Hi, I have just started to use COMSKIP and found that if i setup up GB-PVR to run comskip while recording, comskip crashes. If i use 'post processing' all is OK.
I tryied a 'test version' I found in a recent post re: 'Comskip panics...' as my recordings are also AC3, but still crashes.
Below is an extract from the log file.
################################################################
Generated using Comskip 0.79.46
Time at start of run:
Mon Jul 03 15:38:06 2006
################################################################
Mpeg: D:\Recordings\Less Than Perfect\Less Than Perfect_20060703_15301600.dvr-ms
Exe comskip.exe
Logo: D:\Recordings\Less Than Perfect\Less Than Perfect_20060703_15301600.logo.txt
Ini: comskip.ini
Detection Methods to be used:
1) Black Frame
2) Logo - Give up after 2000 seconds
3) Fuzzy Logic
4) Aspect Ratio
ComSkip throttles back from -0001 to -0001.
The time is now 1538 so it's full speed ahead!
unknown asf object (not loaded)
unknown asf object (not loaded)
found 2 streams
added new video stream(ID:1)
found audio object guid: 0x3afb65e2-0x47ef-0x40f2-0xac2c70a90d71d343 size:5344829101834344
unknown stream(ID:2), assumed to be AudioMPEG2 sound at 48 khz
First Frame
Frame: 1 MinY: 10 MaxY: 566 Ratio: 1.30
Video PTS jumped -1 frames, frame numbers may be incorrect
Resizing scene change array to accomodate 2000 frames.
Audio PTS jumped 1 frames at frame 3, repairing timeline
Video PTS jumped 1 frames, repairing timeline
Audio PTS jumped 1 frames at frame 40, repairing timeline
undeclared stream[Id 0x0]
p_peek[0]&0x80 != 0x80
undeclared stream[Id 0x0]
p_peek[0]&0x80 != 0x80
undeclared stream[Id 0x0]
p_peek[0]&0x80 != 0x80
undeclared stream[Id 0x0]
p_peek[0]&0x80 != 0x80
the 'undeclared stream' etc keeps repeating.
Many Thanks
Cheers
Scott
Comskip crashes when 'parallel processing' DVR-MS files
Erik,
Tried comskip on a TS file that was being written and it ran just fine but did not wait for the file to be closed before exiting.
I.e It started running about 30 seconds after the TS file was created and finished at around the 700 frame mark (on a PAL recording) even though the TS capture continued for another 59 minutes.
Is this what you expected?
I hoped it would realise that the file was still being written to and process the file as it was written. I guess that would be slightly tricky given the initial logo search phase involves multiple passes but it would be possible.
It would certainly speed up post processing tasks if the commercial break file was available almost as the recording finished. This would fix the most common complaint that I get from 'She who must be obeyed'
Tried comskip on a TS file that was being written and it ran just fine but did not wait for the file to be closed before exiting.
I.e It started running about 30 seconds after the TS file was created and finished at around the 700 frame mark (on a PAL recording) even though the TS capture continued for another 59 minutes.
Is this what you expected?
I hoped it would realise that the file was still being written to and process the file as it was written. I guess that would be slightly tricky given the initial logo search phase involves multiple passes but it would be possible.
It would certainly speed up post processing tasks if the commercial break file was available almost as the recording finished. This would fix the most common complaint that I get from 'She who must be obeyed'
Erik,
Been a while since I looked at this due to computer hardware problems . Now normal service has been restored I've had a play with live_tv_retries as suggested but can see no difference.
I've uploaded two verbose log files to you FTP site under a 'Concurrent TS' folder. The '2006-10-28 1434_ts test.concurrent' is the log from comskip kicked off about 30 seconds after recording started, '2006-10-28 1434_ts test.log' was a run of exactly the same command after recording was complete.
Any thoughts?
Been a while since I looked at this due to computer hardware problems . Now normal service has been restored I've had a play with live_tv_retries as suggested but can see no difference.
I've uploaded two verbose log files to you FTP site under a 'Concurrent TS' folder. The '2006-10-28 1434_ts test.concurrent' is the log from comskip kicked off about 30 seconds after recording started, '2006-10-28 1434_ts test.log' was a run of exactly the same command after recording was complete.
Any thoughts?