hello, im running a script thats quite slow, how can i split it the video by 2 or 3 parts and then merge them back? can this be done with many parts of a lossless avi? the source is mpg2

thanks

avisynth is a frameserver, that means it is aimed to serve the picture frame per frame to an other programm (the encoder).
Why do you want to use it as something else (a cut/merge tool) ?

Explain a little more, and may be give your current script.

The source are 3 dvds very very bad interlaced, i could only get great fix with animeIVTC, the problem is with all set takes 1.3 fps ending in like 1day per movie. The source also need cleanning so thats why saving the corrected ivtc to lossless. I though i could maybe serve half and then other half and then mux them together with a script, or even 3 parts and do it in 2 pc's

These are very important dvd to me so thats why im goin in all this trouble

Are you saying is better to cut the vobs, then ecode each other and then mux them when they are completely done (ivtc, cleaning)? Which tools do you recommend?

Ok, so you are working in several step, saving the result in lossless for each intermediate steps.
So simply use:
- trim() to work on a given part of the movie
- then "+" to merge the parts together during the last step.

Hello, thanks thats what i was looking for. This could have problems with the IVTC?

Could you give me a quick example of the use of "+" ?


saluts

Can't be simplier .

For trim and IVTC, the only thing I can see is that you have a chance/risk to cut in the middle of a sequence (IVTC creates 4 progressive frames from 5 interlaced ones). So after the join you may have a glitch at that point.

Thanks for the example, so for the ivtc in theory is ok to cut in frames multiple of 5? this source has no proggresive, is like fully interlaced mixed with hard telecine of even double

I don't think that so simple else a complex IVTC would be useless. Frames needs to be analysed to determine what place in the 5-picture lenght sequence it occupies.
But don't worry for a 1 picture glitch, it's subliminal !

je yeah i think you are right.

thanks

Should lossless pass be saved in fast recompress in virtual dub? i understand that in that mode it does not convert color, so if i have yv12 source that should be the mode?

thanks again

Personnaly I don't know what's the diff with full mode.

It seems full mode also add a RGB conversion for virtualdub filters and then convert back to yuv. When serving with AviSynth fast recompress should be used to avoid the now useless color convertion