This forum uses cookies
This forum makes use of cookies to store your login information if you are registered, and your last visit if you are not. Cookies are small text documents stored on your computer; the cookies set by this forum can only be used on this website and pose no security risk. Cookies on this forum also track the specific topics you have read and when you last read them. Please confirm whether you accept or reject these cookies being set.

A cookie will be stored in your browser regardless of choice to prevent you being asked this question again. You will be able to change your cookie settings at any time using the link in the footer.

Possible Problem with Input 2:3 Pulldown VOBs and MKVs
#1
Things where slightly adjusted in dev. versions when you fixed DVD importer, but i guess some mistakes remained somewhere. This is only guess, so correct me if i am wrong. Here is example of same VOB source:

VOB Mediainfo:
Frame rate 23.976 (24000/1001) FPS
Scan type Progressive
Scan order 2:3 Pulldown

VOB opened in Hybrid:
telecine
29.970/CFR
Image looks normal if i apply Deinterlace->TIVTC or VIVTC

VOB muxed with MKVToolNix GUI to MKV Mediainfo:
Frame rate mode Variable
Frame rate 23.976 FPS
Scan type Progressive
Scan order 2:3 Pulldown
(VTS_01_1-TEST3 example muxed by MKVToolNix to 24.811. Have no idea why)

VOB muxed with MKVToolNix GUI to MKV and opened in Hybrid:
progressive
23.976/VFR
If i set input as progressive, i can see 2 interlaced frames and 3 progressive frames pattern.
If i apply Deinterlace->TIVTC or VIVTC output framerate changes to incorrect 19.1808 fps.

Image looks normal and framerate remains unchanged only if i apply Frame->Misc->TFM or VFM

So the questions are rather complex:
Why Hybrid reads 2:3 Pulldown MKV files in this way?
Is it normal to use Frame->Misc->TFM or VFM in this situation?
Why MKVToolNix changes files metadata like this?
Where came from those strange framerates in some muxed files? Corrupted VOB or maybe a bug in MKVToolNix?

Maybe for situations like these it make sense somehow add an option for Hybrid to manually set input framerate and scan type independently before deinterlacer?
Or somehow else adjust things?

Here is a folder with VOBs and muxed MKVs if it helps: https://drive.google.com/drive/folders/1...sp=sharing
Reply
#2
Will look at it after work.
----
Dev versions are in the 'experimental'-folder of my GoogleDrive, which is linked on the download page.
Reply
#3
Okay, here's what happens:
Since MediaInfo often reports vobs and (mvks created by MakeMKV) as
Frame rate 23.976 (24000/1001) FPS
Scan type Progressive
Scan order 2:3 Pulldown
and VFR
even if the whole clip isn't soft delecined.
Personally I think, such content should be muxed as 23.976 progressive without the 2:3 pulldown and vfr.

Since the answers over at https://forum.doom9.org/showthread.php?t=181825 didn't really help I added special workaround to Hybrid to threat such content as 29.97 telecined. This way processing is a bit slower, but mixed soft and hard telecine content doesn't cause problems.
Hybrid does add those workarounds for mkv files that are not created by MakeMKV.

To properly handle soft telecine, time codes need to be dropped and the used decoder need to support it.
If you can share a !short! and small sample of such an mkv I can look at it.

Cu Selur
----
Dev versions are in the 'experimental'-folder of my GoogleDrive, which is linked on the download page.
Reply
#4
As i wrote earlier - i upload few sample files here, But they are untouched VOBs about 1GB each because i don't know how to trim part of the VOB without possible damage https://drive.google.com/drive/folders/1...sp=sharing
Reply
#5
Since the issue isn't with vob, but with mkv and you use mkvtoolnix, simple split the mkv on remux and upload only a small sample which contains a scene which allows to reproduce the issue.

Cu Selur
----
Dev versions are in the 'experimental'-folder of my GoogleDrive, which is linked on the download page.
Reply
#6
Ok added trimmed versions to same folder.
VTS_01_1-TEST1 (trim) is normal example to illustrate pulldown metadata.
VTS_01_1-TEST3 (trim, strange framerate) just illustrates random framerate problem that happens during mixing. I still don't know if it is due corrupted original VOBĀ file or due something special in that VOB.

Also added VTS_01_1-TEST2 (trim) example that includes more movement to see interlaced artifacts, if it helps.
Reply
#7
About the frame rate: Enabling "Config->Internals->Prefer Original->Frame rate" will help.
Why this happens is probably because MediaInfo analyses the first few time codes, calculates an average and that is that is what happens. Wink
----
Dev versions are in the 'experimental'-folder of my GoogleDrive, which is linked on the download page.
Reply
#8
Quote:VTS_01_1-TEST1 (trim) is normal example to illustrate pulldown metadata.
looks like progressive content

Quote:VTS_01_1-TEST2 (trim).mkv
looks like mixed content

Had a quick look at the file and since ffms2.Source (Vapoursynth) unlike the old FFmpegSource2 (Avisynth) has no 'rffmode' to allow honouring the pulldown flag there is no option with Vapoursynth to handle soft telecine ideally by simply just taking the progressive frame and ignoring timecodes&co.
-> I'll probably adjust Hybrid to treat 23.976fps + Pulldown as telecine with 29.97 fps.
Since telecine + ivtc should be lossless, this would only cost a bit of speed as it does when Hybrid treating vob&MakeMKV-mkvs.

Cu Selur
----
Dev versions are in the 'experimental'-folder of my GoogleDrive, which is linked on the download page.
Reply
#9
VTS_01_1-TEST1 is same as others. There is just not too much movement in it. Interlaced frame visible when hands are moving in the end. Or did you read that file somehow in different way and see that it is progressive?
[Image: mvKA0VT.jpg]

Treat 23.976fps + Pulldown as telecine with 29.97 fps sounds logical because it threats VOBs (and DVDs) in same way.

By the way, will in that scenario IVTC deinterlacer type ("inner" vs "QTGMC") affect speed or quality?
Reply
#10
Quote:By the way, will in that scenario IVTC deinterlacer type ("inner" vs "QTGMC") affect speed or quality?
If the content is real soft telecine probably both, since QTGMC does lots of additional progressing. (not totally sure since it's a while since I read up when ivtc uses the deinterlacer)

iirc I had some sources where MediaInfo reported 'telecine (soft)', but that might have been an old MediaInfo version.

Cu Selur
----
Dev versions are in the 'experimental'-folder of my GoogleDrive, which is linked on the download page.
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)