cce 2.70.02.04 and matrices

[Insomniak4700]

I was wondering.. since CCE 2.70 has "Adaptive Q-Matrix" switching, what happens if you choose a custom matrix for DVD rebuilder?

Will the selection be ignored by CCE? Or will CCE properly use the selected matrix?

Thanks!

[Boulder]

CCE will use the selected matrix as the base matrix from which it creates the adapted matrices. The matrix you choose in DVD-RB will be the one that quantizes the most, the adapted matrices will have lower values.

[Insomniak4700]

Thanks! That could maybe explain why i'm getting so-so results with 2.70 vs 2.67 using the avamat7 matrix..

Is there anyway to disable the feature when using rebuilder? RBObt maybe?

[feedback]

Hmmm! I didn't know that.

@Boulder
How does it work when one has chosen more than one matrix?

For instance, one matrix for the main movie and a different matrix for the extras.

Regards,

[jdobbs]

Is there anyway to disable the feature when using rebuilder? RBObt maybe?
Look in ther REBUILDER.TXT file for information about the CCEAQM switch that can be set in the REBUILDER.INI file.

[Boulder]

Hmmm! I didn't know that.

@Boulder
How does it work when one has chosen more than one matrix?

For instance, one matrix for the main movie and a different matrix for the extras.

Regards,
The idea is still the same. Whatever matrix DVD-RB feeds to CCE in the cell that is being encoded, it will be used as the base matrix.

[feedback]

I understand...thanks Boulder.;)

RB V0.81 - Added the "adjust_q_matrix=0" parameter to ECL files to
prevent CCE from using adaptive quantizer matrices as
there could be incompatibilities on some DVD players.

RB V0.83 -Added a new "hidden" setting in the [Options] area of
REBUILDER.INI for enabling CCE's adaptive quantizer
matrices feature: "CCEAQM=1" -- if you decide to use
this feature, be very careful. The reason it is
"hidden" is because, according to the CCE notes, it
might be incompatible with some DVD players.

Am I correct in noting the change above in RB v.0.81 does not apply to the changes that were made in CCE 2.70, as CCE 2.70 was developed later?

Is the below correct to turn off Adaptive Quantizer in CCE 2.70?
"CCEAQM=0"

Regards,

P.S. Just started playing around with CCE ...it is fast!

[jdobbs]

It applies. If you don't have CCEAQM=1 enabled in the INI file -- it is never being used when you encode. You can double check by looking at the ECL file. If it says:

adjust_q_matrix=0

it means that the AQM is not enabled.

By the way... CCE v2.70 came out in January, 2005. DVD-RB v0.81 came out in April, 2005.

[feedback]

By the way... CCE v2.70 came out in January, 2005. DVD-RB v0.81 came out in April, 2005.
Yes, I goofed on the release dates.:)

I have more Matrices than I should have in my Rebuilt disk.

I have enabled the full backup mode using CCE’s standard matrix for the main movie and the CCE Ultra Low matrix for the extras. After processing with DVD-RB I extracted the matrices from my completed disk (There was no preprocessing).

Looking in the ECL file I see the adjust_q_matrix=0.
I also see the two Matrices that I picked for the encode in the ECL file...one for the main movie and one for the Extras.
However, the AQM still seems to be active as Extras that I have not blanked with the video segment editor have different matrices applied.
I don’t usually use RB in full backup mode…mostly movie only.
The Extras were really low bitrate…around 1500 or less. The movie was “After the Sunset”. The AQM did not apply different matrices on all the Extras just some of them.
VTS 4_01 was one where the following matrices were used…

Intra Luma and Chroma Matrix:
8 9 10 11 12 13 14 15
9 10 11 12 13 14 15 16
10 11 12 13 14 15 16 17
11 12 13 14 15 16 17 18
12 13 14 15 16 17 18 19
13 14 15 16 17 18 19 20
14 15 16 17 18 19 20 21
15 16 17 18 19 20 21 22

NonIntra Luma and Chroma Matrix:
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16

Intra Luma and Chroma Matrix:
8 16 19 22 99 99 99 99
16 16 22 24 99 99 99 99
19 22 26 27 99 99 99 99
22 22 26 27 99 99 99 99
22 26 27 29 99 99 99 99
26 27 29 32 99 99 99 99
26 27 29 34 99 99 99 99
27 29 35 38 99 99 99 99

NonIntra Luma and Chroma Matrix:
16 17 18 19 99 99 99 99
17 18 19 20 99 99 99 99
18 19 20 21 99 99 99 99
19 20 21 22 99 99 99 99
20 21 22 23 99 99 99 99
21 22 23 24 99 99 99 99
22 23 24 26 99 99 99 99
23 24 25 27 99 99 99 99

Intra Luma and Chroma Matrix:
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8

NonIntra Luma and Chroma Matrix:
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8

The second matrix is the CCE Ultra Low matrix that I ticked for the Extras in RB, I don’t know where the other matrices for this VTS came from.:confused:
As I don’t normally do a full backup it is not a big concern for me I was just wondering if the adjust_q_matrix=0 only applied to the main movie.

