setting minimum bitrate problem

[smit_tie]

Hello,

In the .ini file I added under options this line:

min_bitrate=2500.

When DVD Rebuilder starts encoding I can see the minimum bitrate is about 2500.
But when rebuilder is finished I get an output that is about 4,65GB.
Besides that minimum bitrate line I changed nothing.
I did start a new enocde and added a line under options:

CCETargetSectors=2210000 (much lower than the default value)

But again I got an oversized dvd (4,60 GB).
I want the output to be a DVD5.
Normally with these settings (without min_bitrate added) I get DVD5 outputs.
What am I doing wrong?

[jdobbs]

??? If you set a minimum bitrate of 2500 -- and it needs to encode at, say, 2300 to fit on a DVD-5... of course you're going to oversize. It's only doing what you told it to do.

[smit_tie]

What should I do to get a DVD5 output?

[jdobbs]

Get rid of the min_bitrate setting and any other customized settings in your INI file (like your targetsectors settings). Let DVD-RB use its defaults. Then just select DVD-5 for your output size.

[smit_tie]

I am trying to encode the DVD9's from The Walking Dead Season 2.
Most of the episodes have some really dark scenes. When using the defaults I got a perfect dvd5 output but the dark scenes turned out to be very blocky.
That's why I tried to set a minimum bitrate.
Any advice to prevent this blockiness?
Should I change the numbers from VBR_Bias and Qual Prec?

[jdobbs]

Some of those series discs can really pack a lot of video and make reduction to DVD-5 a challenge. You might want to try using 3 passes since you mentioned CCE. I would also give some thought to using either a "Low Bitrate" or "Very Low Bitrate" quantization matrix (from the "Advanced" section under the OPTIONS tab).

[smit_tie]

I will try that, thanks.

[smit_tie]

I used the 'Very Low Bitrate' matrix and it is much much better now. No blockiness. Thank you so much!
What does this matrix do exactly?
Give more bitrate to dark scenes?

[jdobbs]

Look here (http://en.wikipedia.org/wiki/Quantization_(image_processing)) or just do a google search for "quantization".