume you can find RPMs for those too
00:53<Chutt>i'm pretty sure he's got rpms of all the requirements there
00:53<Chutt>he doesn't have rpms of mythmusic, and all the other modules, though
00:54<digital>chut, pardon/
00:58-!-hadees [~blah@pcp01500702pcs.univde01.de.comcast.net] has joined #mythtv
01:01<hadees>hey i have a tv wonder card and the channels are off by one and the picture is almost black and white (really muted colors), i saw in the mailing list archives that i needed to set the confs my self but i can't seem to find what to set them too
01:09-!-WarrenCanuck [] has quit [Read error: 54 (Connection reset by peer)]
01:09-!-WarrenCanuck [malevolent@u167n251.hfx.eastlink.ca] has joined #mythtv
01:10-!-hfb [~hbarrett@pool0643.cvx30-bradley.dialup.earthlink.net] has joined #mythtv
01:10<foom>might try reading the FAQ question labelled almost exactly that
01:11<digital>are you talking to me/
01:11<foom>no
01:11<digital>do you have mythtv up and running/
01:12<foom>yes
01:12<digital>compiled, or pre-packaged?
01:12<foom>both
01:12<foom>i started with debian prepackaged mythtv
01:12<foom>then when I got that working I compiled my own to hack on the code
01:14<digital>debian woodu?
01:14<digital>err woody
01:14<foom>i installed debian unstable because i like to play it dangerous
01:18<Captain_Murdoch>Chutt: you still around?
01:18<Captain_Murdoch>guess not. I shouldn't be either. :)
01:19<hadees>where is the faq about the ati wonder?
01:23<foom>·   All of my channels seem to be off by one, and the picture quality is poor
01:24<foom>The bttv driver is not autodetecting your tuner type correctly, or you are specifying the wrong tuner type. Find out which one your card uses, and read Documentation/video4linux/bttv/CARDLIST to find out how to specify it.
01:28<hadees>thanks
01:47-!-hachi [hachi@24-196-120-102.fdl.wi.charter.com] has joined #mythtv
01:47<hachi>this a channel where I can ask simple questions about what is going wrong with the way I'm doing this?
01:55-!-foom [] has quit [Read error: 60 (Operation timed out)]
01:55-!-rkulagow_ [] has quit [Read error: 60 (Operation timed out)]
01:55-!-hadees [] has quit [Read error: 60 (Operation timed out)]
01:56-!-digital [] has quit ["using sirc version 2.211+KSIRC/1.2.4"]
01:56-!-mykilx_ [1000@dialup-67.75.57.168.Dial1.Boston1.Level3.net] has joined #mythtv
01:57-!-mykilx [] has quit [Killed (NickServ (Ghost: mykilx_!1000@dialup-67.75.57.168.Dial1.Boston1.Level3.net))]
01:57-!-mykilx_ is now known as mykilx
01:57<hachi>well, if it is, my problem is documented (somewhat) here http://nopaste.snit.ch:8001/4072
02:18-!-mykilx [] has quit [Killed (NickServ (Ghost: mykilx_!1000@dialup-67.75.57.234.Dial1.Boston1.Level3.net))]
02:18-!-MykilX [1000@dialup-67.75.57.234.Dial1.Boston1.Level3.net] has joined #mythtv
02:35-!-hfb [] has quit ["Huh! Wasn't me!"]
03:27-!-ribo [~ribo@dont-panic.system42.net] has joined #mythtv
03:28-!-MykilX [] has quit ["Hey Mother, do you like my Drogon boots?"]
03:35-!-poptix [] has quit [Read error: 110 (Connection timed out)]
03:37-!-Cloak [mmen@crackaddict.com] has joined #mythtv
03:37<Cloak>anyone experience a high pitched whine when recording using pvr250?
03:43<hachi>can't say that I have, but I can't get ./setup to load yet anyways, so I'm guessing I have a bit of a base problem ;)
03:43<hachi>er, setup/setup I mean
04:06-!-hadees [~blah@pcp01500702pcs.univde01.de.comcast.net] has joined #mythtv
04:07-!-poptix [poptix@precious.net] has joined #mythtv
05:12<FryGuy>Cloak: not a high pitched whine, but a lot of crackling everywhere not just in myth :p
05:31-!-hadees [] has quit [Read error: 110 (Connection timed out)]
07:24-!-froggy19 [~fryguy@12-246-43-18.client.attbi.com] has joined #mythtv
07:27-!-WarrenCanuck [] has quit [calvino.freenode.net irc.freenode.net]
07:27-!-Chutt [] has quit [calvino.freenode.net irc.freenode.net]
07:27-!-WizFactor [] has quit [calvino.freenode.net irc.freenode.net]
07:27-!-FryGuy [] has quit [calvino.freenode.net irc.freenode.net]
07:27-!-thor_ [] has quit [calvino.freenode.net irc.freenode.net]
07:27-!-froggy19 is now known as FryGuy
07:32-!-WarrenCanuck [malevolent@u167n251.hfx.eastlink.ca] has joined #mythtv
07:32-!-Chutt [~ijr@dsl093-011-148.cle1.dsl.speakeasy.net] has joined #mythtv
07:32-!-WizFactor [rob@12-246-145-135.client.attbi.com] has joined #mythtv
07:32-!-thor_ [~thor@208.185.11.34] has joined #mythtv
07:32-!-WarrenCanuck [] has quit [Read error: 104 (Connection reset by peer)]
07:32-!-WizFacto1 [rob@12-246-145-135.client.attbi.com] has joined #mythtv
07:32-!-WizFactor [] has quit [No route to host]
07:32-!-WarrenCanuck [malevolent@u167n251.hfx.eastlink.ca] has joined #mythtv
08:56-!-rkulagow_ [~rkulagow@12-207-131-29.client.attbi.com] has joined #mythtv
09:28<rkulagow_>chutt: are you here?
09:43<thor_>Chutt, you around?
09:43<Chutt>not quite yet
09:43<Chutt>just got up
09:44<thor_>.... for scrollback ... after your coffee ... real life intervened over the last few days. sorry for the MMusic delays ... something later today
09:44<rkulagow_>chutt: just let us know when you're available.
09:47<Chutt>few minutes while i wake up =)
09:56<-- WarrenCanuckhas quit ()
10:29-!-Chutt [] has quit [Read error: 104 (Connection reset by peer)]
10:32-!-schultmc [~schultmc@zealot.progeny.com] has joined #mythtv
10:36-!-bigguy [] has quit [Read error: 113 (No route to host)]
10:40-!-bigguy [bigman@h60.44.102.166.ip.alltel.net] has joined #mythtv
10:41<rkulagow_>chutt: awake yet?
10:41<hachi>didn't he drop off?