It appears that CCE 2.70 will still try to improve your Extras when the bitrate goes really low (1200-1500).
As I said not all the Extras had the added matrices.
Still, the Extras that did have the added matrices did look better than the ones that only had the CCE ultra low matrix.

The main movie looked fine and only had the one matrix that I ticked in RB.
Hmmm! It's probably just something weird with CCE 2.70.

Regards,

P.S. I also did a full backup with "In Good Company" with a similar result.. the VTS_03_1 had two matrices.

I still love RB Pro.:)

[Rockas]

@feedback
How many passes have you used?

[jdobbs]

If you have the defaults set, any VTS that is less than 50MB will be copied intact -- so the matrices will not change from their original state.

[feedback]

@Rockas
Two pass...I also have something else to discuss with you in a moment.

O.K. lets use "In Good Company" as an example (NTSC R1).

I used CCE's Standard matrix for the main movie in my full backup.
I used CCE's Very Low bitrate matrix/#7 Low_High(1800-2500) matrix for the Extras.

VTS_02_1 which is 304MB is the Extra. It is the Deleted Scenes Extra that starts with the first frame showing "SCENE 8 INT. AIRPLANE DAN, LOUIE, AND MORTY".
The following are the two matrices that I extracted.
Intra Luma and Chroma Matrix:
8 16 19 22 26 27 99 99
16 16 22 24 27 19 99 99
19 22 26 27 29 34 99 99
22 22 26 27 29 34 99 99
22 26 27 29 32 35 99 99
26 27 29 32 35 40 99 99
26 27 29 34 38 46 99 99
26 27 29 34 38 46 99 99

NonIntra Luma and Chroma Matrix:
27 29 35 38 46 56 99 99
16 17 18 19 20 21 99 99
17 18 19 20 21 22 99 99
18 19 20 21 22 23 99 99
19 20 21 22 23 24 99 99
20 21 22 23 25 26 99 99
21 22 23 24 26 27 99 99
22 23 24 26 27 28 99 99

Intra Luma and Chroma Matrix:
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8

NonIntra Luma and Chroma Matrix:
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8
8 8 8 8 8 8 8 8

The first matrix is the CCE Very Low Bitrate Matrix/#7 Low_High (1800-2500) matrix. I don't know where the second matrix is coming from. As I indicated the added Matrices only occur in the Extras not the main movie.
Someone else give it a go and see what you get. BTW, it is not specfic to this movie.

Now, Rockas in checking out the CCE very low bitrate matrix I found that in DVD Rebuilder's Matrix Editor a mistake was made in the
#7 Low_High Matrix which is discribed as "CCE's Very low matrix and is a preferred matrix for heavily compressed extras. Good in the 1800-2500 range".

It should be this...

8 16 19 22 26 27 99 99
16 16 22 24 27 29 99 99
19 22 26 27 29 34 99 99
22 22 26 27 29 34 99 99
22 26 27 29 32 35 99 99
26 27 29 32 35 40 99 99
26 27 29 34 38 46 99 99
27 29 35 38 46 56 99 99

16 17 18 19 20 21 99 99
17 18 19 20 21 22 99 99
18 19 20 21 22 23 99 99
19 20 21 22 23 24 99 99
20 21 22 23 25 26 99 99
21 22 23 24 26 27 99 99
22 23 24 26 27 28 99 99
23 24 25 27 28 30 99 99

But instead we have this...

08 16 19 22 26 27 99 99
16 16 22 24 27 19 99 99
19 22 26 27 29 34 99 99
22 22 26 27 29 34 99 99
22 26 27 29 32 35 99 99
26 27 29 32 35 40 99 99
26 27 29 34 38 46 99 99
26 27 29 34 38 46 99 99

27 29 35 38 46 56 99 99
16 17 18 19 20 21 99 99
17 18 19 20 21 22 99 99
18 19 20 21 22 23 99 99
19 20 21 22 23 24 99 99
20 21 22 23 25 26 99 99
21 22 23 24 26 27 99 99
22 23 24 26 27 28 99 99

Manono was going a little blurry eyed when he came up with this one.:D :)

[jdobbs]

There's no way for DVD-RB to tell CCE what matrix to use other than through the ECL file... so it has to be CCE deciding to use the other matrix on its own. Interesting. I'd guess that the only reason you are seeing it in the extras (but not the main feature) is because the bitrate on the main feature is high enough to not require encoder intervention.

