:: #mythtv

Daily chat history

Current users (89):

abqjp, aloril, andreax, Anduin, Anssi, anykey_, Argilo, beata, Beirdo, bernard_, birds_, brfransen, Chutt, clever, coling, Cougar, dagar, danielk22, Dave123, dekarl, dlblog, elvum, fith, foobum_, foxbuntu, ghoti, Gibby, GreyFoxx, Guest64398, hads, hobiga, holomntn, iamlindoro, ikonia, J-e-f-f-A, j-rod|afk, jams, jannau, jarle, jcarlos, JEDIDIAH__, joe__, jpabq, jpabq-, jstenback, justpaul, jwhite, kc, ke^, kisak, knightr, kormoc, Kunalagon, kurre, laga, leprechau, mag0o, markk, MaverickTech, mike|2, Mousey, mrand, MythBuild, MythLogBot, NightMonkey, okolsi, pheld, poptix, purserj, reynaldo, rhpot1991, rooaus, Snow-Man, sphery, Splat1, stuarta, superm1, sutula, tgm4883, ThisNewGuy, ToadP, tomimo, tris, Unhelpful, weta, wylie, xris, ybot, _charly_
Monday, February 14th, 2011, 00:18 UTC
[00:18:59] okolsi (okolsi! has quit (Read error: Operation timed out)
[00:40:31] Dave123-road (Dave123-road! has joined #mythtv
[00:50:52] elmojo (elmojo!~elmojo@unaffiliated/elmojo) has quit (Remote host closed the connection)
[01:02:16] mrand (mrand!~mrand@ubuntu/member/mrand) has joined #mythtv
[01:10:17] dekarl (dekarl! has quit (Ping timeout: 240 seconds)
[01:14:42] dekarl (dekarl! has joined #mythtv
[01:15:10] Dave123-road (Dave123-road! has quit (Ping timeout: 245 seconds)
[01:16:22] eharris (eharris! has quit (Ping timeout: 272 seconds)
[01:22:28] voyo (voyo! has quit (Quit: Lost terminal)
[01:30:08] elmojo (elmojo! has joined #mythtv
[01:30:08] elmojo (elmojo! has quit (Changing host)
[01:30:08] elmojo (elmojo!~elmojo@unaffiliated/elmojo) has joined #mythtv
[02:29:30] Dave123 (Dave123! has joined #mythtv
[03:06:02] ghoti_ is now known as ghoti
[03:22:54] Dave123 (Dave123! has quit (Remote host closed the connection)
[03:24:36] Dave123 (Dave123! has joined #mythtv
[03:26:40] Dave123 (Dave123! has quit (Remote host closed the connection)
[04:15:36] karatekickz (karatekickz! has quit (Read error: Connection reset by peer)
[04:21:53] hovr (hovr! has joined #mythtv
[04:22:38] hovr (hovr! has left #mythtv ()
[05:23:01] cesman (cesman!~cecil@pdpc/supporter/professional/cesman) has quit (Remote host closed the connection)
[06:11:08] jmartens (jmartens! has joined #mythtv
[06:16:25] jmartens (jmartens! has quit (Read error: Connection reset by peer)
[06:40:56] cesman (cesman! has joined #mythtv
[06:40:56] cesman (cesman! has quit (Changing host)
[06:40:56] cesman (cesman!~cecil@pdpc/supporter/professional/cesman) has joined #mythtv
[06:52:17] tris (tris! has quit (Excess Flood)
[06:52:46] tris (tris! has joined #mythtv
[07:04:12] fith (fith!~iFith@gateway/shell/ has joined #mythtv
[07:16:01] andreax (andreax! has joined #mythtv
[07:21:37] hanasaki (hanasaki! has joined #mythtv
[07:22:00] hanasaki: is there a card that can get hdtv and tune in cable / not ntsc
[07:25:22] hanasaki (hanasaki! has left #mythtv ()
[07:26:10] wagnerrp: ahh... patience...
[07:28:48] markk (markk! has quit (Read error: Operation timed out)
[07:30:29] j-rod|afk (j-rod|afk! has quit (Ping timeout: 240 seconds)
[07:31:49] andreax (andreax! has quit (Read error: Connection reset by peer)
[07:35:59] j-rod|afk (j-rod|afk! has joined #mythtv
[07:42:09] markk (markk! has joined #mythtv
[08:23:36] deaman (deaman!~deaman@nat/trolltech/x-qyyexutsjgnahyzx) has joined #mythtv
[08:28:26] j-rod|afk (j-rod|afk! has quit (Ping timeout: 240 seconds)
[08:33:49] j-rod|afk (j-rod|afk! has joined #mythtv
[08:53:17] martin__ (martin__! has joined #mythtv
[09:26:19] andreax (andreax! has joined #mythtv
[09:30:41] andreax (andreax! has quit (Ping timeout: 264 seconds)
[09:33:48] andreax (andreax! has joined #mythtv
[09:45:10] andreax (andreax! has quit (Read error: Connection reset by peer)
[09:46:50] jamesba (jamesba! has joined #mythtv
[09:47:27] Goga777 (Goga777! has joined #mythtv
[09:50:16] SteveGoodey (SteveGoodey! has joined #mythtv
[09:56:26] andreax (andreax! has joined #mythtv
[09:58:23] andreax (andreax! has quit (Read error: Connection reset by peer)
[10:17:56] Diverdude (Diverdude! has joined #mythtv
[10:17:58] Diverdude: Is there any feature in myth-tv which allows some sort of integration with grooveshark or a similar service? I am thinking a music browser/searcher within mythtv which uses some online music-repository which is streamed when selected?
