Undersizing with Redistribution enabled

[nashcity]

Hey guys,

I was hoping someone could explain to me why the most recent movie I've run through dvdrb gave the following results. I don't really care about the undersizing so much, I would just like to know the reason why the following combination of settings results in different final sizes.

This is the constant in all the examples:

[23:33:58] Phase I, PREPARATION started.
- DVD-RB v1.27.2
- AVISYNTH 2.5.6.0
- CCE 2.70.2.12 encoder selected.
- "Steal Space from Extras" mode (33%) is enabled.
- Source: THE BRAVE ONE
- VTS_01: 2,820,474 sectors.
-- Scanning and writing .D2V & .AVS files
-- Processed 176,076 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 78.2%
- Overall Bitrate : 5,732/4,586Kbs
- Space for Video : 4,110,752KB
- Redistributing using Base_Q: 22
- HIGH/LOW/TYPICAL Bitrates: 6,607/1,571/4,586 Kbs

Example #1 - CCE SP, redistribution enabled, always use HC for redistribution pass, different matrices set up for different bitrate ranges = 4.24 G final size

Example #2 - CCE SP, redistribution enabled, different matrices set up for different bitrate ranges = 4.17 G final size

Example #3 - CCE SP, redistribution enabled, same matrix set up for different bitrate ranges (encoder default) = 4.29 G final size

Other info - When I set up different matrices for the different bitrate ranges (example 1 and 2), this is what I have set up:

Main Feature Matrix - Encoder Default
High Bitrate Matrix - Encoder Default
Med High Bitrate Matrix - Avamat 6 (AUTO-Q1)
Medium Bitrate Matrix - Avamat 7 (AUTO-Q2)
Low Bitrate Matrix - Avamat 7 (AUTO-Q2)
Very Low Bitrate Matrix - Avamat 7 (AUTO-Q2)
Matrix for Extras - Avamat 7 (AUTO-Q2)

Thanks for any help. As I said, I would just love to know the reasoning behind the final size fluctuations

[Fishman0919]

Your getting encoder saturation...- HIGH/LOW/TYPICAL Bitrates: 6,607/1,571/4,586 Kbs
the encoder is maxing out in quality. Add CCEAQM=1 to your rebuilder.ini file under the Option part.... do 3 passes with cce sp

The matrices you have select are good for lower bitrates (Avamat 6, Avamat 7) but I find are not so good for Higher bitrates. But adding CCEAQM=1 to the .ini file.... it will fix that for higher bitrates

[archaeo]

This leads to a question - If encoder saturation is the problem (as it appears to be here), does any particular matrix make as much of a difference as just checking CCEAQM=1 and ticking 3 pass? Does it matter much?
I always wondered in these situations if a High bitrate matrix (like FOX) would have more of an effect at reducing the level of saturation than just the encoder default mtx.

[Fishman0919]

This leads to a question - If encoder saturation is the problem (as it appears to be here), does any particular matrix make as much of a difference as just checking CCEAQM=1 and ticking 3 pass? Does it matter much?
I always wondered in these situations if a High bitrate matrix (like FOX) would have more of an effect at reducing the level of saturation than just the encoder default mtx.


CCEAQM=1 in short turns Avamat 6, Avamat 7 or whatever matrix you are using into a Fox like matrix when needed.

From the CCE SP guide....

What is adaptive Q-matrix switching ?
There are some cases that Cinema Craft Encoder SP cannot keep
the minimum bitrate. It is considered to be caused by that the lowest
quantization scale cannot raise the bitrate under a single quantization
matrix. To solve the problem, Cinema Craft Encoder SP automatically
creates other matrices cutting the value of each element in the
matrices by half, quarter and so on, and applies another one to produce
a much lower distortion for much higher bitrate. This switching
of quantization matrix can occur at every GOP.



Using a matrix like Fox will help with saturation but for me CCEAQM=1 does it for you when needed and you still get the benefits of the matrix you picked.

[archaeo]

From the CCE SP guide....

Using a matrix like Fox will help with saturation but for me CCEAQM=1 does it for you when needed and you still get the benefits of the matrix you picked.

And even with CCEAQM=1, I've found that that 3rd pass has been critical to bringing the final size into the correct range

[Fishman0919]

And even with CCEAQM=1, I've found that that 3rd pass has been critical to bringing the final size into the correct range

Yes it is

[nashcity]

Thanks for the help fishman.

Is there any downside to using CCEAQM=1 all the time?

Also, the quantizations matrices lists selections for the different bitrate ranges, but there is also the "Main Feature Matrix" setting. Which one does the movie use?

[jdobbs]

Custom Technology warns that it may be incompatible with some players... but I haven't heard of any. Here's a quote from the SP2 docs under "Disable Adaptive Q-Matrix Switching":Cinema Craft Encoder SP2 (later than 2.67) adjusts quantization matrix
automatically to improve the quality of image when Multipass
VBR is selected. However, some DVD player does not support such
MPEG streams as encoded with adaptive quantization matrix, and
may cause the block noise when playback. When you encounter the
problem, you can avoid it with this option.

[Fishman0919]

Thanks for the help fishman.

Is there any downside to using CCEAQM=1 all the time?

Also, the quantizations matrices lists selections for the different bitrate ranges, but there is also the "Main Feature Matrix" setting. Which one does the movie use?

Like jdobbs said... not really... some old first Gen DVD Player can have a problem with it.

The list of matrices you have select is fine with CCEAQM=1.... when cce needs to change them, it will.