Hi, Using AutoGK, I'm trying to back up my TV DVD collection in Xvid format so that I won't have to use the disks so much. I've backed up several seasons already and the results were pretty good. Now I'm up to season 1 of Futurama in NTSC format and the problem I'm having is that AutoGK detects it as FILM, when really it is VIDEO and it needs to be inverse telecined (IVTC) otherwise the results gets way too jaggy. I had the same problem (except it was detected as hybrid) on season 7 of The Simpsons, and that was resolved by forcing it to IVTC when detecting video as hybrid (both FILM and VIDEO.) But this solution only works when it detects as hybrid and my Futurama DVD is detected as FILM. Does anyone know a way to force IVTC even when it detects the video as FILM? EDIT: Actually, now that I look closer at the Simpsons rips they have the same problem but not as much as the Futurama rips. On the Futurama rips it is noticeable as soon as there's any movement but on the Simpsons rips I can only see it on characters' mouths when they're talking, but only when the picture is zoomed out so that the characters appear small. Is there some copy protection scheme by those Fox assclowns that's causing this or what?
I think I know what the problem is. I've looked through some logs and it seems on all jaggy encodings, Auto Gordan Knot has decided to switch of B-frames and I think that's what's causing the jaggieness. I can't find a way to force it to have B-frames enabled though.
i think your problem here is called combing. i never used AGK, coz i do the script manually. but you could try using different IVTC filter. i like TIVTC by tritical, avisynth filter of course.