[10:20:56] jamesba: For people who were curious about what it was I was working on and couldn't talk about. is a white-paper detailing a specification for an HTTP-based API to allow the provision of remote clients which interface with set-top-boxes and similar devices. One of the current server implementations (and the most complete) is a pure-python server which interfaces with a slightly modified MythTV on a Myt
[10:20:56] jamesba: hbuntu box providing many of the capabilities discussed in the document. This server will be released (probably on sourceforge) some time in the next month or two once we have all the internal paperwork signed off to do so.
[10:21:37] laga: jamesba: nice, congratulations
[10:24:27] andreax (andreax! has joined #mythtv
[10:43:01] Goga777 (Goga777! has quit (Remote host closed the connection)
[10:47:48] Diverdude (Diverdude! has quit (Quit: Leaving)
[11:05:02] mike|2 (mike|2! has quit (Remote host closed the connection)
[11:05:57] mike|2 (mike|2! has joined #mythtv
[11:14:04] Goga777 (Goga777! has joined #mythtv
[11:30:18] andreax (andreax! has quit (Ping timeout: 276 seconds)
[11:33:12] andreax (andreax! has joined #mythtv
[12:09:39] Goga777 (Goga777! has quit (Remote host closed the connection)
[12:41:58] nutron-home (nutron-home! has joined #mythtv
[13:08:57] voyo (voyo! has joined #mythtv
[13:47:38] SteveGoodey (SteveGoodey! has quit (Remote host closed the connection)
[14:02:32] nutron-home (nutron-home! has quit (Read error: Operation timed out)
[14:07:21] danielk22: jamesba: cool :)
[14:08:57] jamesba: danielk22, laga: thanks
[14:10:00] jamesba: we'll be releasing a lot of stuff licensed under Apache 2.0 so as to make it as open as possible, including a fairly generic python library for building such servers. The Stuff that interacts with MythTV is, naturally, going to be released under the GPL
[14:11:40] jamesba: there are other servers which comply with the spec, but they're all internal only stuff (and not MythTV related)
[14:14:59] danielk22: jamesba: what does this provide above upnp controllers and media renderers?
[14:16:12] jamesba: it's a different thing entirely
[14:16:28] danielk22: (I just read the abstract ;)
[14:16:31] jamesba: UPnP is mostly about routing content around a network
[14:16:56] jamesba: with control elements which are about serving interfaces from the server directly
[14:17:37] jamesba: this is about totally independent interfaces on remote devices, which can be customised to fit a) the needs of the user, and b) the properties of the device
[14:18:30] jamesba: a) is pretty important to us, as the realisation that different disabled people have different and incompatible accesibility requirements was one of our motivators
[14:18:53] jamesba: so making a single UI which suits all of them would be impossible
[14:19:25] jamesba: and although a STB could have multiple UI styles to choose between it can't really have one for everyone
[14:19:34] jamesba: (less of an issue for MythTV, obviously)
[14:20:13] jamesba: so having a client designed around the user's individual needs interfacing with a server speaking a standard protocol was very tempting
[14:20:56] jamesba: UPnP doesn't really provide that without DLNA stuff, and even there the support is anaemic, and, to my mind, employing the wrong technologies to produce a fairly iffy cludge
[14:21:07] danielk22: Hmm, so is this functionally equivalent to adding an XML "UI" instead of a DNLA style HTML interface to UPnP ?
[14:21:24] jamesba: sort of
[14:21:48] jamesba: UPnP really is a different sort of technology
[14:21:59] danielk22: but I guess without all the heft of supporting ancillary UPnP stuff.
[14:22:11] jamesba: that's certainly a consideration
[14:23:51] danielk22: Thanks, I'm going to need to read the spec. Any guess as to industry interest in implementing this?
[14:24:27] jamesba: Can't talk about any specifics on that at this point, but there's been some interest expressed, nothing firm yet.
[14:25:14] jamesba: There's a good chance we'll be submitting the current spec as a member contribution to the W3C fairly soon as well
[14:26:12] danielk22: Yeah, I realize it is early on. Just trying to gouge where this is on the pure-to-applied R&D spectrum.