10:44<hachi>s/he/they/
11:00-!-Chutt [~ijr@dsl093-011-148.cle1.dsl.speakeasy.net] has joined #mythtv
11:00<rkulagow_>welcome back, chutt.
11:00<Chutt>hi
11:00<bigguy>Chutt: problems again?
11:00<Chutt>naw, not really
11:01<rkulagow_>chutt: a few things: 1) can you please sync the website to CVS docs?
11:01<Chutt>sure
11:02<rkulagow_>2) would you care to weigh-in on bruce markey's assertion that changing gbuffers for bttv has no effect vs. advice that other people have said "increase gbuffers to 16 or 32 to help if mythTV gets CPU bound"
11:02<Chutt>it shouldn't help any
11:02<Chutt>mythtv uses 2 buffers
11:03<hachi>did anyone see my question, or should I re-post it?
11:03<rkulagow_>bruce found some code where internal buffers of 20MB and 16MB are setup and that the bttv buffers are commented out.
11:03<mdz>hachi: depends on whether you read he FAQ and HOWTO yet
11:04<Chutt>the bttv buffers aren't commented out
11:04<Chutt>it's always used 2
11:04<Chutt>the whole VIDIOCSYNC/CMGBUF or whatnot
11:04<rkulagow_>sorry, commented out was the wrong phrase.
11:05<rkulagow_>in this thread: "Re: [mythtv-users] Choppy video [no, it's not DMA :)]" he said, "Are you providing enough video-buffer space? The link you provided says ...
11:05<rkulagow_>
11:05<rkulagow_> bttv: using 4 buffers with 2080k (8320k total) for capture
11:05<rkulagow_>
11:05<rkulagow_>There is some disagreement on this list about optimal buffering, but if you
11:05<rkulagow_>are dropping frames, I doubt anyone would dispute the merits of *trying* a
11:05<rkulagow_>higher value to see if it helps. The lowest value I recall seeing anyone
11:05<rkulagow_>actually recommend here is 8, and some of us (with a lot of RAM) use 16 or
11:05<rkulagow_>32 (the maximum possible).
11:05<rkulagow_>"
11:05<rkulagow_>(he, ray o)
11:06<rkulagow_>so, while it's true that it can't hurt, it's probably just a waste of RAM in mythtv's case, correct?
11:06<Chutt>it's just kernel ram
11:06<Chutt>nothing mythtv is using
11:06<hachi>mdz: oh, I can ask you the real question, how do I get mythtv-database to not choke on my mysql server during setup? (my earlier question was because I gave up on the debian packages)
11:07<hachi>it just gets errors trying to auth with mysql and promptly dies
11:07<hachi>says it can't authenticate as root@localhost, but mysql -u root@localhost yields me no errors and works perfectly
11:08-!-hfb [~hbarrett@lsanca1-ar2-4-60-015-094.lsanca1.dsl-verizon.net] has joined #mythtv
11:09<rkulagow_>hachi: i don't know if i'm running into a similar issue. remote backends are connecting just fine to the mysql database, but the masterbackend is now spitting out "can't connect" errors for some reason. mysql -u mythtv -pmythtv mythconverg works fine, but setup and mythbackend are both saying "access denied" messages.
11:09<rkulagow_>haven't had the cycles to trace through the code yet.
11:10<hachi>I'm trying to install debian packages, this is happening during the configuration stage
11:10<rkulagow_>ah, different then.
11:11<hachi>http://nopaste.snit.ch:8001/4072 <--- I am getting this error when I try to not use the debian packagees though
11:12<hachi>packages, even
11:12<rkulagow_>WTF: [mythtv@masterbackend .mythtv]$ cat mysql.txt
11:12<rkulagow_>DBHostName=192.168.10.50
11:12<rkulagow_>DBUserName=mythtv
11:12<rkulagow_>DBPassword=mythtv
11:12<rkulagow_>DBName=mythconverg
11:12<rkulagow_>[mythtv@masterbackend .mythtv]$ mysql -u mythtv -pmythtv -h192.168.10.50 mythconverg
11:12<rkulagow_>ERROR 1045: Access denied for user: 'mythtv@192.168.10.50' (Using password: YES)
11:12<rkulagow_>[mythtv@masterbackend .mythtv]$ mysql -u mythtv -pmythtv mythconverg
11:12<rkulagow_>Reading table information for completion of table and column names
11:12<rkulagow_>You can turn off this feature to get a quicker startup with -A
11:13<rkulagow_><snip>
11:13<Chutt>is that localhost?
11:13<Chutt>or
11:13<Chutt>hachi, you don't have the qt-mysql3 stuff installed
11:13<rkulagow_>yes, masterbackend == localhost
11:13<Chutt>no, is 192.168.10.50 localhost
11:14<hachi>Chutt: that's strange, I wonder why I didn't get an error at compile time then
11:15<Chutt>because it's not a compile time dependency?
11:15<Chutt>it's mentioned in the docs in several places as a dependency
11:15<rkulagow_>yes. i [just] changed dbhostname to localhost on masterbackend (the box hosting the database) and setup can connect just fine.
11:16<rkulagow_>[mythtv@masterbackend .mythtv]$ cat /etc/hosts
11:16<rkulagow_>127.0.0.1 localhost
11:16<rkulagow_>192.168.10.50 masterbackend
11:16<Chutt>rkulagow, so it's probably a permissions error in mysql -- i do know that it's really strict about what you give it to match against
11:16<hachi>I kept installing everything I saw, wonder if debian removed it on my at some point to resolve a dependancy issue
11:17<Chutt>i highly doubt that.
11:17<rkulagow_>chutt: the weird part is that "livingroom" _is_ able to connect. the grant is open to all; i'm using the user@"%" on the grant, which should make it open to every IP on my subnet.
11:17<rkulagow_>anyway, weird.
11:17<Chutt>yeah
11:18<Chutt>mysql's strange like that =)
11:18<Chutt>i've got one more thing to take care of (work stuff), then i'll get to the docs
11:18<rkulagow_>NP, thanks.
11:23<hachi>ahh, the real debian package I want is named libqt3cl02-mt-mysql, how obvious
11:24<Chutt>bitch bitch bitch
11:24<rkulagow_>seems that between bruce and chris pinkham they're a lot closer to the "create a seektable if the recording doesn't have one" than i am.
11:35-!-Drikus__ [~Drikus@cc45940-a.deven1.ov.home.nl] has joined #mythtv
11:41<Chutt>rkulagow, shouldn't the delete stuff make sure that the file exists before deleting it?
11:41<Chutt>otherwise, if the slave's not visible, it's going to remove it from the database, but leave it on the filesystem
11:43<rkulagow_>chutt: i'll take another look. thanks.
11:58<mdz>hachi: I have no idea why it would fail if you can authenticate as root with no password
11:58<mdz>hachi: unless you entered a password during configuration of the package
11:59<mdz>hachi: though maybe you should paste the exact error message
11:59<rkulagow_>chutt: i've looked, and the patch as submitted works.
