Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion. Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules. |
11th July 2006, 17:41 | #1 | Link |
Registered User
Join Date: Oct 2001
Location: Alabama, USA
Posts: 769
|
Finding Glitches in TS
Is there any tool that can analyze a transport stream and find glitches or missing frames in them?
I was recently trying to sync a hd capture with a DVD soundtrack and it took hours of work to track down the frame ranges of the glitches by comparing with the original frames of the DVD. Ultimately, it would have been nice if DgIndex just returned black frames in these glitchy or missing segments. Any suggestions?
__________________
ChapterGrabber - add names to your chapters | AtomSite - open source AtomPub server |
11th July 2006, 18:49 | #3 | Link |
Aging Video Hobbyist
Join Date: Dec 2004
Location: Off the Map
Posts: 2,461
|
Not sure exactly what you're attempting but two tools to look at are:
MPEG2Repair - finds glitches in the video VideoReDo Plus - deals with structural issues in transport streams There are lots of folks who record multiple showings of TV programs, and use M2R to see which of them has the fewest glitches... |
11th July 2006, 22:45 | #4 | Link |
Registered User
Join Date: Oct 2001
Location: Alabama, USA
Posts: 769
|
I familiar with all three of the tools mentioned. However, I haven't found any of them to be that reliable.
MpegRepair only works on Transport Streams and is very slow. Sometimes I'll use this before DGIndex. This is probably the most promising solution. I need to do some research on the log file it creates. VideoReDo Plus isn't free. However, I have had some success using this for some DVR-MS files but they still had delay sync issues. Also, I can't remember exactly how it handles glitches, but I think it just drops the GOP. ProjectX usually doesn't play nice with my hd recordings and I think it also drops the GOP rather than replacing with blank frames. Is it true that transport streams just don't have reliable global timestamping that can be used to easily determine this information? Also, those glitches didn't play well with the telecide().decimate() I was doing. I had to work around by trimming each section individually before the telecide().decimate().
__________________
ChapterGrabber - add names to your chapters | AtomSite - open source AtomPub server |
11th July 2006, 23:31 | #5 | Link | |
Guest
Join Date: Jan 2002
Posts: 21,901
|
Quote:
|
|
12th July 2006, 22:13 | #6 | Link | |
Registered User
Join Date: Oct 2001
Location: Alabama, USA
Posts: 769
|
Quote:
For example, if frames 238053 - 238111 were missing, couldn't this be detected by seeing that there is a gap in the timestamps. So on a global scale, you could determine all missing frames as long as consecutive timestamps were used throughout the stream.
__________________
ChapterGrabber - add names to your chapters | AtomSite - open source AtomPub server Last edited by Emp3r0r; 14th July 2006 at 02:30. |
|
13th July 2006, 00:46 | #7 | Link |
Registered User
Join Date: Sep 2004
Location: Near LA, California, USA
Posts: 1,545
|
I don't know if this is related or not but here it goes. I have transport files that play fine when played on their own but the audio is off sync when I create a .d2v file even if I use the correct ms delay setting. I found that the only way to get the audio and video aligned correctly was to use mencoder without DGDecode or avisynth. I had to create a large Huffyuv video file with audio re-encoded as well. Just demuxing the .ac3 file doesn't work. It was the only thing that worked for me.
__________________
Pirate: Now how would you like to die? Would you like to have your head chopped off or be burned at the stake? Curly: Burned at the stake! Moe: Why? Curly: A hot steak is always better than a cold chop. |
Thread Tools | Search this Thread |
Display Modes | |
|
|