[14:26:53] jamesba: mid-level I'd say. It works, we have implementations (though of a quality for research, not for release as a product)
[14:27:12] jamesba: that said the version number, 0.5.1 should give a good indication that it's far from finished ;)
[14:27:27] danielk22: Hey, we're at 0.24 ;)
[14:28:34] jamesba: true
[14:29:38] jamesba: there will be a supporting document explaining the design decisions in a more high-level way
[14:30:04] jamesba: but it's been held up in approval before release (the spec itself went through with almost no comments, typical)
[14:37:49] okolsi (okolsi! has joined #mythtv
[14:39:08] SteveGoodey (SteveGoodey! has joined #mythtv
[14:46:24] jamesba: oh, a mythtv related issue. I've just checked out the latest stable 0.24 fixes branch of the git repository, and there's no myththemes directory in the root like there used to be on the svn repository. Does this mean the themes are no longer handled seperately and I can just not bother with a seperate build-step in my scripts?
[14:49:00] j-rod|afk is now known as j-rod
[14:57:02] martin__ (martin__! has quit (Remote host closed the connection)
[15:00:51] blackbox (blackbox! has joined #mythtv
[15:05:58] voyo: hi guys
[15:07:39] martin_ (martin_! has joined #mythtv
[15:07:41] voyo: I'v made some nice feature to mythweb – posibility to setup order of channels with drag&drop. But I have some questions to person who is responsible/developing mythweb. (I need to know if I implemented my things right way).
[15:08:50] voyo: any1 here ?
[15:10:09] stuartm: jamesba: actually the opposite, although those themes are still available from a repository (different one, but still under MythTV on github) we're moving towards having most themes downloaded/installed from the new theme browser within mythfrontend
[15:10:20] wagnerrp: jamesba: myththemes are now downloaded through the theme chooser, and a web service CaptM set up a few months back
[15:12:14] stuartm: two reasons for that 1) greater exposure for third party themes 2) it removes the assumption that people can dump themes with us and we'll perform all future maintenance, themers will remain responsible for their themes even after release
[15:12:45] martin_ (martin_! has quit (Remote host closed the connection)
[15:13:57] wagnerrp: while not yet implemented, there is planned a web interface that themes can be uploaded and updated by the author, and automatically validated for DTD compliance
[15:14:24] stuartm: jamesba: we'll maintain a list of themes and the download location on our url, frontend grabs the list from there and displays the preview/description in the UI, when the user choses to install the theme it will be downloaded, unpacked and installed
[15:14:29] hads (hads! has quit (Ping timeout: 255 seconds)
[15:15:32] Jordack (Jordack! has joined #mythtv
[15:16:30] hads (hads! has joined #mythtv
[15:17:02] gregL (gregL! has quit (Remote host closed the connection)
[15:29:45] markk (markk! has quit (Ping timeout: 265 seconds)
[15:46:24] markk (markk! has joined #mythtv
[15:49:48] davide_ (davide_!~david@mythtv/developer/gigem) has quit (Remote host closed the connection)
[15:50:15] davide_ (davide_! has joined #mythtv
[15:50:15] davide_ (davide_! has quit (Changing host)
[15:50:15] davide_ (davide_!~david@mythtv/developer/gigem) has joined #mythtv
[15:52:36] jwhite (jwhite! has joined #mythtv
[15:57:45] blackbox_ (blackbox_! has joined #mythtv
[15:57:50] blackbox (blackbox! has quit (Ping timeout: 240 seconds)
[16:11:26] okolsi (okolsi! has quit (Ping timeout: 245 seconds)
[16:12:42] Jordack (Jordack! has quit ()
[16:31:53] andreax (andreax! has quit (Ping timeout: 264 seconds)
[16:39:17] stuartm: eww
[16:39:50] ** dblain_ wonders if that is directed at me **
[16:44:53] dblain_ is now known as dblain
[16:44:59] birds (birds! has joined #mythtv
[16:45:05] dblain (dblain! has quit (Changing host)
[16:45:06] dblain (dblain!~dblain@mythtv/developer/dblain) has joined #mythtv
[16:46:46] jwhite (jwhite! has quit (Quit: Leaving)
[16:51:57] kormoc is now known as kormoc_afk
[16:53:29] jwhite (jwhite!jwhite@ has joined #mythtv
[16:56:15] nutron-home (nutron-home! has joined #mythtv
[16:57:39] okolsi (okolsi! has joined #mythtv
[17:06:45] kwmonroe (kwmonroe!~kwmonroe@ has joined #mythtv
[17:08:18] natanojl (natanojl! has joined #mythtv
[17:09:23] blackbox_ (blackbox_! has quit (Read error: Operation timed out)
[17:10:21] blackbox_ (blackbox_! has joined #mythtv
[17:13:16] j-rod is now known as j-rod|afk
[17:20:54] Gibby (Gibby!~Gibby@ has quit (Ping timeout: 255 seconds)
[17:22:43] Gibby (Gibby!~Gibby@ has joined #mythtv
[17:29:27] gigem_ (gigem_! has joined #mythtv
[17:29:27] gigem_ (gigem_! has quit (Changing host)
[17:29:27] gigem_ (gigem_!~david@mythtv/developer/gigem) has joined #mythtv
[17:33:36] davide_ (davide_!~david@mythtv/developer/gigem) has quit (Ping timeout: 272 seconds)
[17:34:33] danielk22: dblain: It does seem rather ugly. Do we need the whole explicit visibility cruft on Windows?
