Converting wrongly captured 59.94fps to 29.97fps

[waxstone]

Title sequences are often very different from the main body of the show. Titles are often overlaid as 30i video or 30p video, regardless of the frame rate of the underlying show. You may often find that the underlying video is slowed or sped up to match the theme song -- so the pulldown may not be the standard 3:2 pattern.

Since it's likely your six hour long 59.94 fps caps aren't perfect (no dropped/inserted frames) the easiest thing for you to do is recapture the video at the proper 29.97 fps then double frame rate deinterlace with QTGMC() or maybe BWDIF() or Yadif(), and encode at 59.94 fps.

This seems like the best option to avoid any more issues. To an extent converting the video from the incorrect capture frame rate seemed straightforward but I'd rather eliminate issues where possible.

I know that many suggest re-encoding to 24fps to match the animation but I don't want to lose any frames for any of the title and animation content, there are so many sequences within digital animation in 30fps. 59.94fps means no loss of frame, and it should be able to achieve a buttery smooth deinterlaced effect for the live action commercials and sequenced.

If there are multiple different shows, about 6 to 12, per tape. This is completely due to my lack of knowledge, but is it likely that I will have use a different deinterlacing filter for each segment? E.g. QTGMC for one and BWDIF for another.
This is me assuming that the field order might swap around between each recorded content on the tape, or is that field order is determined on the actual video capture?

[jagabo]

Field order is determined by the capture device and software. Analog video has no field order because it has no frames. It's a continuous stream of alternating top and bottom fields. It's only when it's digitized and packaged as frames that the the frames have a field order -- which of the two fields to display first at playback.

A capture device may start by capturing a top field, then weave in the next field, a bottom field) creating top field first frames. Or it may start with a bottom field, then add in the next field, a top field, to create bottom field first frames. They generally don't switch back and forth between TFF and BFF.

So a normal 29.97i capture will have all TFF frames (or all BFF frames). If a frame is dropped during capture (or edited out), the next frame will still be a TFF (or BFF) so your video will continue to be all TFF (or all BFF). With your 59.94i caps this isn't true. A singled (or any odd number of) dropped frames will result in a field order inversion after SelectEven or SelectOdd.

As far as deinterlacing is concerned, QTGMC is generally the best choice. It sometimes leaves a little ghosting, moire artifacts, or blurring but overall it gives much better results than BWDIF which is a far second-best for most material. The big disadvantage is the speed -- QTGMC is much slower than BWDIF.

[waxstone]

Field order is determined by the capture device and software. Analog video has no field order because it has no frames. It's a continuous stream of alternating top and bottom fields. It's only when it's digitized and packaged as frames that the the frames have a field order -- which of the two fields to display first at playback.

A capture device may start by capturing a top field, then weave in the next field, a bottom field) creating top field first frames. Or it may start with a bottom field, then add in the next field, a top field, to create bottom field first frames. They generally don't switch back and forth between TFF and BFF.

So a normal 29.97i capture will have all TFF frames (or all BFF frames). If a frame is dropped during capture (or edited out), the next frame will still be a TFF (or BFF) so your video will continue to be all TFF (or all BFF). With your 59.94i caps this isn't true. A singled (or any odd number of) dropped frames will result in a field order inversion after SelectEven or SelectOdd.

As far as deinterlacing is concerned, QTGMC is generally the best choice. It sometimes leaves a little ghosting, moire artifacts, or blurring but overall it gives much better results than BWDIF which is a far second-best for most material. The big disadvantage is the speed -- QTGMC is much slower than BWDIF.

Thanks for the very clear explanation, that puts my mind at ease. I haven't had any frame drops logged previously. I'm about to continue with my re-captures, do these settings look fine?


[Attachment 86680 - Click to enlarge]

[Attachment 86681 - Click to enlarge]

[Attachment 86682 - Click to enlarge]

[Attachment 86683 - Click to enlarge]

[Attachment 86684 - Click to enlarge]