Log in

View Full Version : CCEAQM option in RB


Pages : [1] 2

Mr. Monte
12th November 2005, 20:11
O.k..been looking to do some comparisons with different codecs. I did a search here on CCEAQM, but only came up with one thread (Sapster's). He suggest turning it on when using his codec, but didn;t jdobbs recommend it off (by default in RB) due to some possible compatibility issues?

Could someone explain how much better there codings have been with or wihtout this option (when using CCE) and how ofter you may have a compatibility issue. And what is that issue?? (Stuttering, non-playing)

TIA

Fishman0919
13th November 2005, 00:39
that is an option to turn on a matrix correction feat. with CCE... SAPSTAR uses that option to turn on feat. with AQE. CCE can look better with that feat turned on.

Mr. Monte
13th November 2005, 01:31
CCE can look better with that feat turned on.

Thanks fish...but what about the compatibility issues jdobbs talked about?

Have you ever had any?

Fishman0919
13th November 2005, 04:48
From the CCE SP Guide...

Cinema Craft Encoder SP (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.


I worked for a Audio/Video store for about 5 year and when CCE SP 2.70 came out with this feat. I used it and I did a lot of testing with it in the store on a number of DVD Players. I only had 1 DVD Player that didn't like it. A first Gen Pioneer DV-09 LINK to Player (http://www.thedigitalbits.com/reviews/dv09.html)

manono
13th November 2005, 06:07
I second that. I've been making quite a few DVDRs recently with Adaptive Quantization turned on in CCE 2.70, and also think it can help the emcode a lot, especially when compared to using CCE's Standard Matrix alone. I have yet to see a DVD player that has any problems, but haven't done any extensive testing, as Fishman0919 has. In any event, to be spec compliant, DVD players are supposed to be able to decode DVD with AQ flawlessly. If it can't, it's a piece of junk. Just as if it can't handle DVD spec max bitrates, it's also a piece of junk.

In addition, more and more Hollywood DVDs are being released also using the exact same matrices as CCE's Standard Matrix with AQ turned on.

Mr. Monte
13th November 2005, 17:32
Thanks guys..I will start turning it on.

I would imagine it will add a small amount of time to the encode.

Carpo
13th November 2005, 18:54
so just to clarify - your both saying that it should be ok to have it turned on ?

i mainly do backups that i will then play on either my xbox here (main pc) or my laptop - and on occasion the dvd player downstairs which was a cheap ronin £50 job but has played near every disc i have done so i think i should be ok with it.

Just wondering what jdobbs take on this is as he has said not to use use it - but could poss be to cut down on false bug reports.

im gonna try it on a encoder later :)

manono
13th November 2005, 22:42
Hi-

I can't speak for Fishman0919, but in my opinion it's perfectly OK to use the Standard Matrix with AQ turned on. I'll even go further and say that it's an easy way to greatly improve the quality of your encodes without doing the testing to find the optimum matrix to use for each encode. Having said that though, I still prefer myself to use a single matrix when possible, or AQ on a different matrix, for my own backups. You might make a test backup to see if it plays well on all the players that you might be likely to use. I'm betting it will. I think that any problems that might arise will be confined to old and by now obsolete players.

Just wondering what jdobbs take on this is...

Nor can I speak for him (but I'll try anyway :)). He wants DVD-RB to be compatible with as wide an audience and for as many players as possible, and I guess there are people out there with non-spec players.

You want an example of a major DVD release that uses the Standard Matrix with AQ. I was running some matrix tests recently and happened to choose my R1 DVD of Lord Of The Rings Return Of The King Extended Edition with which to test. I was surprised to find it used it. I don't back up that many new movie DVDs, usually preferring the classics, so I checked with a friend,also interested in matrices, and he confirmed that this practice has become very common with the studios of late. Do you really think that New Line and others would release a DVD that might cost them money if very many people couldn't play it well and wanted their money back? I might add that the other 2 earlier R1 LOTR EE DVDs use the matrix commonly called the Fox Home Networks Matrix around here, or number 1 in the Rockas Matrix Editor.

