+- XBMC Community Forum (http://forum.xbmc.org)
+-- Forum: Development (/forumdisplay.php?fid=32)
+--- Forum: PVR Development (/forumdisplay.php?fid=136)
+--- Thread: Testing help needed for Myth 0.25 support (/showthread.php?tid=129021)
RE: Testing help needed for Myth 0.25 support - rwats2011nz - 2012-05-02 16:15
Applied the patch and have disabled commerical skipping. A bit of a pain but at least it fully usable now. Thanks for you help.
RE: Testing help needed for Myth 0.25 support - dteirney - 2012-05-03 13:23
@rwats2011nz, the following change makes auto commskip seeking as fast as normal seeking for me. http://s3.teirney.net/xbmc/patches/commskip-flushed.patch
Can you try and let me know if it works for you as well. Not sure what has changed in dvdplayer, will need to investigate more.
RE: Testing help needed for Myth 0.25 support - dteirney - 2012-05-06 11:57
I've pushed in the Myth 0.25 protocol related changes to XBMC master now.
I'll be looking at the commercial skip seeking problem that others have been having this week.
Thanks everyone for your help. If you do have any Myth 0.25 protocol related problems either raise them here or create a trac ticket.
RE: Testing help needed for Myth 0.25 support - rdoac - 2012-05-06 16:12
Is it possible to do a patch against previous versions too? I have to go back to a tree from mid feb if I want to watch my slingbox. I would love to be able to patch in the differences and use mythbox (now updated to 0.25 and working with your tree) with it?
RE: Testing help needed for Myth 0.25 support - rdoac - 2012-05-07 13:57
Wanted to report back, we watched a recording last night using Mythbox git, which used cmyth for playback. The commercial skips were firing in the wrong place and jumping 7 mins. I am going to apply the patch and see if it's better. (No Commercial Skip would be better). Commercial skip fired if we clicked play or play+skip.
RE: Testing help needed for Myth 0.25 support - dteirney - 2012-05-07 14:27
(2012-05-07 13:57)rdoac Wrote: Wanted to report back, we watched a recording last night using Mythbox git, which used cmyth for playback. The commercial skips were firing in the wrong place and jumping 7 mins. I am going to apply the patch and see if it's better. (No Commercial Skip would be better). Commercial skip fired if we clicked play or play+skip.
I'd need a log and more detail about what the wrong place means. Note that there are some advanced EDL settings that can be configured in XBMC to help clean up the sometimes dubious commercial breaks that are detected by mythcommflag (depending on the algorithm configured). I use Logo Detection only for DVB-T recordings in NZ and found that to be the most reliable for the content broadcast over here. I strongly recommend testing the different options if you have the patience. Curiously the "All available methods" algorithm was the worst performing option based on the testing I did.
For more detail see http://wiki.xbmc.org/index.php?title=Userdata/advancedsettings.xml#.3Cedl.3E
7 minutes sounds curiously like twice the typical 3:30 commercial breaks we get over here so it's possible the frame rate for content was being guessed at half the rate when it perhaps shouldn't be. There are some heuristics applied to determine what frame rate to use to obtain the time markers from the frame markers stored in mythbackend. If the frame rate was guessed incorrectly then the positioning would be off by a factor of 2 as well. Would need a log file to see if the heuristics need adjusting.
You can find the heuristics and the log messages that should appear in the log file in the ReadEditDecisionLists() method at https://github.com/xbmc/xbmc/blob/master/xbmc/cores/dvdplayer/Edl.cpp
RE: Testing help needed for Myth 0.25 support - fiveisalive - 2012-05-07 19:27
(2012-05-07 13:57)rdoac Wrote: Wanted to report back, we watched a recording last night using Mythbox git, which used cmyth for playback. The commercial skips were firing in the wrong place and jumping 7 mins. I am going to apply the patch and see if it's better. (No Commercial Skip would be better). Commercial skip fired if we clicked play or play+skip.
I had the same problem with MythBox (skipping regardless of "Play" or "Play+Skip") and incorrect jumps in some shows. This was when I had set the MythBox options to use "streaming", but when I disabled streaming and set the local recordings directory, all was well. So there seems to be some issues with streaming vs. local recordings, I found this out in the MythBox thread: http://forum.xbmc.org/showthread.php?tid=43115&pid=850552#pid850552 It seems that the issues are with xbmc's myth:// commercial skipping code vs. the commercial skip points supplied by mythtv itself.
RE: Testing help needed for Myth 0.25 support - rdoac - 2012-05-07 22:18
The commercial skip normally works fine in mythfrontend, this leads me to believe that it was xbmc/mythbox not reading the timings correctly.
RE: Testing help needed for Myth 0.25 support - dteirney - 2012-05-08 14:29
If anyone has a USB tuner that's not working with Live TV, I've resurrected some old patches that need testing. Details at http://forum.xbmc.org/showthread.php?tid=131011
Would be great if someone happens to have a USB tuner, e.g. HDPVR, and could check if the changes are still working with a Myth 0.25 backend. The submitted patch was originally against Myth 0.24.
RE: Testing help needed for Myth 0.25 support - dteirney - 2012-05-08 14:43
I've created a new thread to track commercial skip problems as well. http://forum.xbmc.org/showthread.php?tid=131016