11:59<rkulagow_>mysql> select * from recorded where hostname="bedroom";
11:59<rkulagow_><snip>
12:00<rkulagow_>| 1047 | 20030506220000 | 20030506223000 | Futurama | Mother's Day | Mom wants world domination for a Mother's Day gift. | bedroom | NULL | 0 | NULL |
12:00<rkulagow_><bedroom is not running a backend>
12:00<rkulagow_>after delete:
12:00<rkulagow_>mythtv@masterbackend video]$ ls -l 1047_20030506*
12:00<rkulagow_>-rw-r--r-- 1 mythtv mythtv 1934374848 May 6 22:30 1047_20030506220000_20030506223000.nuv
12:00<rkulagow_>[mythtv@masterbackend video]$ ls -l 1047_20030506*
12:00<rkulagow_>ls: 1047_20030506*: No such file or directory
12:00<rkulagow_>[mythtv@masterbackend video]$
12:00<rkulagow_>| 1047 | 20030506220000 | 20030506223000 | Futurama | Mother's Day | Mom wants world domination for a Mother's Day gift. | bedroom | NULL | 0 | NULL |
12:00<rkulagow_>+--------+----------------+----------------+---------------+----------------------------+-----------------------------------------------------------------------------------------------------+----------+----------+---------+---------+
12:00<rkulagow_>6 rows in set (0.00 sec)
12:01<rkulagow_>| 1058 | 20030506210000 | 20030506213000 | Lucky | The Tell | Lucky develops a tell which could jeopardize his chances in an upcoming poker tournament. | bedroom | NULL | 0 | NULL |
12:01<rkulagow_>+--------+----------------+----------------+---------------+----------------------------+-----------------------------------------------------------------------------------------------------+----------+----------+---------+---------+
12:01<rkulagow_>5 rows in set (0.00 sec)
12:01<rkulagow_>meaning, it's gone from the database, and it no longer exists on the disk.
12:01<hachi>mdz, I'm at work right now, so you won't get an error till I finish building a chroot for me to test this, or I go home
12:01<rkulagow_>are you seeing something that i'm not?
12:04<Chutt>rkulagow, no no, if the slave + master _don't_ share disk space
12:04<rkulagow_>ah, ok, that's what i missed.
12:04<rkulagow_>i'll look again.
12:04<Chutt>i wasn't clear about that
12:04<Chutt>sorry
12:09<mdz>hachi: you did tell it that the mysql server is on 'localhost', and not the hostname of the machine, right?
12:10<rkulagow_>chutt: i'll have to look at it later and work through the code. thanks.
12:11<Chutt>mdz, or the ip of the machine
12:11<Chutt>mysql doesn't always translate =)
12:13-!-Drikus___ [~Drikus@cc45940-a.deven1.ov.home.nl] has joined #mythtv
12:15-!-Drikus__ [] has quit [Read error: 110 (Connection timed out)]
12:17<hachi>mdz, is your package supposed to ask me questions? I realize now that I was never asked a thing
12:17<Captain_Murdoch>Chutt: does giving the comm flag thread its own db connection mean calling adddatabase, then setdatabasename, setusername, setpassword, and sethostname? then I can use the new db connection?
12:18<Chutt>captain_murdoch, i've got it done already
12:19<Chutt>but, not quite
12:19<Captain_Murdoch>ok, thanks. I replied to your email on-list asking if you wanted me to take care of it.
12:19<Chutt>just QSqlDatabase::addDatabase()
12:19<Chutt>then gContext->OpenDatabase(db)
12:19<Chutt>i believe
12:20<Captain_Murdoch>ok, I see that won't touch m_db in OpenDatabase(db) in this case, it will just touch db.
12:21<mdz>hachi: it asks questions at certain priorities
12:21<mdz>hachi: there is a system-global setting for the minimum priority
12:21<Captain_Murdoch>what do you think of the post saying it should be a separate process? I already have that done anyway, I have a mythcommflag executable that I run sometimes that can flag one or all shows or just flag the ones that are unflagged already.
12:22<mdz>hachi: if your priority is set higher than that of the question, that means you don't want to see it and it is skipped (uses the default)
12:22<Chutt>doesn't matter, really
12:22<hachi>mdz, the 'default' on debian is medium, I'll have to figure out which package holds that information and reconfigure it
12:23<mdz>hachi: dpkg-reconfigure debconf
12:23<Captain_Murdoch>I didn't think it did much. why work around a bug when you can fix it. :)
12:23<mdz>hachi: and btw, mythtv-database asks for the admin password at priority 'high'
12:23<mdz>so even if your priority is set to medium, you should have seen it
12:24<hachi>mdz: strange, I just built that box last night, never saw any questions, I'll have to try again tonight
12:24<mdz>hachi: that is true since mythtv-database 0.8-10
12:24<hachi>and I'm positive that I chose 'medium' on that box
12:25<mdz>which was months ago
12:25<hachi>I dunno what to say till I try it
12:25<rkulagow_>captain_murdoch: did you see my follow-up to the thread on commercial flagging, and spawning a background process to check files to see if they have a seektable, and create one if there isn't one?
12:26<hachi>mdz, is there another mirror for marillat.free.fr that you are aware of? they appear to have no httpd running today
12:27<Captain_Murdoch>rkulagow_: yeah. I was just looking at some code and wondering what it would take to do something like that.
12:28<hachi>mdz, and filtering ICMP messages possibly
12:31<mdz>hachi: yeah, I noticed that
12:31<mdz>and no, I do not know of another mirror
12:31<mdz>it goes down from time to time but comes back
12:31<mdz>Captain_Murdoch: mixing pthreads and fork is usually a recipe for bad news
12:31<mdz>I don't see why it shouldn't be a thread
12:32<Captain_Murdoch>that's what I'm replying to the guy on the list, and quoting my "why work around a bug when you can fix it." comment here. :)
12:36<Chutt>captain_murdoch, mind testing this once i check it in?
12:36<Captain_Murdoch>sure. want to see it anyway. :) so I don't make the same mistake later.
12:38<Cloak>is there a quick fix to temporarily disable the whining in the ivtv?
12:39<thor_>nziarek_2, you around?
12:40-!-mykilx [1000@dialup-67.75.57.165.Dial1.Boston1.Level3.net] has joined #mythtv
12:41-!-Chutt [] has quit [Read error: 104 (Connection reset by peer)]
12:42-!-Chutt [~ijr@dsl093-011-148.cle1.dsl.speakeasy.net] has joined #mythtv
12:45<Cloak>Chutt: how do you avoid the whine in ivtv?