[17:36:13] danielk22: That said, a few uglies to get MythTV to compile under VS are probably worth it.
[17:36:21] Jordack (Jordack! has joined #mythtv
[17:50:02] jamesba (jamesba! has quit (Remote host closed the connection)
[18:13:57] Beirdo: danielk22: my first reaction is eewwwwww.
[18:14:24] Beirdo: but, if that's actually useful, then, yeah, it could be worth it.
[18:14:59] Beirdo: although, wouldn't it mean we are effectively supporting yet another platform (Windows without mingw)?
[18:18:26] Mousey (Mousey! has joined #mythtv
[18:21:27] andreax (andreax! has joined #mythtv
[18:24:17] kormoc_afk is now known as kormoc
[18:25:24] danielk22: Beirdo: or we just drop mingw support.
[18:25:59] Beirdo: Ummm, not sure that's a good plan.
[18:26:22] Beirdo: dropping the free compiler chain for a Microsoft one will not ingratiate ourselves to the OSS community
[18:26:25] danielk22: we can also compile under VS, but not support it. Like Solaris/BSD for many years.
[18:26:32] Beirdo: true
[18:27:15] Beirdo: but if I understand correctly, the proposal is one new header per library? That's not that bad, really
[18:27:24] Beirdo: still is a bit messy.
[18:28:11] danielk22: Beirdo: seems to me if you're running ms windows already using the vendor's compiler isn't a huge leap.
[18:29:33] martin_ (martin_!~quassel@ has joined #mythtv
[18:29:41] Beirdo: other than cost... and the image it gives. Not all OSS people are sensible, remember :0
[18:30:00] Goga777 (Goga777! has joined #mythtv
[18:30:11] Beirdo: anyways, THAT decision would have to be one that gets thoroughly discussed in email, I would think
[18:30:13] Chutt: isn't the actual compiler bit free?
[18:30:43] danielk22: Chutt: afaik, even the basic VS is gratis.
[18:31:18] danielk22: The whole suite is in the low professional software range >$700, <$1000
[18:32:52] Chutt: so it's still a gratis-free compiler, just not free-free
[18:33:30] danielk22: yup
[18:34:22] j-rod|afk is now known as j-rod
[18:37:41] dblain: just got back.
[18:39:08] dblain: danielk22: VS uses two directives to export and import symbols (dllexport & dllimport) while GCC just uses one.
[18:40:27] dblain: MPUBLIC already exists and is being used, it's just a matter of having a specific define per library.
[18:40:27] danielk22: Right, but the GCC one is optional. Can we just export all symbols under VS?
[18:41:49] dblain: Is it still optional since the use_hidesyms change went in?
[18:42:46] dblain: If I understand it correctly, if you are using GCC 4+ it will use the -fvisibility=hidden compiler option
[18:43:17] dblain: effectively requiring the use of the attribute on public classes/methods
[18:43:59] dblain: either way, I was just getting tired of re-merging my change every time there is a large update and figured I'd put the commit question out there.
[18:45:11] dblain: As for mingw, my changes are a long way off from being able to compile all parts of MythTv, so removing it is not an option right now.
[18:46:45] Beirdo: are they benign for compiling in Linux, OSX, mingw?
[18:47:16] Beirdo: if it breaks any of those, there will be people screaming for sure
[18:48:12] danielk22: dblain: Yeah, I understand. It doesn't sound too onerous, just wondering if something less complex can be done.
[18:49:32] dblain: should be, testing linux again now. It's code that has always been there, just not correct for VS. I was going to change it to also use Q_DECL_EXPORT/Q_DECL_IMPORT which is QT's way of solving this problem. (trying to use as much QT to hide platform differences as possible).
[18:49:48] dblain: But I don't HAVE to use the QT support
[18:50:02] ToadP (ToadP! has quit (Quit: ToadP)
[18:50:32] dblain: danielk22: I'm up for any suggestions for something cleaner.
[18:50:59] danielk22: dblain: heh, you won't be getting them from me. VS 6.0 was the last one I used.
[18:57:19] martin_ (martin_!~quassel@ has quit (Ping timeout: 246 seconds)
[18:59:30] ToadP (ToadP! has joined #mythtv
[19:03:09] Captain_Murdoch: dblain, what's the reason for having a myth*exp.h for each library, is that a VS-ism or for clarity
[19:05:38] dblain: no, it's needed so that it can define the proper MXXX_PUBLIC constant for the library. If it's compiling code that will be exposed, then it needs dllexport. If it's including a header from another library, those headers need MXXX_PUBLIC to be dllimport.
