low level type question

[progoth]

This isn't a complaint or a flame, I'm very happy with the quality of DVDRB, I do have a question though.

I'm pretty sure DVDRB encodes each (something) separately, each one being a few minutes in length, or so, on average.

If I'm wrong already, just move on...

So my thinking is, couldn't this negatively affect the quality of the reencode? If you have a 120 minute movie which consists of 90 "slow" minutes and 30 minutes of action, it seems like feeding the whole movie to CCE would let it boost the bitrate where it's needed...

if i'm wrong, or this is already taken into account somehow, then sorry for the noise

[Paced]

Hello, I'm pretty sure you are right in your thinking - feeding it all in at once would probably allow for more effective bitrate distribution. But I think jdobbs designed it this way to compensate for CCE Basic; something to do with its non-existant handling of chapters I think. Don't hold me on that though :)

[Sir Didymus]

@progoth:

First of all it seems to me your question is not at all a "low level" one... ;)

Then, I should say it has been posed many (too many I would say...) other times...

Suggestion: Search with something like cell, split, bitrate, allocation...

One of the recent answers is given here:
http://forum.doom9.org/showthread.php?s=&threadid=83077

[the-warriners]

I am sure once RB has worked out the reduction in size required this is applied to the original material maintaining the same bitrate ratio's - ie the 'slow' parts will already be using less bitrate than the 'fast' bits so a 20% reduction will only be a small drop in rate, whereas the 'fast' bits can take a bit more of a drop.

I think the original authoring already applied the correct bitrate distribution between fast and slow parts so no need to mess with it.

[jdobbs]

DVD-RB has an advantage over original authoring in that it has the original material available to it for analysis. That's the equivalent of having already done one or two encoding passes. So the allocation of bitrate follows the master -- and as a result overcomes the smaller encoding segment sizes.