12:46<hachi>mdz: I have no clue what went wrong last night, but I just built a chroot and it's working perfectly, thanks for pointing this out
12:47<Chutt>cloak, you don't
12:47<Cloak>ah
12:48<Chutt>unless you want to mess around with different driver versions
12:48<Cloak>gotcha
12:48<Cloak>barrier to entry is a little high for me to try to go in and help
12:48<Chutt>aye
12:48<Cloak>this is a new problem?
12:48<Chutt>not really
12:49<Chutt>couple months old
12:49<Chutt>or a month or so
12:49<Chutt>i told the ivtv guys 'bout it when i first got my card working, apparently i was the first
12:49<Cloak>weird
12:50<Chutt>happens on both my 250 and 350
12:51<Chutt>captain_murdoch, it's in there, now
12:51<Chutt>but i didn't really test things
12:51<Chutt>so if you could, that'd be great =)
12:53<Captain_Murdoch>already checked out and wondering how many qt v3.0.x vs 3.1.x posts on the list it will generate. :) I'm assuming that's what it is, I'm getting "`QMutexLocker' undeclared"
12:53<Captain_Murdoch>I need to upgrade sometime. :)
12:53<Chutt>bah
12:53<Captain_Murdoch>or is that an include header?
12:54<Chutt>just replace those all with a lock/unlock pair
12:54<Chutt><sigh>
12:54<Chutt>qt 3.1's been out for almost 6 months now :p
12:54<Cloak>heh
12:54<thor_>#warn Qt 3.1 sucks
12:54<Captain_Murdoch>yeah, but I don't upgrade my distros but every 2 years or so. :)
12:54<thor_>#warn 3.0
12:54<thor_>never mind
12:55<Cloak>#warn never mind
12:55<thor_>#warn thor not making sense
12:57<Chutt>captain_murdoch, i'll fix it in a sec
12:58-!-mchcoastal [~michael@fw.teamstudio.com] has joined #mythtv
12:59<poptix>[12/1158] <+poptix> kirtk: what do you need linux for anyway?
12:59<poptix>[12/1158] <+kirtk> poptix, i just need the visual basic source for linux, where can i get it?
12:59<poptix>heh.
13:00<Cloak>sounds coherent enough, did you give it to him?
13:01<Chutt>captain_murdoch, try now?
13:03<Captain_Murdoch>I was just editing the file. :) ok, rechecking out. so the thread didn't need it's own connection if locked properly?
13:03<Chutt>it's getting its own connection
13:03<Chutt>in tv_rec
13:03<Captain_Murdoch>ok, didn't look at that file, sorry.
13:04<Chutt>it would need to share the lock with the TVRec object, otherwise
13:04<Chutt>i just added the extra locking because i was being paranoid
13:05<Captain_Murdoch>ok, I see you pass the new db to the new player object.
13:23<Captain_Murdoch>Chutt: compiled, installed, recording, and setup to record 2 additional shows once this one is finished in 6-7 minutes.
13:23<Chutt>i just need to know that the commercial skip thread ends properly and stuff
13:23<Chutt>i'm not _entirely_ sure about that removeDatabase call
13:25<Captain_Murdoch>I'll keep an eye on it as they finish and check the db to make sure flagged data was written.
13:26<Chutt>thanks
13:29<Captain_Murdoch>thanks for fixing it. I see you also added some locks around other updates as well. better safe than sorry.
13:29<Chutt>yup
13:30<Chutt>i should go through and replace all the pthread_mutex_t's with QMutexes
13:30<Chutt>and the pthreads with QThreads
13:30<Chutt>one less direct dependency
13:30<Chutt>and would aid porting elsewhere, if anyone had the inclination
13:34-!-justin_ [] has quit [Read error: 54 (Connection reset by peer)]
13:35-!-justin [~justin@H233-225.STATE.RESNET.ALBANY.EDU] has joined #mythtv
13:40<Cloak>after a few hours of listening to the high pitched whine i think it starts messing with your mind
13:40<Cloak>i'm seeing flying girls with super powers
13:40<Cloak>its surreal
13:45-!-hadees [~blah@pcp01500702pcs.univde01.de.comcast.net] has joined #mythtv
13:49-!-mykilx [] has quit [Read error: 113 (No route to host)]
13:49-!-CyberPsyko [~CyberPsyk@bgp01015761bgs.rosvle01.mi.comcast.net] has joined #mythtv
13:49<CyberPsyko>hi all!
13:50<CyberPsyko>moegreen, are there?
13:51<CyberPsyko>anybody?
13:53<hadees>hi
13:53<CyberPsyko>hi
13:53<Captain_Murdoch>Chutt: first recording finished successfully, backend is still running recording the next show and the flag thread finished and correctly updated the database.
13:54<CyberPsyko>can some answer a question about MythWeather?
13:54<Captain_Murdoch>go ahead and ask and if someone can answer they will
13:55<CyberPsyko>Is the Radar map available in CVS and/or Stable reslese?
13:55<Captain_Murdoch>cvs
13:55<Chutt>captain_murdoch, great
13:55<CyberPsyko>is there known issues with RH9 and the cvs (I think i'm not compiling it right)?
13:56<poptix>CyberPsyko: it's known that people don't rtfm
13:56<CyberPsyko>I did read everything I could find, but all I can find is theQTDIR... and it's set correctly
13:56<poptix>CyberPsyko: have you checked the documentation at www.mythtv.org?
13:57<poptix>okay, and what error are you receiving?
13:57<CyberPsyko>forgive me for being a novice...
13:57<CyberPsyko>hang on a min.. it's on another computer
13:59<CyberPsyko>I get errors about undeclared indetifiers (main.o) Error 1 (BTW I did have the Stable release working correctly)
13:59<Chutt>the exact error message, if you expect people to be able to help you.
14:00<CyberPsyko>k, hang on....
14:02<CyberPsyko>unittype.h (no such file or directory) and xmlparse (same thing) am I missing files?
14:03<Chutt>you need to install cvs mythtv, of course
14:03<CyberPsyko>I have Mythtv up and working fine
14:04<Morph>CyberPsyko: you're trying to use mythweather-cvs with mythtv-stable?
14:04<CyberPsyko>I take it that won't work?
14:04<Morph>doubtful.
14:05<CyberPsyko>ok, I'll install the complete CVS and try that... Thank you everone for putting up with me...
14:06<Chutt>morph, hey, how warm does your nehemiah run under load?
14:06-!-CyberPsyko [] has quit ["Client Exiting"]
14:06<Morph>Chutt: hmm good question, ive not checked it out exactly. it gets warm but not overly hot.
14:06<Chutt>could you check?
14:07<Morph>yeah.
14:07<Chutt>mine gets up to > 80c (according to lm_sensors)
14:07<Morph>wow.
14:07<Chutt>after about a half hour at 100% cpu
14:07<Morph>thats hot.
14:07<Chutt>yeah.