[19:07:25] dblain: if both used MPUBLIC, then it can only be defined as one or the other so one of the conditions will be wrong.
[19:08:10] ** dblain re-reads his comment and is confused... could of worded it better **
[19:09:05] martin_ (martin_!~quassel@ has joined #mythtv
[19:23:46] Captain_Murdoch: ok, I understood enough of it. :) thanks.
[19:26:58] nutron-home (nutron-home! has quit (Ping timeout: 272 seconds)
[19:31:28] martin_ (martin_!~quassel@ has quit (Read error: Operation timed out)
[19:36:02] kwmonroe (kwmonroe!~kwmonroe@ has quit (Ping timeout: 240 seconds)
[19:40:15] blackbox_ (blackbox_! has quit (Remote host closed the connection)
[19:52:17] kwmonroe (kwmonroe!~kwmonroe@ has joined #mythtv
[19:58:43] abqjp (abqjp! has joined #mythtv
[20:08:23] stuartm: sphery: symbol visibility does make a more obvious (though not huge) difference when you're building without debugging symbols – was your test a release build?
[20:09:52] voyo (voyo! has quit (Quit: Leaving.)
[20:13:17] stuartm: dblain: does VS require the use of those directives, i.e. it won't export/import all by default, or are they just nice to have?
[20:14:07] abqjp: iamlindoro: actually, moving the channel-change to be part of the signal-monitor is fairly new, and could be contributing to the problem.
[20:15:56] iamlindoro: abqjp: Ah, could be-- Though I still have done a fair amount of reading that suggests v4l in modern ubuntu has tons and tons of issues
[20:16:09] iamlindoro: abqjp: Crashing drivers that never crashed before, spontaneous resets
[20:16:26] laga: :92
[20:16:28] laga: oops
[20:17:40] abqjp: I run Fedora 13, and the only zero byte files I get are because my HD-PVR needs power cycled. Happens about once every two months.
[20:18:55] sphery: stuartm: don't remember--but I still think the difference is small enough to not worry about. Now, if we were making a library that's the basis of tons of applications and/or entire systems, I might feel differently. But since we're only really using symvis in libmyth* (and, IIRC, not all of those), there's not much savings to be had. I'm showing a release build of MythTV 0.24-fixes as having 41.3MiB of libs (my /usr/local/lib ...
[20:19:01] sphery: ... structure)--and that includes 11.5MiB of plugins (that aren't using symvis).
[20:19:04] iamlindoro: abqjp: In the case of thise thread, the user suggests that it affects all of his cards, including digital tuners
[20:19:15] abqjp: I do agree that Myth could do a better job of error recovery. However, it is hard to do when you are trying to please everyone. I wanted to put a time-out in the Signal Monitor, but was told I couldn't because in DVB land it can take a while for a channel to come "
[20:19:19] abqjp: "on line"
[20:21:56] stuartm: sphery: it's certainly not the primary benefit, that's avoiding the symbol noise from stuff that's entirely internal to the libraries
[20:23:09] abqjp: Myth should never get stuck in a 't' state, though. There is a 45 (I think) time-out for the actual tuning part of the operation. I should probably take a look at that — maybe with gigem's help
[20:24:12] sphery: stuartm: yeah, I just think the benefit isn't worth the cost (and I will fully admit I hate having dev-tool-specific garbage spread throughout the repo)
[20:24:20] NightMonkey (NightMonkey!debian-tor@pdpc/supporter/professional/nightmonkey) has joined #mythtv
[20:28:45] stuartm: the symbol visibility stuff appeals to the part of me that likes neat and tidy code, but that's in it's current form (GCC only atm)
[20:29:39] stuartm: it forces people to think about the public API of the library they are creating etc, not nearly as much as I'd like, but still
[20:37:13] deaman (deaman!~deaman@nat/trolltech/x-qyyexutsjgnahyzx) has quit (Quit: deaman)
[20:47:01] elmojo: sphery, abqjp, iamlindoro: yes, I'm running 10.04 Ubuntu and only had to power-cycle one of my HD-PVRs once in about 5 weeks of use – my other tuners (dvb pci) are rock solid and haven't encountered any issue at all
[20:47:40] elmojo: abqjp: do you have any ideas why the "Payload start" takes so long with the HD-PVR?
[20:47:48] elmojo: sounds like it might be out of our control
[20:47:53] iamlindoro: elmojo: as a counterpoint, I have one DVB tuner that was rock solid on 10.04 and earlier, which now resets the driver every time you try to record
[20:48:06] iamlindoro: my dmesg is full of firmware re-re-re-re-reuploads
[20:48:13] elmojo: for 10.10?
