CYMAT Matrix
 

In French forum already are talking about that matrix, from ammon82 (ex avalon?).
I would like to know your opinion, specially Kwag one, as designer of KVCD matrix.
CYMAT is completely different to KVCD matrix, but it seems reach more compression (I don't tested it, only read about this).

http://www.brckomania.net/CYBERNETSO.../htm/cymat.htm

Search for "avalon" in this forum.
You'll see who he was, what he did, and what he stole (from many!)
I'm sure you'll get the clear picture ;)

-kwag

Hi Pro,

The Cymat or also as known as old Avamat, is a Matrix which gives nice compression BUT it cuts a lot of frequencies, ......

http://www.digitalfaq.com/archives/error.gif


...not mention the ex 2.0 .... 8O

http://www.digitalfaq.com/archives/error.gif
which IMHO is only valuable for high resolution xvid encodings as those ones wont be scaled up that much.

The optimum of a mtrix is not just a result of more compression, its the optimal middle between compression, keeping details and less quantisation.

Like the MVCD Matrix. It compresses more even if only GOP 15 is used and also much frequencies are kept. But KVCD Notch for example is better for movies which do got dark and plain areas like underwater movies and Animatics also. Karl wanted to support a wide range of movie types. MVCD is more for typical Movies.

Also the Bach01 (my favorite at the moment) is a very good developement from Ddog which results in much compression and less Q average in Bitrateviewer finally.

So when comparing Matrixes, you also have to watch beside the final target size the Q curve in Bitrateviewer.

Also it depends on which avg Bitrate youre encoding, which encoder you use and what kind of movie u do encode ;-)

Heres a nice Link where u can learn a lot about matrixes and how they do work:
http://www.edv-tipp.de/gastbeitraege/kika001_dct.htm

I already read who is avalon, for that I remarked that cymat matrices are very differents to notch one.
I wanted an answer as Inc, give me in the other post.
Thanks.

I know that Inc. wouldn't be with MVCD if he think that it is illegal or not etic. Could somebody say me hot was resolved problem between KVCD and MVCD?

Well you posted: "from ammon82 (ex avalon?). ", which is clearly a question :!:
So I answered ;)

-kwag

As MVCD was equal to KVCD matrix, I though that more compresibility was reached for P or B (dont remember) picture spoilage setting.
1 - Is it in this way?
2 - Was it so, but now it changed?
3 - If it keep in this way, how can I use MVCD enhanced compresibility with Mencoder (related with picture spoilage).
4 - I will search it, but could you point me a Batch01 link?

I didn't know if ammon82 was really avalon.

The Problem between KVCD Notch and MVCD Matrix has been solved since Mole is no more active in that community (and also doesnt support it anymore). So "Roc" (his name is Martin) got in with a new own Matrix which is totally different from Notch. From there on the templates have been called 2.5 (incl. std. GOP).
I did enhance these MVCD Templates a bit within TmpgEnc specific Settings and by that 2.6 was born.

Karl is aware about that and I do agree that Mole forced the real heat by just copying the Notch in a disrespectable way. So Roc went in and cleared up things in there, and so for me HE is the real developer of that actual "MVCD" Format ... so you can call it Martin-VCD :wink:

It "was" and since the begginning of 2003 its a unique matrix:


MVCD 2.5/2.6

8 10 16 20 25 28 33 44
10 10 14 26 27 29 34 47
16 14 18 27 29 34 37 48
20 26 27 31 36 37 38 50
25 27 29 32 37 40 48 58
28 33 36 42 45 50 58 68
33 37 39 44 48 56 66 79
37 39 45 48 56 66 79 83

16 18 20 22 23 27 32 40
18 20 22 24 26 32 36 42
20 22 24 26 33 38 42 44
22 24 26 34 38 42 44 46
23 26 33 38 42 44 46 48
27 32 38 42 44 46 48 50
32 36 42 44 46 48 50 52
40 42 44 46 48 50 52 58

B-Picture spoilage setting isnt needed anymore.


Just use it in mencoder and do your tests. But as I said, it depends on the movie .... there will never be one optimal matrix for all movies and all bitrates ;-)
*Maybe* in Mencoder the "spoilage" is the same as the quant diff. range from I to P and B. But Im not shure how you can imitate that in a 1:1 way in avcodec like tmpgenc does.

Bach01 (not Batch01)

08 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

16 18 20 22 24 26 28 30
18 20 22 24 26 28 30 32
20 22 24 26 28 30 32 34
22 24 26 30 32 32 34 36
24 26 28 32 34 34 36 38
26 28 30 32 34 36 38 40
28 30 32 34 36 38 42 42
30 32 34 36 38 40 42 44

OT: Until now I thinked that MVCD is a copy for KVCD, as all MVCD related stuff in here, talked in that way, and I really didn't see the date of last messages (from 2003).
I think that this could happend to another people also, and this is, as when a bad news is given in a newspaper in first page, and when corrected error it is placed in last page, small letters. Don't think, it would be better that a single message was posted so things become clarified?
End OT.

@Inc.: Thank you very much for yours always so good explanations.

Inc,

what matrix do you use for your interlaced encodes?

Till now I havent found a very good low bitrate interl. matrix, as that Notch Interlaced one where I just reordered the values is not correct. Ligh told me that some month ago. So I got 2 ver well known ones from mb1 and also one from Angel:

mb1 DV 720x576 min 2000 avg 4000 max 5000

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 99 99 99
26 27 29 34 99 99 99 99
27 29 35 99 99 99 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 99 99 99
22 23 24 26 99 99 99 99
23 24 25 99 99 99 99 99


mb1 interlaced DV

8 13 13 17 17 21 21 28
13 13 17 17 19 21 23 30
13 17 19 19 21 23 28 34
13 17 19 19 21 23 28 48
17 19 19 19 23 28 34 48
19 19 23 25 28 32 34 48
19 21 23 25 28 32 34 48
21 21 25 25 28 32 34 48

8 11 11 15 15 17 17 24
11 11 15 15 17 17 21 24
13 15 15 17 21 21 26 34
13 17 15 17 21 21 26 48
17 21 21 23 21 30 34 48
17 21 21 23 28 30 34 48
19 19 25 26 28 30 48 48
19 19 25 26 28 30 48 48


Angel SVCD 4-3 SVCD Interlaced 99
(a more cutting at high freqs one of that mb1 above)

8 13 13 17 17 21 21 28
13 13 17 17 19 21 23 30
13 17 19 19 21 23 28 34
13 17 19 19 21 23 28 99
17 19 19 19 23 28 34 99
19 19 23 25 28 32 34 99
19 21 23 25 28 32 34 99
21 21 25 25 28 32 34 99

8 11 11 15 15 17 17 24
11 11 15 15 17 17 21 24
13 15 15 17 21 21 26 34
13 17 15 17 21 21 26 99
17 21 21 23 21 30 34 99
17 21 21 23 28 30 34 99
19 19 25 26 28 30 99 99
19 19 25 26 28 30 99 99

Thanks, I have only seen mb1's older interlaced DV matrix (with the max 48). I'll have to experiment with my next TV cap, although it looks that mb1's newer DV matrix cuts the frequencies quite a lot.