An idea. Needs somebody to implement it.

[AngusMacGyver]

Hi, guys.
I don't know if anybody has thought that already...
When we convert DVD -> (S)VCD vbr, we have problems to know the final size of file, unless we use 2 pass vbr.
I think we could in someway use vobs from DVD to know in wich parts movie needs more bitrate and where it needs fewer. I think it could be used in someway by TMPGEnc, like a table, to determine the actual bitrate of a part of film, proportionally to the average of whole film in (S)VCD. It would be necessary an amount of programming, but I think it would be possible.
I'd like to see your impressions about it.

[sterno]

Uh, you're more or less describing the first pass of two-pass VBR encoding. The first pass is largely an analysis where it decides how much bitrate in the final output everything will get.

[AngusMacGyver]

But there would be only one pass. TMPEnc would use information of vobs to determine bitrate of a given scene.
I hope you understand me.

[ViperXT]

Seems like a good idea to me. Would cut the encoding time in half. If I had the programming know-how I would help.

[wway]

So, you're saying use the vbr information already encoded in the vob file as a guide when encoding with tmpgenc?

Lets say your DVD movie is encoded at 5000 average. And you need to encode at 1700 average to fit the movie on 2 cd-r's. The encoder would follow the bitrate change in the vob file (on a percentage basis) and adjust your bitrate setting by the same percentage.

Hmmm. You would have a one-pass vbr encode that matches the original vob. Interesting.

I'm not knowledgeable enough to tell you if this can be done.

[Cornucopia]

IIRC, this is how ReMPEG2 already works.

Would be nice if it was updated/bugfixed, or if TMPGEnc were to incorporate some of those features.

Scott