digitalFAQ.com Forums [Archives] - Is it true that the Bach1 is a modified KVCD matrix?

digitalFAQ.com Forums [Archives] (http://www.digitalfaq.com/archives/)

- Video Encoding and Conversion (http://www.digitalfaq.com/archives/encode/)

- - Is it true that the Bach1 is a modified KVCD matrix? (http://www.digitalfaq.com/archives/encode/13731-true-bach1-modified.html)

Is it true that the Bach1 is a modified KVCD matrix?

Hi Karl,

I read somewhere on the net, I can't remember where might have even been over at D9, to my horror that the DDog's Bach1 matrix is based on the KVCD matrix and is a modified version of it.

Is this true?

What is your point of view on this?

I suppose it is perfectly legit to modify someone else's matrix and hard work but in my opinion it is a bit cheeky.

I just wanted to know what you thought about it, I have never used the Bach1 and prob never will there is only one Matrix for me and that is the 'Notch'. ;)

I would appreciate your thoughts please Karl.

Thanks. :)

Hi Michael,

I haven't even looked closely at bach's matrix, so I can't tell.
However, at the time when the notch matrix was designed, actually the original KVCD matrix without the notch pattern, there was no bach matrix, no B matrix, no Q matrix, etc. :lol:
It just so happens that when the KVCD notch matrix was released, a lot of people went crazy trying to do matrixes.
I do recall the "bach" character at doom9, who was very knowledgeable, but one day he went bananas and deleted all of his posts, and was never seen again. There were a lot of good theoretical posts from him at doom9, and all were lost.
But coming back to the point, I still prefer the notch matrix, because *none* of the other matrixes has the notch pattern which is what kills most of DCT blocks on low lit scenes. And I'm sure bach's matrix doesn't have the notch. If it has it, then IT IS the KVCD notch matrix ;)

-kwag

Hi Karl,

Thank you very much for clearing this up, it was a very interesting read.

To be honest I have never used the Bach1 matrix it just something I happen to come across by chance while web browsing, I can't even remember where I read it from that it was a modified KVCD matrix. I wish I kept it to show you. :oops:

Yes I heard a lot about this Bach guy apparently like you said he was very knowledgeable about a lot of things.

I was very interested to find out he offered a lot of good info and it was one of his theoretical posts that help lead to the fine tweaking Robshot method in DVD2SVCD after which Tylo improved apon apparently.

It is a shame he went bananas and deleted all his posts all that knowledge lost is in my eyes a real crime and a shame.

To be honest although I have never looked at this Bach1 matrix if it did have the 'Notch' pattern in it, then to me this is a blatent rip-off of your matrix and to me this is bad.

Still for me at least there is only one matrix and it is the KVCD 'Notch' matrix, it has never let me down and I know it never will. :D

Thanks again Karl for info. :)

Actually, the bach1 non-intra IS the KVCD non-intra.
The bach1 intra is a slightly modification of QLB (or the opposite) :

Code:

*MATRIX NOTCH

 8  9 12 22 26 27 29 34   16 18 20 22 24 26 28 30

 9 10 14 26 27 29 34 37   18 20 22 24 26 28 30 32

12 14 18 27 29 34 37 38   20 22 24 26 28 30 32 34

22 26 27 31 36 37 38 40   22 24 26 30 32 32 34 36

26 27 29 36 39 38 40 48   24 26 28 32 34 34 36 38

27 29 34 37 38 40 48 58   26 28 30 32 34 36 38 40

29 34 37 38 40 48 58 69   28 30 32 34 36 38 42 42

34 37 38 40 48 58 69 79   30 32 34 36 38 40 42 44



*MATRIX QLB

 8 16 19 22 26 27 29 34   16 17 18 19 20 21 22 23

16 16 22 24 27 29 34 37   17 18 19 20 21 22 23 25

19 22 26 27 29 34 37 39   18 19 20 21 22 23 24 26

22 22 26 27 29 34 38 42   19 20 21 22 23 24 26 28

22 26 27 29 32 36 40 50   20 21 22 23 25 26 28 29

26 27 29 32 36 40 50 61   21 22 23 24 26 28 29 31

26 27 29 35 40 50 59 75   22 23 24 26 28 29 31 34

27 29 35 40 50 59 75 89   23 24 25 28 29 31 34 38



*MATRIX BACH1

 8 16 19 22 26 27 29 34   16 18 20 22 24 26 28 30

16 16 22 24 27 29 34 37   18 20 22 24 26 28 30 32

19 22 26 27 29 34 34 38   20 22 24 26 28 30 32 34

22 22 26 27 29 34 37 40   22 24 26 30 32 32 34 36

22 26 27 29 32 35 40 48   24 26 28 32 34 34 36 38

26 27 29 32 35 40 48 58   26 28 30 32 34 36 38 40

26 27 29 34 38 46 56 69   28 30 32 34 36 38 42 42

27 29 35 38 46 56 69 83   30 32 34 36 38 40 42 44

Thanks for the added info Phil. :)

So what are the implecations then, is it a part rip-off of the Karl's matrix then? This seems very naughty to me.

So its a mix of the two, more likely imho QLB took some of the Bach1 intra as to my knowledge the bach1 is older.

Thanks again Phil for this added info it was very interesting.

In a sense, it feels good to know that the rip-offs are based on the KVCD matrix :!:
That only means one thing: :arrow: "KVCD Notch set a standard on matrixes" :P
Thanks for the info guys :cool:

-kwag

Quote:

Originally Posted by kwag

That only means one thing: :arrow: "KVCD Notch set a standard on matrixes" :P

Yes that is good to know that the KVCD 'notch' matrix is setting the standards. :D

Quote:

Originally Posted by kwag

Thanks for the info guys :cool:

-kwag

Your welcome. :)