2010-09-29, 11:29 PM
In another thread, pBS said:
Can anyone tell me precisely what this does to the file? Could leaving it enabled throw off the data in the file relative to the NPVR Timing.Info? I've become EXTREMELY frustrated trying to use comskip with NPVR lately; the skip points are sometimes so far off I have to manually skip back through the commercial break to catch 10-20 seconds of show that I missed when it skipped. I'm almost ready to throw in the towel on using comskip (yet without it I would start to question (again) why I bother with all this). Comskip itself works perfectly: if I bring up a problem file in the comskip debugger display, the detected skip points are frame-perfect from its point of view. NPVR and comskip just can't agree on the timeline in a lot of files, so NPVR skips in the wrong place half the time.
I've just set the max_repair_size to 0, but I won't know for sure right away since some recordings do seem to skip pretty accurately.
Can anyone say whether this could in fact have anything to do with this issue, or what else might cause it?
pBS Wrote:there is only 1 part of files comskip can sometimes alter, the pts timeline can be 'fixed' by comskip up to 200 frames by default..This setting is described in the comskip.ini file thusly:
set max_repair_size=0 in comskip.ini to turn that off completely...
Code:
max_repair_size=200 ; Will repair maximum 200 missing MPEG frames in the timeline, set to 0 to disable repairing for players that don't use PTS.
Can anyone tell me precisely what this does to the file? Could leaving it enabled throw off the data in the file relative to the NPVR Timing.Info? I've become EXTREMELY frustrated trying to use comskip with NPVR lately; the skip points are sometimes so far off I have to manually skip back through the commercial break to catch 10-20 seconds of show that I missed when it skipped. I'm almost ready to throw in the towel on using comskip (yet without it I would start to question (again) why I bother with all this). Comskip itself works perfectly: if I bring up a problem file in the comskip debugger display, the detected skip points are frame-perfect from its point of view. NPVR and comskip just can't agree on the timeline in a lot of files, so NPVR skips in the wrong place half the time.
I've just set the max_repair_size to 0, but I won't know for sure right away since some recordings do seem to skip pretty accurately.
Can anyone say whether this could in fact have anything to do with this issue, or what else might cause it?
server: NextPVR 5.0.7/Win10 2004/64-bit/AMD A6-7400k/hvr-2250 & hvr-1250/Winegard Flatwave antenna/Schedules Direct
main client: NextPVR 5.0.7 Desktop Client; LG 50UH5500 WebOS 3.0 TV
main client: NextPVR 5.0.7 Desktop Client; LG 50UH5500 WebOS 3.0 TV