Carpo
13th November 2005, 22:57
1800-3500 AVAMAT7.mtx and cceaqm - using that on an encode - i tend to use that matrix for most and rb defaults - with the backups i have done without aqm on they look good enuff and play ok

Mr. Monte
14th November 2005, 00:23
manono,

How can you tell what matrix or that CCE with AQM was used on a studio DVD?

Is there a proggy to look at it and determine this?

Also, I encoded a DVD with AQM..however, nothing popped up on the CCE encode screen or in the log in RB showing I used it. IS there a way for me to verify it was used?

SAPSTAR
14th November 2005, 00:52
jdobbs just gave a standard warning about the possible incompatibility, but every encode I made with this options always worked on different DVD players, anyway most DVD players are now using the same MPEG2 chip....so no pb.
Something else, if you are talking about AQE (my encoder), the option is used to trigger a special mode which is NOT the same than CCE and it's 100% compatible with any MPEG2 chip...it's fully MPEG compliant.

Mr. Monte
14th November 2005, 01:34
Thanks Sapster...was wondering if you can comment on any of the other questions I asked above?

Thanks Again.

manono
14th November 2005, 04:18
Hi-

How can you tell what matrix or that CCE with AQM was used on a studio DVD?

There are several ways to tell. You can open a vob, m2v or mpv in ReStream and it will tell you the matrix for the first GOP only. You can run them through the Rockas Matrix Editor, and it will give you up to 2 matrices (I think) used by the entire video. You can open them in DGIndex and go Options->Log Quant Matrices and save the Project File. If you don't want to do the whole movie or an entire vob, you can do part of it by using the [ and ] buttons. When done, there will be an xxx.quants.txt. You open that and you'll see all the Intra and Non Intra matrices used. I've seen a maximum of 3 used in a single movie.

If you're using a version of CCE that allows the use of different matrices and you have the "Disable adaptive Q-matrix switching" box unchecked, and you choose the Standard Matrix, then you'll get 3 matrices used (usually, depending on the "overhead" allowed by the movie you're encoding. These matrices will be the same as the commercial DVD releases that I've seen that use the same method. Maybe they use a CCE hardware encoder. I don't really know, as I don't know much about that side of it. These 3 matrices are the Standard Matrix:

Intra Luma and Chroma Matrix:
8 16 19 22 26 27 29 34
16 16 22 24 27 29 34 37
19 22 26 27 29 34 34 38
22 22 26 27 29 34 37 40
22 26 27 29 32 35 40 48
26 27 29 32 35 40 48 58
26 27 29 34 38 46 56 69
27 29 35 38 46 56 69 83

NonIntra Luma and Chroma Matrix:
16 17 18 19 20 21 22 23
17 18 19 20 21 22 23 24
18 19 20 21 22 23 24 25
19 20 21 22 23 24 26 27
20 21 22 23 25 26 27 28
21 22 23 24 26 27 28 30
22 23 24 26 27 28 30 31
23 24 25 27 28 30 31 33

The Standard Matrix with the numbers halved:

Intra Luma and Chroma Matrix:
8 8 10 11 13 14 15 17
8 8 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

NonIntra Luma and Chroma Matrix:
8 9 9 10 10 11 11 12
9 9 10 10 11 11 12 12
9 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

And the Standard Matrix with them halved again:

Intra Luma and Chroma Matrix:
8 8 8 8 8 8 8 9
8 8 8 8 8 8 9 9
8 8 8 8 8 9 9 10
8 8 8 8 8 9 9 10
8 8 8 8 8 9 10 12
8 8 8 8 9 10 12 15
8 8 8 9 10 12 14 17
8 8 9 10 12 14 17 21

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

That's what DGIndex gives you. You'll see them switching back and forth up to hundreds of times during a movie.

manono
14th November 2005, 04:53
Hi-

...anyway most DVD players are now using the same MPEG2 chip...

