|
|
Before I get started, let me say that I am not complaining, I am only curious.
I was observing the bbMPEG multiplexing and cutting portion of DVD2SVCD. Based on individual settings, the cut points relative to file size are set and bbMPEG then is used to multiplex the audio and video for the size specified and proceed to the next cut point. Diring this operation the following occurs (Note, I may not be using the proper wording for these steps, but please bare with me):
1. scan the video for the entire movie
2. scan the audio for the entire movie
3. multiplex the audio and video, with offsets, for the first cut
4. add scan offsets to the file created in step 3
5. scan the video for the entire movie
6. scan the audio for the entire movie
7. multiplex the audio and video, with offsets, for the second cut
8. add scan offsets to the file created in step 7
etc.
My question is, why does it have to repeat scanning the entire movie for both video and audio for each cut pass? If you use bbMPEG as a stand-alone, you can do one scan of the entire movie for video and one for the audio. The offset and cutting then proceeds without having to again read the entire movie.
Is it possible that by rereading at each cut step you reduce the potential for audio/video sync problems? If that is the case, then the extra time spent in rereading is well worth it.
The point is the overlapping of the CDs. With automatic cutting in bbMPEG you won't get any overlapping, so DVD2SVCD doesn't make one muxing with cut-points, but a seperate muxing with offset for each MPEG-file. For this reason bbMPEG scans the entire movie every time again.
Thanks for the info, that makes sense to me. |
|