[20:48:15] iamlindoro: yep
[20:48:30] kwmonroe (kwmonroe!~kwmonroe@ has quit (Ping timeout: 240 seconds)
[20:48:38] elmojo: maybe we should post a warning or put it on the website to avoid 10.10
[20:48:40] abqjp: elmojo: I am guessing that it may have something to do with keyframes being over 2 seconds apart.
[20:48:59] sphery: elmojo: coupled with hearsay from the list (where it seems most who are reporting the issues upgraded to 10.10 /and/ 0.24-fixes at the same time)...
[20:49:22] iamlindoro: elmojo: I'd hate for my single data point to be responsible for a warning like that-- what we really need is people to go from .23 to .24 on 10.04, then test the same upgrade on 10.10, etc.
[20:49:29] sphery: but--at least for me--it's all conjecture and I have no proof (and it may well be upstream kernel or V4L/DVB changes that just haven't settled out right or something)
[20:49:40] elmojo: abqjp: oh yeah, forgot about that, isn't there a backend setting to tell it to wait for that or not?
[20:49:43] abqjp: elmojo: To me, the "proper" fix would be no interruption at all .... Have Myth do all the preparation work for the new file before making the "switch", so it is seemless. Might take spawning a thread to do the work.
[20:50:47] elmojo: abqjp: sounds scary – seemless file switching for a stream would be cool though – right now this causes the user to lose 2 – 9 seconds of recordings at the transition point
[20:52:45] tgm4883: elmojo, iamlindoro i've had to powercycle my HDPVR on 10.04 w/ 0.24 as it will apparenlty get stuck "recording" and all following recordings are 0-byte length
[20:52:54] tgm4883: as it looks like that is what you are discussing
[20:53:13] iamlindoro: tgm4883: Resetting HD-PVR hardware is aside from all that
[20:53:27] iamlindoro: the HD-PVR hardware getting hung up isn't something we can do anything about
[20:53:27] elmojo: tgm4883: we are discussing the state of the dvb drivers in recent distro/kernel releases
[20:53:32] abqjp: I have not looked at the code, so I don't know how scary it would be. I did notice that you get that two second pause even if you are behind Live, when the file switch happens. So, it is not just loosing some of the show, but the delay in starting to *read* from the new file. That delay in reading could be harder to fix.
[20:53:45] iamlindoro: The issue at hand here appears independent of tuner type
[20:54:26] Goga777 (Goga777! has quit (Ping timeout: 240 seconds)
[20:54:27] elmojo: abqjp: yes, the actual playback ringbuffer switch is an entirely different issue
[20:54:29] iamlindoro: But *does* tend to coincide with a total distro upgrade, often to the latest, which is in turn blamed on myth as the most obvious probably culprit-- just that in this case, it may not actually *be* the culprit since little has changed that could have caused such a pandemic
[20:55:10] elmojo: tgm4883: so when your HD-PVR gets in the "dead" state does cat'ing the video device yield video?
[20:55:22] elmojo: wondering if it's the control side that is getting hosed
[20:55:28] sphery: abqjp: out of curiosity, which distro/version are you using?
[20:55:33] abqjp: Fedora 13
[20:55:34] tgm4883: elmojo, haven't tried, but I'll certainly do it next time to test
[20:56:08] elmojo: tgm4883: even if it's still recording that doesn't help us if we can't communicate/control the device
[20:56:26] sphery: someone on the thread mentioned seeing the problems after upgrading to F14 (from F13)... maybe it's just upstream kernel or V4L/DVB or drivers or ... Anyone want to compare version info for F13->F14 and Ubuntu 10.04 -> 10.10?
[20:56:32] tgm4883: elmojo, ok, the only indication that I see that it thinks it's recording is the blue halo is lit up
[20:56:52] abqjp: tgm4883: I have a patch which (in theory) tells the HD-PVR to switch to a different video input, and then back, when it has trouble with the HD-PVR. I have not had any luck testing it though, since I only have that problem about every 7–9 weeks — So I don't know if it helps or not.
[20:57:39] elmojo: abqjp: are you saying that since you've put your patch into use you haven't had an issue? :)
[20:57:42] tgm4883: abqjp, yea it would be nice to know if something common causes it that we can reproduce
[20:58:09] tgm4883: elmojo, the issue is infrequent
[20:58:18] abqjp: elmojo: according to my logs, that code has not been triggered.
[20:58:48] elmojo: ah, cool, should be interesting when it happens if it helps
[20:58:58] abqjp: tgm4883: if you want to try the patch give me your email address and I will send it to you.
[20:59:39] elmojo: abqjp: I'd like to try it too – between the 3 of us we should be to trigger more quickly
[20:59:53] elmojo: you can send to my alias
[21:00:12] abqjp: elmojo: will do.
[21:00:54] tgm4883: abqjp, possible bad side effects of the patch?
[21:01:08] abqjp: I can't imagine any.
