The Difference of Video Type (FILM/NTSC/PAL/HYBRID)

FILM Forced FILM ON	23.976 fps progressive frame coded on DVD Forced FILM skips the RFF (repeat_first_field flag) of MPEG-2 stream
FILM Forced FILM OFF	29.970 fps 3:2 pull-down (by DVD2AVI) based on orderly RFF RFF sequence behaves as 01010101... every four frames add two fields (one frame) That's the theorem of 23.976/24->29.970/30 fps (4:5) conversion
NTSC/PAL Forced FILM OFF	29.970/25.000 fps interlaced or progressive frame coded on DVD all RFF of MPEG-2 stream equal to 0, unnecessary to repeat disable Forced FILM to set correct frame rate, NTSC 29.97 fps or PAL 25 fps
HYBRID Forced FILM OFF	29.970 fps irregular pull-down (by DVD2AVI) due to unorderly RFF Try to skip RFF will result in meaningless stuff and cause A/V asynchronism.

• By obeserving the relation between coded and playback frame number, I think you will catch my meaning.

• From another point of view, HYBRID is compounded of both FILM & NTSC and must accomdate to NTSC.
  It's impossible to turn HYBRID back to FILM automatically with 100% accuracy cause of irregular pull-down.
  What we can do is Inverse Telecine by TMPGEnc automatically and check/correct frame by frame manually.
  If you are skilled in Inverse Telecine, you can even turn the NTSC which pull-down from FILM before MPEG-2 encoding
  back to FILM also. Most Anime DVDs are encoded in such a way. Indeed it's not a good process of DVD mastering.
  A few of Anime DVDs are encoded as FILM, such as Ah! My Goddess! OVA and the movie of CardCaptor Sakura.

• Forced FILM is based on RFF detection and frame decimation/duplication.
  NTSC or PAL + Forced FILM ON -> garbage
  FILM + Forced FILM ON -> synchronous 23.976 fps flawless FILM (equals to IVTC)

• HYBRID + Forced FILM ON -> synchronous 23.976 fps semi-FILM
  Depending on the nature of video sources, unsmooth playback or interlaced frame is possible.
  If you can't satisfy with the result, switch to HYBRID + Forced FILM OFF -> IVTC -> FILM

• Another alternative is using adaptive deinterlace filter of TMPGEnc/AviUtl/VirtualDub to deal with such kind of video,
  though the quality downgrades and frame rate still remains 29.970 fps that wastes more disk space.

The AC3/MPA/DTS Stream DELAY Computed by DVD2AVI

• The delay time of AC3/MPA/DTS stream is recorded on filename.

• This value is derived from the difference between audio and video PTS (Presentation Time Stamp),
  the real playback time of each audio and video packet.
  Please look at the real log ripped from CardCaptor Sakura OP1:

  Video Packets Audio Packets Video Packets Audio Packets Video Packets Audio Packets Video Packets Audio Packets Audio Packets Video Packets Audio Packets Video Packets Audio Packets Video Packets Audio Packets Video Packets Audio Packets Audio Packets Video Packets Audio Packets

  [PTS: 0.223 sec] [PTS: 0.223 sec] [PTS: 0.287 sec] [PTS: 0.319 sec] [PTS: 0.351 sec] [PTS: 0.383 sec] [PTS: 0.415 sec] [PTS: 0.447 sec] [PTS: 0.479 sec] [PTS: 0.511 sec] [PTS: 0.575 sec] [PTS: 0.790 sec] [PTS: 0.607 sec]

  I P B P B B P B B P B B P B B I

  The beginning of selected region is aligned to I-frame by DVD2AVI and only I-frame owns a PTS.
  DVD2AVI will start to rip audio packet from the decided I-frame.
  Assuming that we decide to rip from the second I-frame...
  The audio delay of second I-frame is computed as PTS(A) 0.607 - PTS(V) 0.790 sec = -183 ms (audio is ahead)
  The playback time of audio/video packet is different, so we should cut out 183 ms of audio in order to synchronize.

• The delay time of WAV (Dolby Digital Decode/Linear PCM) is already compensated by DVD2AVI.
  Only raw streams (AC3, MPA, DTS demux) need to be compensated manually.

• According to statistics, the perceptual tolerance of audio/video skew is about 80 ms (2~3 frames)
• A few of DVDs are badly encoded. Even played on a DVD player still results in asynchronism, like this one.