VideoHelp Forum

Try DVDFab and download streaming video, copy, convert or make Blu-rays,DVDs! Download free trial !
+ Reply to Thread
Results 1 to 9 of 9
Thread
  1. Member
    Join Date
    Sep 2015
    Location
    China
    Search Comp PM
    Sometimes,some interlacd still exist when I add some deinterlacing filters which can solve most interlacd.How can I solve them?
    What's the solving thinking?Here's my example.

    https://files.videohelp.com/u/245858/Young.m2v

    Code:
    MPEG2Source("Young.d2v")
    Telecide(guide=1)
    TDeint(mode=2)




    QTGMC can solve all interlacd in this case.
    But QTGMC always takes away too many details.
    So I don't like QTGMC too much but have to depend on it often for the complex hybrid sources.
    Damn...
    Quote Quote  
  2. Member
    Join Date
    May 2014
    Location
    Memphis TN, US
    Search PM
    The sample is hard-telecined.
    Why did you use deinterlace filters on telecined film ?

    Why didn't your script use field matching for inverse telecine:
    Code:
    AssumeTFF()
    TFM()
    TDecimate()
    Originally Posted by hintlou View Post
    QTGMC can solve all interlacd in this case.
    No, it can't.

    Originally Posted by hintlou View Post
    But QTGMC always takes away too many details.
    So I don't like QTGMC too much but have to depend on it often for the complex hybrid sources.
    Damn...
    You're not using it correctly. Why would would you use it in the first place for complex hybrid sources?

    The sample has illegal video levels and blown-out highlights.


    Your question has nothing to do with ripping DVD. Re-encoding and ripping are two different things.
    - My sister Ann's brother
    Quote Quote  
  3. Originally Posted by LMotlow View Post
    The sample is hard-telecined.
    Basically yes, but some (blown out) scenes with true 59.94 interlaced video, if I am not mistaken.
    Quote Quote  
  4. You're images aren't showing residual interlacing, they are showing aliasing artifacts from bad deinterlacing. Your source is a mix of telecined film, 30p and 30i, with some orphaned fields. Use QTGMC. There are settings that make it retain more detail -- see the SourceMatch and Lossless options.
    Last edited by jagabo; 15th Mar 2020 at 16:06.
    Quote Quote  
  5. Member
    Join Date
    May 2014
    Location
    Memphis TN, US
    Search PM
    I should think that if there are 30p and 30i (interlace and telecine), different parts of the movie would need different styles of cleanup. Likely everything should be 23.976fps for original film, then combined. Pulldown could then be applied to make it all 29.97. Why would you want QTGMC for the telecined parts and the progressive parts?


    Agreed, tdeint is a terrible choice.
    - My sister Ann's brother
    Quote Quote  
  6. Originally Posted by LMotlow View Post
    Why would you want QTGMC for the telecined parts and the progressive parts?
    Because it's a pain in the ass to break the video down into 0.3 to 5 second clips. And using QTGMC() on the entire video won't make it look any more jerky than playing the original 30i video in a 60p display.
    Quote Quote  
  7. Member
    Join Date
    Sep 2015
    Location
    China
    Search Comp PM
    Originally Posted by LMotlow View Post
    The sample is hard-telecined.
    Why did you use deinterlace filters on telecined film ?

    Why didn't your script use field matching for inverse telecine:
    Code:
    AssumeTFF()
    TFM()
    TDecimate()
    Originally Posted by hintlou View Post
    QTGMC can solve all interlacd in this case.
    No, it can't.

    Originally Posted by hintlou View Post
    But QTGMC always takes away too many details.
    So I don't like QTGMC too much but have to depend on it often for the complex hybrid sources.
    Damn...
    You're not using it correctly. Why would would you use it in the first place for complex hybrid sources?

    The sample has illegal video levels and blown-out highlights.


    Your question has nothing to do with ripping DVD. Re-encoding and ripping are two different things.
    Thanks for correcting my words.I'm a tiro about Re-encoding.
    (What's the correct meaning of "dvd ripping"? )

    How do you find that the sample has sample has illegal video levels and blown-out highlights?
    I don't know what is "illegal video levels and blown-out highlights" exactly.
    Thanks for your guide.
    Quote Quote  
  8. Member
    Join Date
    Sep 2015
    Location
    China
    Search Comp PM
    Originally Posted by jagabo View Post
    You're images aren't showing residual interlacing, they are showing aliasing artifacts from bad deinterlacing. Your source is a mix of telecined film, 30p and 30i, with some orphaned fields. Use QTGMC. There are settings that make it retain more detail -- see the SourceMatch and Lossless options.
    I added the SourceMatch and Lossless parameters,and tried different settings several times.
    They would influence the result badly.

    QTGMC(preset="veryslow",fpsdivisor=2) can solve there problems,
    but QTGMC(preset="veryslow",fpsdivisor=2,SourceMatch=X ,Lossless=X,...) can't :





    I am willing to solve these problems on clips.
    I'm not afraid of complicated steps because it's my learning process.
    If I would like to trim,what can I do to solve these problems?
    Thanks!
    Quote Quote  
  9. Originally Posted by hintlou View Post
    I added the SourceMatch and Lossless parameters,and tried different settings several times.
    They would influence the result badly.

    QTGMC(preset="veryslow",fpsdivisor=2) can solve there problems,
    but QTGMC(preset="veryslow",fpsdivisor=2,SourceMatch=X ,Lossless=X,...) can't :
    Yes, lossless and sourcematch do reduce the effectiveness of the antialiasing. With heavy processing like this everything is a matter of compromise. You can try an antialiasing filter afterward but that will blur other parts of the video a bit. Using Overlay() with an edge mask to limit the antialiasing to only high contrast edges might help.

    Why are you using fpsdivisor=2? That is loosing half the frames and motion during the 30i portions of the video and making the 24p portions jerky. Though with all the half second cuts it's hard to see.
    Quote Quote  



Similar Threads