Saving raw video data before 2-pass enc.?

[LRDX]

I wondered, if there was a feature in GK to save the video picture after it had benn resized, filteres, cropped, etc. and before encode in az uncompressed/Huffyuv/lossless compressed format?

It would speed up the encoding process: why does the job of the decopresson of the MPEG stream, cropping, resize, and additional filters have to be done twice when compressing 2 pass (or n times @ n-pass)?

It would take an enormous HDD space, but this is the "age" of hundreds of GBs of HDDs.. (a 2H film@res 640x256, 16bit color, 25FPS, Huffyuv "encoded" would be around 20-30Gb..)

I would except a 2x or more speedup, based on encodings of DV => DivX and DVD => DivX. (I dont have any statistics, though.)

[len0x]

the video has never raw form as it is proccessed in YV12 mode anyway... Optimization you're describing is only possible if encoding was done via one tool, but we have three: avisynth -> vdubmod -> codec.

[LRDX]

I dont see why is it a problem:
Why is this impossible:
Avisynth->VDM->(uncompressed)->codec
(but.. what is this VDM stage, anyway? Isnt VDM "just" to run the codec?)

[len0x]

apart from implementation issue this will slow things down to hell (as we never get raw data have to do full decompress)