Avisynth: DVD to CVD Conversions?
 

Hi, I have been farting around with so many filters that I think I just S**T myself. I do DVD to CVD Conversions and would like to know what you all think of my script conclusion. Thanks

[AVISYNTH_Simple Resize_Convolution3D_Blockbuster]
0=loadplugin(!simpleresize.dll)
1=loadplugin(!convolution3d.dll)
2=LoadPlugin(!Blockbuster.dll)
3=simpleresize(^TargetWidth,^TargetHeight)
4=BlockBuster(method="noise", variance=1, seed=1)
5=Blockbuster(method="sharpen", detail_min=^detail_min,detail_max=^detail_max, strength=^StrengthValue)
6=convolution3d(^A,^B,^C,^D,^E,^F,^G)
!simpleresize.dll=c:\dvd2svcd\simpleresize\simpler esize.dll
!convolution3d.dll=C:\DVD2SVCd\convolution3d\Convo lution3D.dll
!Blockbuster.dll=C:\DVD2SVCd\blockbuster\Blockbust er.dll
^A=0
^B=3
^C=4
^D=3
^E=4
^F=2.8
^G=0
^detail_min=10
^detail_max=99
^StrengthValue=10

Re: Need a Yay or Nay. Please :)
 

That's never good ;).

Personally I wouldn't use Blockbuster's sharpen method. As for resizing, I like either bilinear or a neutral bicubic. As for smoothing, I like FluxSmooth :mrgreen:.

Thanks for responding. I also like Bilinear and Unfilter. What do you use to sharpen?

Would this maybe better;

Bilinear
Blockbuster ... -> noise only for getting rid of DCT's
Unfilter (20,20) -> for sharpen
Convolution3d -> for final, all around cleanup

I have always wondered about the noise thing, as to where to put it. If it's put before resize you get smaller file sizes and if I use Unfilter do think that it does a good enough job with the DCT blocks?