eedi2 and tivtc

[codemaster]

There are several types of deinterlacers: field elimination, spatial adaptive, motion adaptive, motion search adaptive, blend, bob.

If I understand correctly, the motion search adaptive deinterlacers (for example, using EEDI2 + TIVTC) seem to output the highest quality results. The drawback is that they are very slow.

I do not fully understand how EEDI2 + TIVTC should be used for deinterlacing. Can someone write an avs script with an example of how to use EEDI2 + TIVTC properly ?

Or, what other hi-quality-no-matter-how-low-the-speed deinterlacers would you recommend ?

The author of QTGMC says that it is "a very high quality deinterlacer", but is one of the QTGMC presets in the motion search adaptive category ?

[manono]

TIVTC isn't a deinterlacer. It's an IVTC, a field matcher and decimator. It shouldn't be used just for deinterlacing. The use of EEDI2 and other deinterlacers within TIVTC is for when TIVTC doesn't find a suitable match for fields and they get deinterlaced. With properly telecined material (not always the case) a post-processor (deinterlacer) shouldn't be used very often. As for a sample script, it's in the manual:

My usual script these days for anime ivtc is:

mpeg2source("")
inter = separatefields().eedi2(field=-2)
deint = tdeint(mode=2,edeint=inter)
tfm(d2v="",clip2=deint,MI=60)
tdecimate(mode=4+5)

That's from the 'Common Usage Examples.txt'. Or, from the 'TFM - Readme.txt':

clip2 -

When using PP > 1 this parameter may be used to specify a clip to take deinterlaced
frames from instead of using one of TFM's inbuilt deinterlacing methods. If PP < 5
then frames will be taken as is from clip2, if PP >= 5 (i.e. using motion adaptation)
then TFM will build the mask as usual and only pixels in moving areas will be taken
from the frames in clip2. Don't specify anything for clip2 to use TFM's internal
deinterlacing. For an example of using clip2 see the clip2_sample_avs.avs file.
.
.
###From the clip2_sample_avs.avs:
d2vpath = "c:\mysource.d2v"

mpeg2source(d2vpath) #assume tff source
deint = last.bob().selecteven() #simple cubic deinterlacing (keep top field)
tfm(PP=5,clip2=deint) #motion masked post-processing since PP > 4

If you really want to deinterlace something as well as possible, then use QTGMC. If you still want to use EEDI2, EEDI3, or NNEDI3 as deinterlacers, read their respective manuals.