[jdobbs]

One other possibility I just thought of... is there a STILL frame associated with the VTS? STILLs are not reencoded but are brought over intact (and that looks like a matrix that may be used for a STILL). Look in your log file and see if there was a line saying "Extracting STILL" during the encode phase for that VTS.

[Rockas]

Ummmm... you are right... the values are different :p

About the "where did it came from" matrix (:D)... I really can't explain that... I've been extremely occupied on the last couple of weeks and my tests came to a complete 0.

One advice... you should try 3 passes when using CCE's adaptive quantizer... I remember to make some tests on that area... the final matrices were a little different between 2 and 3 passes... don't ask me why... I don't even remember wich matrices I was using...

I hope that next week my dayjob will leave me a few more free time :)

[feedback]

One other possibility I just thought of... is there a STILL frame associated with the VTS? STILLs are not reencoded but are brought over intact (and that looks like a matrix that may be used for a STILL). Look in your log file and see if there was a line saying "Extracting STILL" during the encode phase for that VTS.

- Extracting STILLS for VTS_02 segment 10

Wow! you are good.http://users.pandora.be/eforum/emoticons4u/happy/1197.gif

And therein lies the mystery uncovered.:)

[manono]

Good find, feedback. It looks like the kind of careless mistake I might make. So, for everyone that might use that matrix (I use it for almost all my extras), you'll have to go and make the change in the Matrix Editor, Matrix Folder, 7-Low_High(1800-2500).mtx:

08 16 19 22 26 27 99 99
16 16 22 24 27 29 99 99
19 22 26 27 29 34 99 99
22 22 26 27 29 34 99 99
22 26 27 29 32 35 99 99
26 27 29 32 35 40 99 99
26 27 29 34 38 46 99 99
27 29 35 38 46 56 99 99
16 17 18 19 20 21 99 99
17 18 19 20 21 22 99 99
18 19 20 21 22 23 99 99
19 20 21 22 23 24 99 99
20 21 22 23 25 26 99 99
21 22 23 24 26 27 99 99
22 23 24 26 27 28 99 99
23 24 25 27 28 30 99 99

I don't remember if I was blurry eyed since it was a while back, and making a mistake like that is understandable when you're typing countless numbers in a short period of time. But it does speak to a lack of quality control on my part, and I humbly apologize.

I like that matrix a lot for extras. But I find that Ultra-Low Bitrate Matrix you tried out to be pretty bad. I tried it a couple of times, but it gave everyone clay faces. Facial detail was pretty much destroyed, and everyone looked like they had on masks.

[feedback]

No need to apologize my friend. Nobody and I mean nobody is perfect.:)

I appreciate your work involved in compiling the Matrices with the added descriptions of when to use them. Rockas and your efforts with the DVD Rebuilder Matrix Editor have given the community a very useful tool to use in conjunction with RB. ;)
BTW, I love playing around with different matrices.

On another note you are so right about the CCE Ultra low matrix. IMO it looks like merde. That was the first time I had used it.

Also, have you seen the following Matrix? I have started seeing it used more often in movies like "The Island, War of the Worlds, Lord of War" and others. I call it the "DreamWorks Matrix" as that seemed to be who was using it the most.
It seems to be similar to #2 High/Medium (4000) Matrix, but is used where a bit more compression is needed.

08 08 10 11 13 14 15 17
08 08 11 12 14 15 17 19
10 11 13 14 15 17 17 19
11 11 13 14 15 17 19 20
11 13 14 15 16 18 20 24
13 14 15 16 18 20 24 29
13 14 15 17 19 23 28 35
14 15 18 19 23 28 35 42

08 09 09 10 10 11 11 12
09 09 10 10 11 11 12 12
09 10 10 11 11 12 12 13
10 10 11 11 12 12 13 14
10 11 11 12 13 13 14 14
11 11 12 12 13 14 14 15
11 12 12 13 14 14 15 16
12 12 13 14 14 15 16 17

Regards,

P.S. As the thread title speaks to Matrices I don't believe we are totally off topic.:)

[manono]

Yeah, that one is used quite a lot these days on new movie DVDs, as part of the Standard Matrix with AQ turned on. That's what you get with the Standard Matrix numbers halved. In addition, you also get quite a bit of Standard Matrix with the numbers quartered. You won't often see the Standard Matrix itself in such encodes, as it'll usually only be used during the most complex scenes, and then only for a very short time. If you have the latest DGIndex, you can now see the frame numbers each matrix is used.

If you saw that matrix used for the whole movie by itself, then no, I haven't seen that, as I don't back up new movie DVDs all that often. I'm only aware of it being used as part of the Standard Matrix with AQ turned on ("Disable Adaptive Quant Switching" box unchecked).