TFM documentation should mention that modes >0 are good for fixing perverse telecine

[Katie Boundary]

I've always wondered why TFM bothers with including modes other than 0. I've always set it to 0 to save a few CPU cycles and reduce the chances of an incorrect match.

I tried ripping and converting "Due South" yesterday, only to find that TFM at my preferred settings was having difficulty with finding proper matches for most fields. Suspecting that it might have been shot at true 60 hz, I tried bob(), which revealed nonlinear field behavior. Suspecting a field order error, I tried complementparity().bob(), but that didn't fix it. On a hunch, I tried setting TFM to Mode 1, and lo and behold, progressive film frames magically appeared. It just turned out to be one of those whack telecine jobs done by people on PCP.

I think that mentioning this sort of use case in the TFM documentation would be more helpful than just "increases your chances of finding a match if one exists".

[johnmeyer]

Another trick is to force the field order to upper or lower. This can also solve problems like the one you describe.

[Katie Boundary]

Originally Posted by johnmeyer

Another trick is to force the field order to upper or lower. This can also solve problems like the one you describe.

That would only work in situations where complementparity() or assumebff() would also work.

[StainlessS]

Perhaps of interest Katie.

https://forum.doom9.org/showthread.p...99#post1897899 [Dont know if you know but, little blue arrow after StainlessS {EDIT: in quote} below is this same link]

Originally Posted by StainlessS

Here an explanation of Phase Shifted (AKA Perverse) Telecine [click a 2nd time for magnified image].


Taken from Scharfis_Brain Exotic Interlacing [originally in German],
see DATA Folder @ MediaFire in my sig below this post for PDF's in English or German.
[There is also a zip containing Web page]

EDIT: The Website is no longer online, but LigH posted a link to his saved copy somewhere.

EDIT: Original web page on WayBack Machine:- https://web.archive.org/web/20140411...esInterlacing/

And a thread I started on it (forgot about it) some time ago:- https://forum.doom9.org/showthread.p...ic+Interlacing

EDIT: LigH hosted original web page in German:- https://www.ligh.de/scharfis_brain/E...esInterlacing/

NOTE in the IMAGE from Scharfis_Brain

Recovery in Avisynth

DoubleWeave().SelectOdd()

Perhaps it works for U.

EDIT: Also, TFM(d2v=d2v_Filename, ...) # works better with d2v arg where source is DgMpgDec's Mpeg2Source(d2v), rather than other source filters.
eg

        Video = Mpeg2Source(VideoFileName)
        Video = Video.TFM(d2v=VideoFileName) # + any other args

Also, see the 2.1.3 Telecine with PAL speedup - dynamic phase SHIFT section in image with correction using Telecide().
ie

Telecide(order=1,guide=2,post=0)