Avisynth: Converting Episodes to SKVCD?

LadyMiles · #1 07-15-2004, 08:47 PM

Okay I just upgraded to avisysnth 2.5 and now my video looks weird, like the colors are smeared. I don't know what's is wrong. The video source is fine. My colors look weird.

Here is my script
(I needed to deinterlace the source right ?)

LoadPlugin("C:\Filters\MPEG2Dec3.dll")
LoadPlugin("C:\Filters\GripFit_YV12.dll")
LoadPlugin("C:\Filters\STMedianFilter.dll")
LoadPlugin("C:\Filters\asharp.dll")
LoadPlugin("C:\Filters\unfilter.dll")
LoadPlugin("C:\Filters\undot.dll")
LoadPlugin("C:\Filters\Convolution3DYV12.dll")

Mpeg2Source("C:\CURB_ENTHUSIASM\VIDEO_TS\CEone.d2v ")

ConvertToYV12

SeparateFields()
odd=SelectOdd().Kwag_MA()
even=SelectEven().Kwag_MA()
Interleave(odd,even)
Weave()

Function Kwag_MA(clip input) {
undot(input)
Limiter()
asharp(1, 4)
STMedianFilter(3, 3, 1, 1 )
MergeChroma(blur(1.5))
MergeLuma(blur(0.1))
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) ")
Convolution3D(preset="movieLQ")
Limiter()}

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

Dialhot · #2 07-16-2004, 03:45 AM

Question : Which version did you have before ? Which script ? All was perfect on the same source ?

Note: use ALWAYS ConverttoYV12()
It seems that there is a strange issue in avisynth that drop the speed in case of missing '()' (not sure, but I read that in some places).

Boulder · #3 07-16-2004, 04:37 AM

I'd say that script looks extremely weird, a huge amount of filtering and not an optimal way to treat an interlaced source. What kind of a source do you have, is it NTSC, interlaced 29.97fps?

LadyMiles · #4 07-16-2004, 02:40 PM

Interlaced

But I just reverted back to my old apps. Upgrading was a pain. There should be sticky about upgrading from avisynth 2.0 to the current.
I got all types of weird runtime errors and unexpected characters.

Or a sticky with all the new encoding programs (i.e. diko, mencode) listing what you will need for each.

Boulder · #5 07-16-2004, 03:00 PM

There should be nothing hard about upgrading. Use the uninstaller in Avisynth v2.0, remove any instances of avisynth.dll from your computer and delete all the plugins from your plugins folder. Then install the Avisynth 2.5 and unpack all the v2.5 compatible plugins that you need in your plugins folder.

It is important that you use MPEG2DEC3.dll!

Find all the plugins here: http://www.avisynth.org/warpenterprises

Here's a modified version of your script for AVS v2.5:

Code:

Mpeg2Source("C:\CURB_ENTHUSIASM\VIDEO_TS\CEone.d2v") 
AssumeTFF() # Put AssumeBFF() if your video is bottom field first!
KernelBob(7)
#CROP HERE
undot(input) 
asharp(1, 4) 
#RESIZE HERE
STMedianFilter(3, 3, 1, 1 ) 
MergeChroma(blur(1.5)) 
MergeLuma(blur(0.1))
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, 2) ") 
Convolution3D(preset="movieLQ") 
Limiter()
SeparateFields()
SelectEvery(4,1,2) # SelectEvery(4,0,3) for bottom field first video
Weave()
#ADD BORDERS HERE

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

function kernelbob(clip a, int th) 
{   ord = getparity(a) ? 1 : 0 
   f=a.kerneldeint(order=ord, sharp=true, twoway=true, threshold=th) 
   e=a.separatefields().trim(1,0).weave().kerneldeint(order=1-ord, sharp=true, twoway=true, threshold=th) 
   interleave(f,e).assumeframebased() 
}

Remember to encode as interlaced if you use this script. I'd also ditch Convolution3d, you're already doing quite heavy filtering.

The plugins you'll need for this script are UnDot, aSharp, STMedianFilter, UnFilter, Convolution3D and KernelDeint.