14:08<Cloak>egg cooking time
14:08<Chutt>but, stop the compile (or whatnot), and it drops down to mid 40s pretty quickly
14:08<Chutt>like, 20 seconds or so
14:11<poptix>sufficient cooling but insufficient head spreader
14:11<poptix>head/heat
14:11<Morph>hmm gotta update a bunch of stuff.
14:12<Chutt>poptix, standard with what came with the board
14:12<Chutt>i'd _assume_ it'd be fine
14:12<poptix>Chutt: yeah
14:14<Chutt>i like this case
14:14<poptix>a friend of mine setting up a dual athlon mp thought the retail cpu fans + pink crap would be sufficient too
14:14<Chutt>but i think a bigger one would be more open
14:14<Chutt>poptix, you can't easily remove the hsf from these
14:14<poptix>Chutt: epoxy?
14:14<Chutt>believe so, yeah
14:15<poptix>hrm
14:15<Chutt>and i really don't want to freeze the thing :p
14:15<Chutt>i'm going to write the via guy that sent me the board and ask about temps
14:16<poptix>it's possible lm_sensors is off by a little bit
14:16<Chutt>exactly
14:16<poptix>(or a lot)
14:16<poptix>does it have hardware monitoring in the BIOS?
14:16<Chutt>in the bios
14:16<Chutt>but, the monitor's in the chip
14:16<poptix>right
14:16<Chutt>so it's kind of hard to load it up in the bios
14:17<poptix>how does the cpu temp look in an idle bios, as opposed to an idle system
14:17<Chutt>ah, i should check, hadn't thought of that
14:18<Chutt>ah well, i'll do that later
14:22<poptix>let me know how it goes
14:22<poptix>that's how i usually calibrate my sensors
14:22<Captain_Murdoch>Chutt: does it make sense to reset framesWritten every time a keyframe is written the same way we reset framesPlayed every time a keyframe is read?
14:23<Chutt>which section of code?
14:23<Chutt>i suppose, but, i dunno
14:23<Chutt>shouldn't be an issue unless you're dropping a lot of frames
14:24<Captain_Murdoch>I'm thinking in NuppelVideoRecorder::WriteVideo when the 's' frameheader is written. shouldn't ever be a problem unless dropping frames, but it does help keep the actual frameswritten value in sync with framesplayed.
14:28<hadees>anyone use a guillmot 3d prophet? or a card with geforce 256
14:29<Captain_Murdoch>nevermind, can't do that, since tv_rec.cpp uses the number of frames written to determine filesize. I'll juse use "(fnum-startnum)>>1" when marking the blank frame since that's what the 's' frameheader uses.
14:29<Captain_Murdoch>s/filesize/filelen/
14:29<Captain_Murdoch>I'm testing the "write blank frame list to db while recording" code.
14:31<Chutt>it'll eventually be possible to do the blank frame stuff with the pvr card
14:31<Chutt>at the cost of some additional cpu usage
14:32<Chutt>since it'll be able to pull down raw YUV data in addition to the mpg data
14:34<Captain_Murdoch>how much cpu does it use to decode the mpeg2 for you?
14:34<Chutt>eh, 15% or so on the 1800+
14:35<Chutt>something like 70% for full-sized on the mini-itx
14:35<Captain_Murdoch>wonder if it would take more cpu for fetching the yuv data or just decode the mpeg2 and use that? probably the former?
14:36<Captain_Murdoch>having a playback thread running alongside the record thread could be an option, just play the video back and flag commercials there eliminating the post-record thread. not if it uses too much cpu though.
14:36<hadees>is there anyway to strech the nvidia binary drivers, they don't fill up my tv
14:37<Chutt>shouldn't take all that much to grab the yuv
14:37<Chutt>just some additional memory usage
14:37<Chutt>it's assuming, of course, that the yuv + mpeg data is in sync
14:40<Captain_Murdoch>yeah. I've got the code working to write out the blank frame list during recording, just doing a little more testing by letting some stuff record. currently it writes the new entires every 15 blank frames. just need to test the impact on my slow Athlon 700 later to see if it impacts recording any. have to step out for a while now though so will get back to it later.
14:41<Chutt>just make sure you're locking db access
14:41<Chutt>and it should be pretty minimal
14:42<Captain_Murdoch>yeah locking already before you discovered the other problem.
14:43<mdz>new mythv FAQ entry: there is no need to start every message to mythtv-dev with an explanation about whether you are using GnuPG or not
14:46<Captain_Murdoch>not sure if it's related to your patch or not, but I've had the master backend die twice on me in the past few minutes. I've been bringing up and down a slave backend as well as a 2nd frontend watching livetv to test my stuff and the main backend has died. my code was only in the slave with the master being current cvs. I'm going to leave it running while I run out and see if it's me causing it or something else. might be that
14:46<Captain_Murdoch> you were wondering about.
14:46<Chutt>what's that from?
14:47<Chutt>unless you didn't paste it =)
14:47<Chutt>get a bt
14:47<Captain_Murdoch>I didn't paste it. if I can make it happen again I'll do a bt. don't want to bother you if it's something I'm causing with my mods to the slave.
14:47<Chutt>could very well be my patch
14:48<Chutt>the slave shouldn't be making the master die, though
14:49<Captain_Murdoch>that's what I thought. the recordedmarkup table did get updated the first time though, so I wasn't sure if it was something I was causing. I'm running my master inside gdb right now and will take a look when I get back. it's about to start my wife's afternoon of TLC recordings. :) nice guinea pigs.
14:50<Chutt>heh
15:02<FryGuy>those changing rooms shows are silly
15:02<FryGuy>i don't understand why my mom watches them
15:03<FryGuy>she gets so pissed at the designers for putting paint on good wood, and they do it every time, yet she still watches them
15:04<hachi>maybe that supports the theory that people can only be happy knowing that someone else is suffering?
15:05<FryGuy>i don't know.. she just things it's bad to "ruin" the wood
15:06<FryGuy>like it is to put mustard on steak :)
15:43-!-Chutt [] has quit ["dead"]
15:55-!-LeandroD [leandro@host154-242.pool8021.interbusiness.it] has joined #mythtv
15:55<LeandroD>hello
16:02-!-hadees [] has quit [Read error: 110 (Connection timed out)]
16:14<Captain_Murdoch>FryGuy: my wife use to think it would be so cool to let them come and do a room at our house, then I kept pointing out stupid stuff they do like paint fireplaces silver & purple checkerboard and glue hay to living room walls and I eventually convinced her it's not worth the risk.
16:22-!-hadees [~blah@pcp01500702pcs.univde01.de.comcast.net] has joined #mythtv
16:24<LeandroD>let's go talk about stupid thing... what case have you choose for your living room?
