dv to cvd CCE setting

[paul8883]

Hi,
I have just done some conversion from my home videos (mini dv) to CVD following the guids in vcdhelp with avi2svcd or cce together with avisynth. I found the following setting give me the good results even I don't understand some setting:

- one pass vbr with
- Q : 7
- min: 300
- max: 2520
- quality prec: 18
- anti noise: 4
- matrix: mpeg standard
- DCT: 8
- resolution: 352x576

the avs:
AVISource("C:\MSP6 Preview Files\michael1.avi")
ConvertToYUY2()
BicubicResize(320,538)
AddBorders(16,19,16,19)

Here are the questions hoping someone can help me understanding:
- I don't know if it's necessary to add TV overscan borders in CVD resolution (352x576)?
- if I increase DCT from 8 to 10, I can't see any difference in TV, but with bitrateviwer, the Q scale (both average and peak) is smaller (does it means better quality)? which DCT is better for DV source?
- what matrix is the best for dv to cvd conversion with max bitrate less than 3000 kbps?

[ulfschack]

You could use a test video where you've placed multi colored borders upon borders to determine the range of your overscan. There's a easy way to do this in VDub , allthoguh not in avisynth AFAIK.

DCT of 10 is best for DV ... yes, smaller Q means better quality.

search for posts by "mb1" who's made a long list of matrices for TmpgEnc (prolly works for CCE as well), where one is especially designed for DV. Haven't tried it yet but am going to.

cheers