[21:01:13] kwmonroe (kwmonroe!~kwmonroe@ has joined #mythtv
[21:01:28] tgm4883: ok, shoot it to my email, i'll review it and talk it over with superm1
[21:01:45] tgm4883: I'll see if I can get some other volunteers as well
[21:02:01] abqjp: I suppose it is possible that the first input switch could succeed, and the switch back could fail...
[21:02:31] abqjp: It may have no effect at all, but it is the only thing I can think of to "reset" the HD-PVR in software.
[21:03:41] SteveGoodey (SteveGoodey! has quit (Remote host closed the connection)
[21:05:57] sphery: sounds like you guys may be able to trigger the patch more easily by upgrading to F14 or Ubuntu 10.10 :)
[21:06:24] birds_ (birds_! has joined #mythtv
[21:07:35] elmojo: abqjp: one of my HD-PVRs failed while I was away this past week – I have the logs – are they of any interest to you?
[21:08:33] tgm4883: abqjp, elmojo would you want to gather any version info on other packages besides mythtv to assist in this issue?
[21:08:56] birds (birds! has quit (Ping timeout: 276 seconds)
[21:09:05] elmojo: abqjp: this repeated in my logs until I rebooted -> 2011-02–12 20:03:56.051 AnalogSM(/dev/video-hdpvr0), Error: Start encoding failed
[21:09:05] elmojo: eno: Resource temporarily unavailable (11)
[21:09:31] elmojo: abqjp: is that the issue your patch tries to help with?
[21:09:32] abqjp: elmojo, Five of those in a row would trigger the new code
[21:09:40] elmojo: awesome!
[21:10:10] abqjp: elmojo: is you email elmojo at myth ?
[21:10:41] elmojo: sent PM
[21:12:15] abqjp: I sshed into my home machine and got the patch, so you should have it.
[21:12:40] elmojo: abqjp: does your patch save the current recording?
[21:15:06] abqjp: I am not sure I understand the question. When the recording starts, the Signal Monitor is started to make sure the HD-PVR is producing valid output. The recording won't start until the Signal Monitor indicates that everything is fine. That patch tells the Signal Monitor to toggle the input of the HD-PVR if it is getting those resource errors.
[21:15:29] elmojo: abqjp: ok, I understand now
[21:17:44] tgm4883: hmm, i wonder what happens if it toggles it and the errors continue
[21:17:52] tgm4883: keeps toggling thousands of times?
[21:18:49] abqjp: tgm4883: it would toggle once for every 5 resource errors.
[21:19:15] tgm4883: abqjp, hmm, might want to build something in there as a limiter
[21:19:46] tgm4883: just seems bad to toggle it repeatedly as that could be thousands of times if it doesn't stop the errors
[21:19:56] abqjp: tgm4883: agreed. This was really just a proof of concept. Spur of the moment for me to offer to let others try it.
[21:20:53] abqjp: tgm4883: if you are comfortable modifying the code, then go for it. Otherwise, I can send you a more sane implementation when I get home.
[21:21:22] tgm4883: abqjp, i'm not able to look at it right now
[21:21:32] Jordack (Jordack! has quit ()
[21:21:46] abqjp: np. I will try to remember to send you something new tonight.
[21:21:53] tgm4883: I'm assuming it is C, which isn't Python, so I'll probably be no good at editing it :/
[21:22:20] abqjp: tgm4883: yup
[21:28:27] martin__ (martin__!~quassel@ has joined #mythtv
[21:32:39] sphery: tgm4883: as far as version differences between F13 and F14, j-rod may be able to help collect those?
[21:41:19] iamlindoro:
[21:41:22] iamlindoro: Someone post that in theming
[21:41:26] iamlindoro: since I can't :P
[21:42:48] iamlindoro: Captain_Murdoch: ^^
[21:43:00] iamlindoro: thanks :)
[21:43:27] iamlindoro (iamlindoro! has quit (Changing host)
[21:43:27] iamlindoro (iamlindoro!~iamlindor@mythtv/developer/iamlindoro) has joined #mythtv
[21:44:00] j-rod: whassat?
[21:44:48] iamlindoro: Just some random goofing around since I am not a big fan of our logo
[21:45:50] wagnerrp: i know its our namesake, but ive never really liked that phrase 'mythical convergence device'
[21:46:57] j-rod: must have netflix for true convergence...
[21:47:07] kwmonroe (kwmonroe!~kwmonroe@ has quit (Quit: Ex-Chat)
[21:47:24] wagnerrp: how could you possibly consider building a home theater application without netflix support???
[21:47:25] j-rod: but yeah, the logo really could use a refresh
[21:48:05] j-rod: the 'launch vmware player' approach is interesting
[21:48:40] wagnerrp: i was actually thinking about that a few days ago... would qtwebkit be able to support silverlight on windows/osx?
