KVCD GOP Structure?

[vdk_au]

I've never converted a cartoon from dvd to cartoon, just movies. And i was reading about something about changing the GOP Structure for cartoons, what does GOP Structure do? And i also read from i think the kvcd admin, that for new cartoons, computer animated, this isn't neccessary. Would i have to change it for converting finding nemo to vcd using the kvcd pal 704-x576 templates and if so to what?
Thanks

[Dialhot]

Don't you think Nemo enters perfectly in the definition of " computer animated" cartoons ?

BTW, if that is your first KVCD, do not do Nemo ! It gave headaches to the most experienced guys here (myself included)

The underwater scenes, even computer generated, are always very hard to convert decently.

[vdk_au]

I've actually a fair bit of dvds into kvcd but this is my first one for cartoons. It finally finished encoding, it was quiet good quality as I used the prediction programs and i got CQ Values around 80. But in some scenes its a bit blocky, especially in the water scenes, but what do you expect when the bitrate is between 300 and 2500. Is there any other way to lessen this (and not through resolution or bitrate), I also use the latest optimal scripts from kvcd.

[Critter]

Quote:

Originally Posted by Dialhot

BTW, if that is your first KVCD, do not do Nemo ! It gave headaches to the most experienced guys here (myself included)

Is there some modified avisynthscript for nemo ? Or shall I use that kwags optimal script ?

[Dialhot]

Answer for both of you :
Just use the MA script and add this line at the end :

Blockbuster(method="noise",detail_min=1,detail_max=10,variance=0.3,seed=5623)

That should reduce a lot the blocks in underwater scenes.

[vdk_au]

I'm not sure what u mean by the MA Script, can you please explain?
I used the optimal script from KWAG.

[Dialhot]

Quote:

Originally Posted by vdk_au

I'm not sure what u mean by the MA Script, can you please explain?
I used the optimal script from KWAG.

This is the MA script .

MA stands for "Motion Adaptative" and means that the script adapts its behaviour to the complexity of the scene.
This is done in this part of the script :

ScriptClip(" nf = YDifferenceToNext()" +chr(13)+ "unfilter( -(fmin(round(nf)*2, 100)), -(fmin(round(nf)*2, 100)) ).TemporalSoften( fmin( round(2/nf), 6), round(1/nf) , round(3/nf) , 1, 1)  ")

YDifferenceToNext() is used to find how the next frame will be different from the current one. And according to that, the weight of the parameters of unfilter and temporalsoften are adjusted.

[Prodater64]

Quote:

Originally Posted by Dialhot

Answer for both of you :
Just use the MA script and add this line at the end :

Blockbuster(method="noise",detail_min=1,detail_max=10,variance=0.3,seed=5623)

That should reduce a lot the blocks in underwater scenes.

@Dialhot:

Do you mean in this way:

Quote:

Originally Posted by "MA Script for Finding Nemo"............................................. .............................

Mpeg2Source("E:\DockingGate\Samples\Temp\Nemo.d2v" )

undot()
asharp(1, 4)
GripCrop(480, 576 , overscan=2, source_anamorphic=false)
GripSize(resizer="BicubicResize")
STMedianFilter(3, 3, 1, 1 )
MergeChroma(blur(1.5))
MergeLuma(blur(0.1))

## Linear Motion Adaptive Filtering ##

ScriptClip(" nf = YDifferenceToNext()" +chr(13)+ "unfilter( -(fmin(round(nf)*2, 100)), -(fmin(round(nf)*2, 100)) ).TemporalSoften( fmin(

round(2/nf), 6), round(1/nf) , round(3/nf) , 1, 1) ")

Blockbuster(method="noise",detail_min=1,detail_max =10,variance=0.3,seed=5623)

GripBorders()
#LetterBox( Your_Values_Here ) # Depends on situation. Use MovieStacker!

function fmin( int f1, int f2) {
return ( f1<f2 ) ? f1 : f2
}

or exactly at end of the script?

[Dialhot]

Do it just like you write it. Else there will be noise on the borders and that is not very clean.