16:25<Captain_Murdoch>a $27 one from compgeeks.com. It's still in it's proof of concept phase for me even though I've been using it for 6 months constantly. :) I plan on putting in a low-power setup for my main frontend and moving the backend with existing server to my server cabinet at some point.
16:26<LeandroD>what about the WAF? (wife acceptance factor)
16:28<Captain_Murdoch>she's so happy that I coded up the commercial skip functionality that she doesn't care right now. she gets frustrated watching real live tv right now without going though Myth. :)
16:28<Captain_Murdoch>I debated picking up one of these -> http://www.softwareandstuff.com/dsk_delloptigx100466.html to see if it's powerful enough, but thought about trying a spare p2-400 I have sitting here first to see if it had the horsepower to playback my video first.
16:29<Captain_Murdoch>that sucks, I just hit up-arrow after turning on debugging and doing a recompile and accidentally started another "make distclean && make" :)
16:30<LeandroD>eheheh... it is not a great mistake compared with a rm -r /
16:31<hadees>hey i am using kde and i am runing mythtv in fullscreen but when i go to watch live tv the live tv goes into a window, i can fix it by hitting the full screen button but i want it to just go fullscreen when i hit the button
16:31<hadees>for live tv
16:31<LeandroD>about commercials, just curiosity, do you live in US? how many commercials per hour are there?
16:31<Captain_Murdoch>yeah, at least I'm compiling on my faster machine on an nfs mounted dir and then just running "make install" on the backend.
16:31<LeandroD>hadees, don't use kde...
16:31<hadees>i thought there were problems with gnome
16:32<LeandroD>gnome and kde are not the only two window manager, take a look into contrib directory
16:33<Captain_Murdoch>LeandroD: looking at my recordedmarkup table, it looks like an average of 3 commercial breaks per 30 minute show, and 5 or 6 per 1-hour show. figuring 4-6 commercials per commercial break (30-second spots), then that's 12-18 per half-hour. I think it averages about 7.5 minutes of commercials per half-hour timeslot
16:34-!-Chang [~coffman@fw.pqbs.com] has joined #mythtv
16:34<Captain_Murdoch>we can normally watch a 1-hour show in about 47 minutes with commercials flagged and hitting the 'Z' key to skip them when one comes up.
16:35<LeandroD>what the hell! here in italy we are lucky :)
16:36<Captain_Murdoch>commercials used to be annoying, now I barely notice them. occasionally I will watch one or two if I see them. I still don't automatically skip them, although the detection is pretty accurate on most shows/channels.
16:37<LeandroD>I admin I haven't tried yet... usually I skip forward 30s
16:38<Captain_Murdoch>might not work too well for non-US users right now. most people say that their channels don't have blank frames between commercials or shows.
16:38<hadees>what window manager do most people use?
16:38<Captain_Murdoch>we have some here that don't, so I'm going to start working on other detection methods as well as the blank-frame stuff.
16:38<LeandroD>I know
16:39<Captain_Murdoch>fvwm2 with .fvwm2rc included in mythtv/configfiles
16:40<rkulagow_>blackbox
16:44-!-choenig [~choenig@p5085927A.dip.t-dialin.net] has joined #mythtv
16:48<LeandroD>I am looking for a cool looking htpc case, price is not important, any advice?
16:58<hachi>mdz: if you're still there, do you happen to have a liblame .deb file sitting around? as I just lost mine a moment ago
17:00<hachi>scratch that, marillat seems to be back up, though very slow
17:01-!-Chang [] has quit [Remote closed the connection]
17:26-!-mchcoastal [] has quit ["Client Exiting"]
17:28<LeandroD>'night
17:29<-- LeandroDhas quit ()
17:35-!-hadees [] has quit [Read error: 60 (Operation timed out)]
17:35-!-Drikus___ [] has quit ["leaving"]
17:50-!-Chutt [~ijr@dsl093-011-148.cle1.dsl.speakeasy.net] has joined #mythtv
17:53<Captain_Murdoch>Chutt: looks like the segfault was in the patch. the new program_info wasn't being created in time because it was moved to after the new db connection was created. I moved it back up to the top of DoFlagCommercialsThread() before the flagthreadstarted = true; line and it seems to be working right again. I did do a bt and it was dying in the "new ProgramInfo" line since by that time prevRecording was NULL.
17:53<Chutt>ahh
17:53<Chutt>right
17:53<Chutt>commit a fix?
17:54<Captain_Murdoch>not yet, testing to make sure.
17:54<Chutt>that'd be it
17:54<Captain_Murdoch>will commit the stuff that saves the blank list during recording at the same time.
17:58<Captain_Murdoch>what do you think about changing NuppelVideoRecorder to keep track of the seek table in a QMap as well as the current method, and making it write the list out to the DB every X frames. mpegrecorder.cpp could be modified as well. Then nuppeldecoder could be modified to use the db list if no seektable was in the file. soa usable seektable (in the db) would exist for .nuv and mpeg2 files even if the backend were to die.
17:58<Chutt>sure
17:58<Captain_Murdoch>doesn't sound hokey?
17:58<Chutt>i'd like it if you could commit that tiny bug fix
17:59<Chutt>before too much longer, though
17:59<Captain_Murdoch>I'll go ahead and commit. I'll know in 2 minutes whether it fixed it. I'm 99.99% sure anyway.
18:00<Captain_Murdoch>committed.
18:00<Chutt>thanks
18:01<Captain_Murdoch>just the 1-line change, will get my other stuff in a few.
18:02<Chutt>thanks
18:05<hachi>mdz: if you're there and you have a moment, say my nick please
18:06<Cloak>its a destiny's child irc day
18:09-!-bline [office-3@office.gossamer-threads.com] has joined #mythtv
18:16<Captain_Murdoch>Chutt: I am noticing a bunch of defunct processes since putting the new version into place. One is an old commercial flagging thread that finished, but the rest look like normal mythbackend threads. one had lots of cpu time so I'm guessing it was an encoder. Is that just cause I'm running under gdb?
18:18-!-choenig [] has quit [Read error: 54 (Connection reset by peer)]
18:18-!-schultmc [] has quit ["Client exiting"]
18:19<Chutt>yup
18:19<Chutt>threads don't go away properly when run in gdb
18:20<Captain_Murdoch>ok, figured it must be.
18:23<Cloak>handle SIGCHILD pass nostop noprint ?
18:31* Captain_Murdochis not a gdb expert but will try it next time
18:32<Cloak>how does this commercial flagging work? it goes through, finds blank frames and marks the locations?
18:36<Captain_Murdoch>finds blank frames then does some intelligent analysis to determine which surround commercials. then merges commercials into commercial breaks and flags them in the DB also. eventually it can/will analyze things like scene changes, audio, closed-captiioning, etc. to help detect commercials.
