AssumeFrameBased

[MattO]

Hello all,

I use the following line in my AVS scripts:

AssumeFrameBased().SeparateFields().SelectOdd()

TRBarry replied to one of my previous posts stating if I was using AVISynth 2.5 this line may not have the same effect, can anyone elaborate on this?
If this line will not work is their an AVISynth 2.5 equivilent?

Also an IVTC question: In AVISynth 2.07 I always use:

ivtc(70,10,25).AssumeFrameBased().SeparateFields().SelectOdd().AssumeFPS(25)

with NTSC sources as part of my process to convert them to PAL. I am perfectly happy with this method, so please don't flame me ;)
Does DeComb have an equivilent (standard) line to ivtc(70,10,25) ?
or is there another IVTC plugin (YV12) which can do something similar?

thanks

MattO

[sh0dan]

http://www.avisynth.org/index.php?page=AviSynthTwoFiveZeroBugs

and

http://www.avisynth.org/forum/viewtopic.php?t=13



In short - It's a known bug, being worked on.

[Guest]

Originally posted by MattO
I use the following line in my AVS scripts:

AssumeFrameBased().SeparateFields().SelectOdd()

TRBarry replied to one of my previous posts stating if I was using AVISynth 2.5 this line may not have the same effect, can anyone elaborate on this?
If this line will not work is their an AVISynth 2.5 equivilent?I don't know what Tom said, but I can tell you something about it. If you have YV12 data then SeparateFields() only works properly if the source was chroma sampled using the interlaced algorithm, and not the progressive one. Separating the fields of a progressive sampled YV12 gives you similar problems to that of the infamous chroma upsampling bug.

This is in direct conflict with the limitation that SeparateFields() requires frame-based, and frame-based is currently being assumed to imply progressive content. Complete sanity has not yet been brought to field handling in Avisynth 2.5, but that area is being actively addressed for future releases.

Separating the fields of an interlaced-sampled YV12 frame creates new frames that are progressive-sampled. I would be happy if all the parity handling code was removed and we just had a Separate() function that took a parameter for ordering of the resulting stream. It would work on any content and it would be up to the user to ensure that what he does makes sense, given the source material. I've made a proposal in this regard here:

http://www.avisynth.org/forum/viewtopic.php?t=13

Does DeComb have an equivilent (standard) line to ivtc(70,10,25)? or is there another IVTC plugin (YV12) which can do something similar?
Decomb and IVTC() use different basic approaches. There is no way they can be equivalent with any settings. When you say "something similar" do you mean performing IVTC? :) Or is there some specific behavior of IVTC() you are looking to duplicate?