Hardly. There must be dozens of proprietary and 3rd party MPEG-2 decoder chips in common use. I don't even know if there's a dominant one, so I don't know which one you might mean. That isn't to say that they aren't all supposed to be able to handle Adaptive Quantization. They are. If you're curious about some of the chipsets in common use, then you can check out which ones are used by well known NTSC DVD players here:

http://www.hometheaterhifi.com/cgi-bin/shootout.cgi?function=search&articles=all

Just click on the various players to see the chipset(s) used (if known).

Beginning about half way down this page is a discussion of some of the more common and well known chipsets:

http://www.hometheaterhifi.com/volume_7_4/dvd-benchmark-part-5-progressive-10-2000.html

m.rup
14th November 2005, 13:53
Hi, I have a probably dumb question: when I use DVDRB Pro, I can choose different matrices for different bitrates (low, very low and so on). Does this mean that CCEAQM option is on by default or have I to set it manually? These matrices aren't concerned with those from Rockas Matrix Editor?

SAPSTAR
14th November 2005, 14:31
Hi, I have a probably dumb question: when I use DVDRB Pro, I can choose different matrices for different bitrates (low, very low and so on). Does this mean that CCEAQM option is on by default or have I to set it manually? These matrices aren't concerned with those from Rockas Matrix Editor?
CCEAQM is off by default, it's not related to the matrices you chose. It's a special mode in CCE and AQE.

@manono : I never said that the QMatOp function was the Adaptive Quantization from CCE SP ?!? QMatOp is an algo which tries to generate the best matrix for an encode....so it's fully supported by any chip, because the matrix is the same for the whole movie....but optimized prior to start ! In the case of DVDRB, for each segment (AVS script), an optimized matrix is generated.
Once again, it's NOT the same thing than CCE SP.

m.rup
14th November 2005, 14:45
Thanks for reply. So is it right that DVDRB chooses appropriate matrix (loy, very low) without having CCEAQM activated? And if I had CCEAQM activated, DVDRB would manipulate this choosen matrix furthermore?

SAPSTAR
14th November 2005, 14:52
... And if I had CCEAQM activated, DVDRB would manipulate this choosen matrix furthermore?
Wrong....DVDRB does nothing here...it's the encoder itself !!! See DVDRB as a smart robot, but never forget that what really produces the movie is still the encoder......

m.rup
14th November 2005, 15:01
OK, then let's say: If I had CCEAQM activated CCE would manipulate the choosen (by DVDRB?) matrix furthermore. Is this right? Sorry, I'm not very experienced in this area.

SAPSTAR
14th November 2005, 15:30
OK, then let's say: If I had CCEAQM activated CCE would manipulate the choosen (by DVDRB?) matrix furthermore. Is this right? Sorry, I'm not very experienced in this area.
Yep...

m.rup
14th November 2005, 15:51
Thanks, that makes it clear.

@manono:


I can't speak for Fishman0919, but in my opinion it's perfectly OK to use the Standard Matrix with AQ turned on. I'll even go further and say that it's an easy way to greatly improve the quality of your encodes without doing the testing to find the optimum matrix to use for each encode.

Does that mean i mustn't select different matrices for different bitrates in DVDRB or is it ok to select different matrices and set CCEAQM to active?

jdobbs
14th November 2005, 16:58
Removed.

manono
14th November 2005, 17:24
Hi-

Sorry, but I don't know the answer to your question. You can no longer use AQ directly in DVD-RB, and I've never used CCEAQM. I do these things when encoding manually with CCE.

SAPSTAR-

I never said that the QMatOp function was the Adaptive Quantization from CCE SP ?!?

Neither did I. At least I don't think I did. As stated above, I've never used your (very fine, I'm sure) app. If you're referring to my very long post earlier, I was trying to answer the question about (I thought) how to tell what matrices were used in a DVD-RB encode and in Studio DVDs, and if that Studio DVD used the Standard Matrix with AQ turned on. Now I think I see to what you're referring:

...or that CCE with AQM was used on a studio DVD?...