18:36<Cloak>neat
18:36<Cloak>vcrs have been using audio level changes for years to try to detect commercials
18:37<Captain_Murdoch>it flags blank frames at record time for software-encoding since it's not that much cpu to do that. after recording it scans the file again, this mainly benefits hardware encoders right now, but this is where the enhancements will come, so it will benefit everyone in the future.
18:37<Cloak>:)
18:38<Cloak>blank frame tends not to be a US thing?
18:39<Captain_Murdoch>some people say blank-frame is only working in the U.S.. that's another reason to add better/more detection methods.
18:39<Cloak>oh, only in the usa
18:39<Captain_Murdoch>even some U.S. channels and shows don't use blank frames around and between commercials.
18:40<Captain_Murdoch>current code works really well on some shows and channels (like close to 100% accuracy on stuff I record), but on others it doesn't work anywhere near as well as 100%)
18:41<Cloak>is there alot of heavy lifting to do lets say fft for audio to determine shifts?
18:41<Cloak>or is there a pretty clean way to extract the information?
18:42<Captain_Murdoch>it's going to be a lot more cpu intensive than the current easy blank-frame detection. code wouldn't be extremely hard, just might need to rework some of the architecture to make sure that commercial detection didn't consume cpu and cause problems recording other shows at the same time.
18:43<Cloak>gotcha, it can also mark and sweep in downtime
18:43<Cloak>doesn't need to act concurrently
18:45<Captain_Murdoch>yeah, one of the benefits of the blank-frame detection is that if it works for you, then you can skip commercials even while watching an in-progress recording. other detection methods have to be done post-record or offline.
18:47<Captain_Murdoch>they could be implemented so as to allow skipping using on-the-fly detection, but that's going to be slow, kinda like the first blank-frame detection patch I did where it took 10-15 seconds to skip a commercial break just scanning blank frames. that was before the code that checked for blank frames during the recording process.
18:49<Captain_Murdoch>dinner calls, be back later..
18:49<Cloak>c-ya :)
18:49<Cloak>thanks for the insight
18:50<Captain_Murdoch>yw, anytime
18:53<hachi>mdz: I've got a strange glitch that you'll want to know about, and I'm getting segfaults, but I'm asuming the segfaults are me doing something wrong, I however did get everything else working
19:01<bline>are their any plans to make different output displays supported such as gl2 or sdl
19:01<bline>hey hachi
19:02<hachi>doesn't this run on SDL? or am I confusing my reading with freevo?
19:02<hachi>hi bline
19:03<bline>myth uses Xv exclusivly for video
19:04<hachi>ahh, that explains the segfault then, I'll go and get a new videocard now
19:05<bline>oh, I wasn't refering to your segfaults, it was just a general question
19:06-!-hadees [~blah@pcp01500702pcs.univde01.de.comcast.net] has joined #mythtv
19:07<hachi>I think I'm serious though, this videocard doesn't run with XV
19:07-!-mdz_ [~mdz@216-15-124-77.c3-0.smr-ubr3.sbo-smr.ma.cable.rcn.com] has joined #mythtv
19:08<hachi>uhh, welcome back mdz?
20:04-!-hfb [] has quit ["Huh! Wasn't me!"]
20:08<hachi>will mythtv run without a soundcard?
20:08<hachi>I mean, a dsp of any kind
20:09<Captain_Murdoch>frontend or backend?
20:09<hachi>frontend
20:09<Captain_Murdoch>I think backend will if you have a pvr-250 or 350 since their audio is in the mpeg2 stream
20:09<hachi>though, I guess it's both, since it's the same box
20:09<Captain_Murdoch>how are you going to hear the audio?
20:10<hachi>my system is panicing
20:10<hachi>I want to elimitate one driver
20:12<hachi>do you think mythtv will tickle bugs in the bt878 driver which xawtv does not? or does it only address it roughly the same?
20:18<Captain_Murdoch>xawtv does overlay, myth does frame grabbing, they do different things with the tv card.
20:20<hachi>crap, in my house I have 1 PCI sound card, and this machine only has PCI slots in it
20:23-!-hadees [] has quit [Read error: 110 (Connection timed out)]
20:28* Cloakwhines like a pvr.
20:38<hachi>mine doesn't whine, mine simply crashes
20:39<hachi>bad sound card I'm pretty sure
20:51-!-ribo is now known as ribo`away
20:51-!-ribo`away is now known as ribo
20:51<hachi>crashes on XV failure
20:55<hachi>it's working with no sound card, I think I need to do something with dma though
20:59-!-hadees [~blah@pcp01500702pcs.univde01.de.comcast.net] has joined #mythtv
20:59<hachi>yow, I am taking up 100% of my processor time
21:03<-- ribo(~ribo@dont-panic.system42.net) has left #mythtv
21:03<Captain_Murdoch>hachi, if you don't have Xv support, then it will consume huge amounts of cpu because it has to do colorspace conversion and scaling in software rather than in hardware using the Xv driver.
21:03<hachi>nah, I fixed that
21:03<Captain_Murdoch>ok.
21:03<hachi>either dma isn't on, or I chose a really intensive codec
21:09<-- Cloak(mmen@crackaddict.com) has left #mythtv
21:15-!-froggy19 [~fryguy@12-246-43-18.client.attbi.com] has joined #mythtv
21:16-!-FryGuy [] has quit [Read error: 104 (Connection reset by peer)]
21:16-!-froggy19 is now known as FryGuy
21:40-!-Ndiin [~ruler@ip68-3-77-245.ph.ph.cox.net] has joined #mythtv
21:40<Ndiin>Hey folks.
21:41<Ndiin>Is there an iulius theme thats up to date with cvs?
21:44<hachi>wheee, it's running, sound card is sucky
21:58<rkulagow_>nziarek, are you here?
22:01-!-Chang [~Tony@a11c133.neo.lrun.com] has joined #mythtv
22:03<hachi>GAH, blocks on channel change
22:04<Chutt>soundcard, or database issue
22:04<hachi>no soundcard, could be that... but it's not because the drivers suck :]
22:05-!-Chang [] has quit [Client Quit]
22:06<hachi>I guess it's also possible that I fried the DB with all the panics
22:08-!-hadees [] has quit [Read error: 110 (Connection timed out)]
22:10-!-Ndiin [] has quit ["raven[0m[36m:[0m ."]
22:25-!-bigguy_ [bigman@h39.129.39.162.ip.alltel.net] has joined #mythtv
22:26-!-bigguy [] has quit [Killed (NickServ (Ghost: bigguy_!bigman@h39.129.39.162.ip.alltel.net))]
22:26-!-bigguy_ is now known as bigguy
22:50-!-nyquiljer [coyote@12-211-10-191.client.attbi.com] has joined #mythtv
22:51<rkulagow_>chutt: is there a simple way of dumping the thread apply all stuff directly to a file once there's a segfault?
22:52<rkulagow_>(cutting and pasting between windows is getting boring...)