[21:48:44] j-rod: I presume mythnetvision could do netflix if there were appropriate silverlight support
[21:48:59] j-rod: heh. exactly what I was thinking :)
[21:49:02] iamlindoro: assuming that silverlight support extended to Qt/Webkit, sure
[21:49:25] j-rod: webkit and silverlight definitely get along fine
[21:49:36] iamlindoro: If their streaming catalog has an API, it could theoretically already work on Windows/OS X
[21:49:46] j-rod: at least, there's a silverlight plugin that works fine with netflix under safari on mac os x
[21:50:01] iamlindoro: Would need some minor tweaks to MNV to handle the authentication parts, but probably not all that much
[21:50:06] j-rod: its got a pretty rich api, from what I understand
[21:50:14] iamlindoro: Well, sure, but Webkit != Qt/Webkit
[21:50:21] iamlindoro: close, but not identical anyway
[21:50:31] j-rod: yeah, that's why I was qualifying it
[21:50:54] iamlindoro: Anyway, probably wouldn't be the world's toughest project to add, but I don't have Netflix so no interest
[21:51:17] iamlindoro: (which obviously again presumes Qt/Webkit + Silverlight on Mac and Windows are all cool)
[21:52:51] wagnerrp: bug reports would indicate its at least partially functional...
[21:55:43] iamlindoro: Maybe I could get someone to bankroll a few years of netflix if I add support ;)
[21:55:46] j-rod: plex has a netflix "plugin"
[21:56:08] j-rod: not sure exactly what its using under the covers though
[21:56:09] iamlindoro: Then again, that would also presuppose fixing issues with python bindings installing on both windows and OS X
[21:56:21] iamlindoro: Which means free netflix for wagnerrp too ;)
[21:56:39] j-rod: heh
[21:56:46] j-rod: they do offer a month free to try it out
[21:57:03] j-rod: I don't care enough to bankroll it myself, I've got umpteen other devices that do netflix now
[21:57:46] j-rod: wd tv live hub is working reasonably well for mythtv recordings playback over upnp, netflix streaming, and movies via local copies
[22:00:39] wagnerrp: iamlindoro: yeah, i should look into how to install those properly one of these days
[22:00:45] wagnerrp: i dont think it would take much
[22:00:56] Unhelpful (Unhelpful!~quassel@rockbox/developer/Unhelpful) has joined #mythtv
[22:01:08] wagnerrp: i dont think any of the actual code uses stuff that wouldnt work on linux
[22:01:28] iamlindoro: wagnerrp: It tends to be the primary limiting factor in getting MNV to work on those platforms-- or rather, it theoretically is the only one, as MNV should work fine on either so long as its deps are met
[22:03:06] wagnerrp: the problem i have is that every windows application ive ever seen that uses python includes its own copy of the interpreter
[22:03:10] wagnerrp: i dont want to do that
[22:03:46] Unhelpful: haven't used mythtv in ages, and i'm trying to set it up on a new system. mythtv has appropriate GRANTs on mythconverg, but it hangs filling the initial DB. -v database shows *no* queries executed after it gets the schema lock, it just prints 'Upgrading to MythTV schema version 1226' and then sits.. this is mythtv v0.24 on arch linux. any ideas?
[22:03:53] wagnerrp: i dont know what the proper mechanism for finding it through the environment would be
[22:04:11] iamlindoro: Unhelpful: wrong channel, see /topic
[22:04:36] Unhelpful: oh! terribly sorry, didn't even notice! actually used to develop, but clearly this is a user problem. ;)
[22:05:21] martin__ (martin__!~quassel@ has quit (Remote host closed the connection)
[22:37:53] natanojl (natanojl! has quit (Ping timeout: 264 seconds)
[22:55:15] j-rod: wagnerrp: I care not about what Windows does. Mac OS X at least has a sane python of its own out of the box. :)
[22:55:34] j-rod: (granted, its a somewhat dated one now — 2.6.1)
[23:03:24] sphery: markk: Why I think we should just drop the animation rate and sort out the broken animations later: (and that topic has been coming up, regularly, since 0.22 or so).
[23:25:30] Dave123 (Dave123! has joined #mythtv
[23:35:07] Kunalagon (Kunalagon!~Kunalagon@ has joined #mythtv
[23:45:41] j-rod is now known as j-rod|afk
[23:54:24] markk: sphery: I couldn't agree more – but after last week's discussion, I decided that I needed to step away from the UI stuff. It's a can of worms. I'm more than happy to help out on any rendering/painting type issues and implementations but I just don't have the time to start pushing/leading on it
[23:55:41] sphery: heh, fair enough. I'll try pushing stuartm to let us change it to /30 (or some other value the UI guys can agree on) and see what breaks
[23:56:05] sphery: (knowing full well he's said he doesn't have any connection to the /70 that's there, now)
[23:56:56] sphery: just want his (and iamlindoro's and ...) agreement since he has a couple of themes that may need fixing in the wake of the change
[23:57:44] iamlindoro: I've never had any control over animation speed before, I doubt I'll have to make any changes :)

IRC Logs collected by BeirdoBot.
Please use the above link to report any bugs.