I didn't realize that AQM might be referring specifically to your app (and I'm still not convinced it is). Earlier I had stated that the Standard Matrix with AQ was showing up in commercial DVDs. Nowhere did I refer to your program. So I was assuming that Mr. Monte was asking about Standard Matrix with AQ. Obviously no studio is using any uncommon (even if spec-compliant) matrix created by your program. If I was somehow unclear, I apologize. If I misunderstood Mr. Monte's question, I apologize.

DK
14th November 2005, 17:56
fyi:

i wanted to try autoqmatenc with rb yesterday

realizing that the encoder stated that CCEAQM was not active i added CCEAQM=1 to the rebuilder.ini

result: still INactive

i changed the postition of this entry from the first lines down to cce-options as i wasnt really sure

result: INactive


finally i realized that the spelling of rebuilder.ini was REBUILDER.INI

changing the capital letters of the extension to REBUILDER.ini made a change finally

Carpo
14th November 2005, 18:06
i used it in an encode last night with cce looks ok havent had time to test it on the player down stairs but seeing i play most things on here its no biggie :)

Mr. Monte
17th November 2005, 14:49
I would imagine..it should look or test better than with it off. Has anyone ever did a PSNR and SSIM test on the same material (with AQM off vs on)?

Mr. Monte
18th November 2005, 00:15
Also....does it matter where this option is placed in the RB INI file?

I assumed it belongs under the [CCEOPTION] header..but since there is nothingconfirming it is enabled..I just wanted to make sure

TIA

jdobbs
18th November 2005, 04:31
It goes in the "[Options]" area.

CCEAQM=1

Mr. Monte
18th November 2005, 14:36
Thanks Jerry. I found a post last night that showed a log with it enabled. Apparrently DVDRB does indicate in the log when it is enabled. Could you add where it gets placed in your help file on the next update?

Thanks for a great program again :)

fluis
19th November 2005, 02:34
jdobbs, why did you deleted your 14/11 post ? Monono answer is out of context.

Best regards

jdobbs
19th November 2005, 05:03
???

He is answering m.rup -- how is it out of context? The post I deleted was only there for 2 minutes and manano answered almost a half hour later. I don't even remember what I said -- I just remember it didn't properly fit.

thewonderer
19th November 2005, 13:58
does the freeware version of dvd-rb work with this cceaqm=1 switch? I don't see a message saying it was detected...

jdobbs
19th November 2005, 14:39
No. It's always off in the freeware version

fluis
27th November 2005, 05:41
Sorry, my brain stoped. Regards

happycase
27th November 2005, 08:10
Ok, just so I have this right, AQM is something that scans a scene and applies the best matrix to the scene and all I have to do, if I have CCE 2.67 or higher to enable this function is put CCEAQM=1 in the options section of the .INI?

Does this program automatically overwrite any matrix that I have checked other than 'encoder default?' Is there anything else I should know about this option before I can start using it properly?

jdobbs
27th November 2005, 14:11
@happycase

It overrides any matrix you have set.

@everyone

Also please note the warning earlier in this thread and in the CCE documentation... CCEAQM may create in incompatible stream. That's why it is only available as a "hidden" option. When you buy a new player for Christmas and your backup doesn't play right -- please don't blame DVD-RB. Deciding to use AQM was your choice.

DK
27th November 2005, 14:34
And you have to set AutoQMatEnc.exe instead of CCE in setup as your encoder.


if you dont want CCE to make use of that function this is true

on the other hand this thread pretty much looks like it dealt with the aqm function that cce is able to, doesnt it?


++++++++++++++

i should think rippraff has noticed his mistake already as the posting is gone

Mr. Monte
27th November 2005, 17:00
@happycase

It overrides any matrix you have set.

jdobbs,

Thanks for that tidbit..I was unaware it overrided any matrix we selected. I assumed I guess that it would "modify" the matrix or more allocate the bits. Do you know whether it only uses it's built in available matrixes then? Would this technically be better than running a project through bitrate viewer and figuring out the compression then the resulting bitrate, then select a custom matrix ?