23:00<Chutt>quite possibly, but i'm not sure
23:00<rkulagow_>i've been googling, but nothing is jumping out as a likely possibility
23:02<Chutt>i believe you can log the entire session to a file
23:08<Chutt>oops
23:08<Chutt>accidently recorded at 720x480
23:08<Chutt>and there's some pci bus bandwidth issues with my system at that res for analog capture
23:09<Chutt>got a black line going through the middle of the picture
23:13<sandeen_>has anyone ever seen mythbackend die with a "lame error?
23:13<Chutt>nope
23:13<sandeen_>k
23:13<sandeen_>hrm.
23:14<Chutt>turn off audio compression if it happens
23:14<sandeen_>calling into libmp3lame and dying, latest versions of everything :/
23:14<sandeen_>how?
23:14<sandeen_>:)
23:14<Chutt>tv->settings->recording
23:15<sandeen_>I'm guessing this is from mythfrontend
23:15<Chutt>yup
23:15<sandeen_>which I can't run, because when I do it kills mythbackend with a "lame error" :)
23:15<sandeen_>where are configs stored?
23:15<Chutt>um
23:15<sandeen_>sql?
23:15<Chutt>the only way it can kill it with that is if it's playing something back
23:15* sandeen_shrugs
23:15<Chutt>when you run mythfrontend, it's not playing anything back
23:15<Chutt>or recording
23:16<sandeen_>I just did mythsetup, then tried to run mythtv
23:16<Chutt>if you can't run the frontend, you can't select anything to run
23:16<sandeen_>I start mythbackend and get...
23:16<Chutt>mythtv is not mythfrontend.
23:16<sandeen_>Probed: /dev/video0 - Television
23:16<sandeen_>Probed: /dev/video0 - Composite1
23:16<sandeen_>Probed: /dev/video0 - S-Vide
23:17<sandeen_>well, either way.... let me try hte othre
23:17<sandeen_>after I start mythtv the backend dies with
23:17<sandeen_>adding: porter as a remote ringbuffer
23:17<sandeen_>Changing from None to WatchingLiveTV
23:17<sandeen_>lame error, exiting
23:17<Chutt>wow, you read really well
23:17<Chutt>mythtv is not mythfrontend.
23:18<sandeen_>oh wow
23:18<sandeen_>mythfrontend works ;-)
23:18<Chutt>no shit.
23:18<sandeen_>I'm sorry, I thought I tried both
23:24<sandeen_>so could I deign to ask what "mythtv" is for?
23:25<Chutt>just does the livetv stuff.
23:27<sandeen_>ah
23:27<sandeen_>and, in fact, when I try to watch live tv it does die with the lame error
23:27<sandeen_>*sheepish grin*
23:27<Chutt>exactly
23:27-!-billytwowilly [~chris@h24-86-147-220.ed.shawcable.net] has joined #mythtv
23:29<sandeen_>ok guess I'll have to do some debugging to figure out why it can't talk to lame
23:30<Chutt>that'd be cool.
23:30<Chutt>but, easiest thing to get around it would be to just turn off the mp3 encoding
23:30<Chutt>and, it's using liblame, not the lame binary
23:30<sandeen_>wow, you read really well ;-)
23:31<sandeen_>"calling into libmp3lame and dying"
23:31<billytwowilly>heh. my archos jukebox mp3 player hates your no mp3 policy;)
23:31<Chutt>eh
23:31<billytwowilly>mp3 encoding..
23:31<Chutt>ah
23:31<Chutt>i just didn't feel like adding that
23:31<Chutt>no reason to, really, for me
23:31<sandeen_>is there support for album art in MythMusic?
23:31<Chutt>not yet
23:31<billytwowilly>nah, it makes sense from an opensource perspective, it's just a pain in the arse for someone who is already invested in an mp3 player.
23:32<Chutt>shoulda bought a neuros
23:32<Chutt>well, not really, since they were buggy and stuff
23:32<Chutt>but they'll do ogg soon =)
23:33<billytwowilly>why don't I just strap a brick to my archos and pretend it's a neuros;)
23:34<sandeen_>how do various filesystems keep up with the video recording, any trouble?
23:34<Chutt>shouldn't be
23:34<Chutt>it's really not that much data, as long as you've got dma turned on
23:35<sandeen_>XFS has a "realtime subvolume" that allocates data in big chunks, I'm always looking for some project to try to use it :)
23:36<sandeen_>although I think dvgrab or something like that could make better use of it
23:36<Chutt>heh
23:36<Chutt>really, though, it's not that much data
23:36<Chutt>now, dumping uncompressed video to disk would stress things
23:37<sandeen_>if I read things right, can it "buffer" uncompressed video to disk if the CPU can't keep up?
23:37<sandeen_>or did I make that up
23:38<Chutt>if you're using rtjpeg
23:38<Chutt>and the cpu's falling behind
23:38<Chutt>and there's disk space, yeah, it'll skip the compression step
23:38<sandeen_>ok
23:38* sandeen_looks at his paltry dual PIII 350s :/
23:39<Chutt>ah
23:40<Chutt>you probably don't want mp3 compression on anyway =)
23:40<sandeen_>:)
23:41<sandeen_>it used to be fast :(
23:41<Chutt>almost certainly
23:43-!-billytwowilly [] has quit ["Client exiting"]
23:51<bline>nice commit message "Changing stuff." heh
23:52<Chutt>yeah, i couldn't think up a good one
23:52<bline>I should search through my cvs change long and find my worst commit messages
23:52<bline>s/long/log/
23:53<Chutt>heh
23:53<Chutt>i try, generally =)
23:53<bline>I'm pretty bad
23:54<Captain_Murdoch>Chutt: got the code tested and working to write the blank frame list out to the DB every 15th blank frame during record. Also converted all of commercial_skip.cpp into a CommDetect class and changed all calls to it. Still a few more things I'll probably do to it also as I add more detection methods.
23:54<Chutt>cool
23:55<Captain_Murdoch>debating whether to move the MARK_ #define's from commercial_skip.h to programinfo.h or somewhere else since they aren't all commercial related. any preferences? I was thinking programinfo.h or tv.h
23:56<Captain_Murdoch>they're used by both player and recorder in some cases.
23:58<hachi>what window manager are you people using with mythtv, I'm having issues with running it over none
23:58<bline>windowmaker
23:58<nyquiljer>i'm using fvwm2
23:58<Captain_Murdoch>you have to have a window manager running. I use fvwm2 with the included mythtv/configfiles fvwm2rc example.
23:59<Chutt>either would be fine
23:59<Chutt>programinfo.h would probably be more appropriate
23:59<bline>windowmaker seems to work well with no changes
23:59<Captain_Murdoch>that's what I thought. already put it there but figured I'd ask first before saying so. :)