Posts: 785
Threads: 16
Joined: Mar 2020
Probably animations like these need some smarter frames extraction methods.
Sort of somehow auto detect:
Is there is a normal 5 frames cycle (3 progressive, 2 interlaced)
Is there some 1 frame progressive, 1 interlaced cycle.
Or if there some other unusual pattern (like 4 frames progressive, 1 interlaced)
And based on this detection somehow select better frames based on detection of line jaggies and line smoothness.
Guess this is near impossible in real life
It just feels very offensive that all good frames are there and potentially they may be extracted somehow back to normal looking original 23.976 clip without any additional post processing and smoothing filters. But detelecine filters can't do it perfectly due that "variable telecine" pattern. Or we just miss some secret about proper processing of this animation.
Posts: 785
Threads: 16
Joined: Mar 2020
I got some feedback from authors of good looking DVD rips:
"Used the Yatta software, in which you can detect patterns and enforce them (for the 3:2 sections) as weell as do other manual stuff like freezeframing the cuts where one field is missing and you can also select the 30fps sections so that they don't get decimated, the software generates timecode files for you and the end file is VFR then - 24fps where it was IVTCed and 30fps elsewhere.
It was an awful lot of work to do it that way (besides watching out for 60i fades done on top of the ^ and having to apply debanding in places)"
Yatta software is some very special Avisynth tool
https://forum.doom9.org/showthread.php?t=161653
So it is possible to autodetect those variable telecine patterns, but it is not a one click workflow.
There is also YATTA for VapourSynth (real name: Wobbly)
https://forum.doom9.org/showthread.php?t=172496
Posts: 10.618
Threads: 57
Joined: May 2017
Yeah, I know Yatta&Wobbly, but I probably would simply bob and filter the clip instead of spening the time to create vfr content from it.
Quote:So it is possible to autodetect those variable telecine patterns, but it is not a one click workflow.
Not really auto detection if you have to tweak tons of values, but yeah it's possible just not with simple VIVTC/TIVTC; like I wrote before AnimeIVTC might also be up for the job.
Cu Selur
----
Dev versions are in the 'experimental'-folder of my GoogleDrive, which is linked on the download page.
Posts: 785
Threads: 16
Joined: Mar 2020
Seems there is a quick and easy problem fix provided by PRAGMA here
https://forum.doom9.org/showthread.php?p...ost1954089
Any chance to implement PD2V class of pvsfunc directly in Hybrid UI somewhere in future?
Posts: 10.618
Threads: 57
Joined: May 2017
Somewhere in the future: may be
Any time soon: no
On MacOS: Probably not unless I either find a way to make python portable or give the user has to manage the python environment etc. on it's own.
also since
https://github.com/rlaphoenix/pvsfunc relies on DGIndex D2V project files the question is whether the project files created by d2vwitch contain the all the info the d2v files from DGIndex contain (not sure whether DG included some additional meta data), since DGIndex is not available on MacOS.
On Windows with a special addOn for it: may be, but probably not.
Cu Selur
----
Dev versions are in the 'experimental'-folder of my GoogleDrive, which is linked on the download page.
Posts: 785
Threads: 16
Joined: Mar 2020
Yes, that is not optimistic...
Posts: 10.618
Threads: 57
Joined: May 2017
After some further thinking about it, the python portable might not be the problem since Python must be present to use Vapoursynth to begin with.
But the other points are all valid. So I don't plan it for the forseeable future atm.
Cu Selur
----
Dev versions are in the 'experimental'-folder of my GoogleDrive, which is linked on the download page.