Thanks for all the feedback in this thread.

Rippraff
27th November 2005, 17:11
i should think rippraff has noticed his mistake already as the posting is gone
Yes, I realized that it was a misunderstanding so I've deleted my post, before there was any answer. ;)

Cu Rippraff

thewonderer
30th November 2005, 13:16
Another question about cceaqm option. If you are doing 1 pass, Quality encoding, and have enabled cceaqm in the ini file, would cceaqm still work, is cceaqm only for VBR passes?

What about using OPV with rb-opt 0.21, standard cce matrix, with the tickbox empty for disable Adapative Q Matrix switching...? would cceaqm still be honoured in this..?

Thanx!

Insane2986
8th March 2006, 09:26
SAPSTAR or others,

How does CCEAQM=1 in CCE compare to AutoQMatEnc in deciding which matrix to use?

Fishman0919
8th March 2006, 19:49
SAPSTAR or others,

How does CCEAQM=1 in CCE compare to AutoQMatEnc in deciding which matrix to use?

CCEAQM=1 is just a switch in the rebuilder.ini file to set adjust_q_matrix=1 in the REBUILDER.ECL. When CCE SP 2.70.##.## see this it turns on adaptive Q-matrix switching.

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.
Thus, adaptive Q-matrix switching can improve the quality of image,
especially with lower bitrate sources or CG works or the edge of
fading out.

In AQE it turn on the Built-In advanced QMatOp technology... in short AQE looks at the segment of movie to be encoded with dctune2.0.exe...DCTune is a patented technology available by NASA (Link (http://vision.arc.nasa.gov/dctune/))... to find the "best" matrix for that part and then uses it to encode the segment.

archaeo
8th March 2006, 21:48
@Insane2986

I have been using AutoQMatEnc on and off for awhile now, and had been recently testing the latest beta for SAPSTAR, v .33.0.1. I find the program holds a lot of promise, and have had some very good results with it, at times. I have also found that DCtune is not always better than I am at selecting matrices. I find the matrix selection to be pretty aggressive most of the time, really filtering out a lot of detail when it appears that lower thresholds would suffice to get the compression I need. Perhaps this has something to do with the fact that DCTune was primarily designed to be used for JPEG image compression, a slightly different animal that MPEG.

SAPSTAR has added a nice function to the latest AQE version, which is a hidden setting that can allow me to limit the thresholds on the intra and inter matrices generated by DCtune. So, if I see that AQE is creating some heavy compression matrices (multiple values of 255 for instance), I can set the parameters to hold the maximum at 99, or 127, or whatever. The same for the minimum thresholds. I've tested this a few times now on some challenging sources, with good results. This feature has given me a little more control over the creation of the DCTune generated matrices than I had before. It's a good addition to a good program.

omega6666
22nd December 2006, 14:31
jdobbs,

Thanks for that tidbit..I was unaware it overrided any matrix we selected. I assumed I guess that it would "modify" the matrix or more allocate the bits. Do you know whether it only uses it's built in available matrixes then? Would this technically be better than running a project through bitrate viewer and figuring out the compression then the resulting bitrate, then select a custom matrix ?

Thanks for all the feedback in this thread.

I'm also very curious about the answer for this question. I was hoping that AQ would adapt the custom chosen matrix for the main movie where needed. But if it chooses the best possible matrix from my list of custom matrices intelligently, I guess that would be cool as well. But if AQ is only possible with auto-adapting the few standard matrices, then I don't like the feature too much...

Voodoochild
22nd December 2006, 21:25
cut the matrix you selected by half or quarter, depend on what you chose .. for example this is avamat 6 matrix

8 16 19 22 26 27 29 34
16 16 22 24 27 29 34 35
19 22 26 27 29 34 35 38
22 22 26 27 29 34 35 40
22 26 27 29 32 35 40 48
26 27 29 32 35 40 48 50
26 27 29 35 40 48 50 60
27 29 35 40 48 50 60 62

16 20 24 28 32 36 40 44
20 24 28 32 36 40 44 48
24 28 32 36 40 44 48 52
28 32 36 40 44 48 52 56
32 36 40 44 48 52 56 58
36 40 44 48 52 56 58 60
40 44 48 52 56 58 60 62
44 48 52 56 58 60 62 62

this is after encoding with CCEAQM=1

NonIntra Luma and Chroma Matrix at encoded frame 36694:
8 10 12 14 16 18 20 22
10 12 14 16 18 20 22 24
12 14 16 18 20 22 24 26
14 16 18 20 22 24 26 28
16 18 20 22 24 26 28 29
18 20 22 24 26 28 29 30
20 22 24 26 28 29 30 31
22 24 26 28 29 30 31 31

Intra Luma and Chroma Matrix at encoded frame 36802:
8 8 8 8 8 8 8 9
8 8 8 8 8 8 9 9
8 8 8 8 8 9 9 10
8 8 8 8 8 9 9 10
8 8 8 8 8 9 10 12
8 8 8 8 9 10 12 13
8 8 8 9 10 12 13 15
8 8 9 10 12 13 15 16

hope I helped

Elad

Fishman0919
23rd December 2006, 00:22
kinda quoting CCE SP guide...

For whatever matrix you use, if CCE SP cannot keep the minimum bitrate with the given matrix for that GOP or...that the lowest quantization scale cannot raise the bitrate under a single quantization matrix, CCE SP cuts the value of each element in the matrices by half, quarter and so on till the values in the given matrix used fix the need of the GOP being encodes.


08 16 19 22 26 27 29 34

may become...

08 08 10 11 13 14 15 17

or...

08 08 08 08 08 08 08 08

If the matrix you are using is for a low bitrate encoding and the bitrate spikes real high, CCE SP well adjust the matrix per GOP as needed to give a better picture.

jdobbs
23rd December 2006, 01:14
I probably should have clarified what I meant by "it overrides the matrix" -- that means what you will see in the stream may not be the matrix you chose, because the setting of CCEAQM could very well change it. I thought Fishman0919's quote from the guide along with Voodoochild's comments have explained it better much more clearly than I could.

DVD-RB will still apply the matrix you've chosen to the ECL file when you select it regardless of the setting of CCEAQM. If you have CCEAQM selected -- it will also be applied and could possibly modify the matrix.

Of course, if you have the minimum bitrate set to the DVD-RB default of 300Kbs -- I wouldn't think it would get applied very often except in black, fade to black, or very low detail scenes.

In AutoQMatEnc it is an entirely different story. It is set using a different flag in DVD-RB Pro ("AutoQMatAQM=1") -- that kicks in an entirely different method of building adaptive matrices.

linx05
23rd December 2006, 05:03
I thought in recent versions of DVD-Rebuilder you did not need to but in the AutoQMatAQM=1 into the .ini? Doesn't Settings > AutoQmatEnc Settings > Enable QMATOpt (AQM) do the same thing?

techmule
23rd December 2006, 07:38
I thought in recent versions of DVD-Rebuilder you did not need to but in the AutoQMatAQM=1 into the .ini? Doesn't Settings > AutoQmatEnc Settings > Enable QMATOpt (AQM) do the same thing?

Yes, it does that only through the GUI, but as JDobbs commented AQMenc Qmatop enabling works in a totally different way, analyzing the video segment every time to come up with a custom matrix for it intelligently. It takes more time to encode overall but the quality you get is unparalleled specially under low bitrates.

Sharc
23rd December 2006, 09:28
Of course, if you have the minimum bitrate set to the DVD-RB default of 300Kbs -- I wouldn't think it would get applied very often except in black, fade to black, or very low detail scenes.


Hmmm..... Is that so? See my attachement (for CCEAQM=1).