Panasonic 24p footage - workflow in CS4

Question

Hi there,
Although it’s my first post on this forum I’ve been reading and browsing for answers to my questions before – it’s not the case for me to give advises, not yet at least!!!
But this thread just confuses me a lot and I don’t think I fully understood the previous related discussions I’ve read. So here is my question:
What’s the workflow in Premiere CS4 with DVX100 footage, 24pA (although I also have some 24p Standard – meaning 2:3:2:3 pulldown cadence) to obtain a 24p video with Nooooo interlaced artifacts?
For      the 24pA, from what I reasoned myself and read on this forum, I chose      the 24p Standard present under DV-24p. Captured the footage      and applied the Interpret footage -> Remove 24p DV pulldown.  Set the in and out points; dragged      it on the timeline of a sequence (24p preset). None of the Source or the      Monitor windows show any interlaced artifacts and the timeline shows NO red line above. So it should be a 23.97fps footage, right?
BUT, as soon as exported (in the Microsoft AVI format with, for e.g.,  the NTSC DV 24p preset) I DO see the video as interlaced in BSplayer, while with Windows Media Player it shows no interlaced artifacts. I thought that my BSplayer is configurated wrong, but, any further compression I apply to this like exported video (a DivX codec, for e.g.), gives me a new video file with interlaced look in BOTH players. I’m loosing my mind – where do I do wrong? Project settings? Export settings?
(the footage was shot on tripod, but with a lot of motion in it - sport! And is designed to be exported for DVD and to be viewed on computer.)
2.      How could I mix 24pA footage with 24p Standard footage of the same DVX100? As, for mistake, the 1st day I shot 24p Standard instead of 24pA?
From what I  understood the 24p Standard need a 29,97 fps sequence – is it right?
Thank you,
Karmen

Answer 1

It works--I've done it dozens of times with hours of 24p (not 24pA) footage. The 24pA cadence wasn't necessarily designed to make 24p editing possible; it was designed to make 24p editing less taxing for a computer and somewhat higher quality, because a whole "frame" (and I say "frame" in quotes because there are no such things as "frames" in interlaced video) could simply be discarded.
I understand the logic of your thinking, but again, keep in mind that are no such things as frames in interlaced video. Each field is written independently, but we like to think of a video frame as being comprised of a top and bottom (or upper and lower, or odd and even) field because it's easier to comprehend that way. For the most part, this thinking works; that's the whole source of the "is it 30i or 60i?" nomenclature question.
Now, while I'm not too sure about the science or math of it--I try to stay out of the part--I believe the way that DV compression works is to package those two fields together as a "frame", simply for compression purposes. The fields are still independent, and can be recovered as independent entities, but it involves decompression to reverse the DV compression. Once you do that, you have two separate fields, and in the case of 24p, those independent fields can now be reassembled in the proper order to recreate 24 independent progressive frames. It is the 2:3 pulldown cadence that tells the software which fields it needs to use to recreate the independent frames. Of course, it's still DV, so in order for it to be played back properly, there must be a recompression step.
And this is why 24pA was invented. It's still recorded with fields, because it's interlaced DV at its heart, but the 2:3:3:2 cadence allows for the third "frame" of five to simply be ignored: everything the editing software needs to create a 24p stream is already encoded in the DV stream, without any decompression or reassembly needed. Recording using 24pA maintains a somewhat higher quality on those frames that would otherwise have to be reconstructed if the video was recorded with 24p, and there is theoretically less of an impact on computer performance because the decompression/recompression cycle is being avoid (though I doubt that has much bearing on any computer manufactured within the last few years).
I know it seems like voodoo, but that's really how it works (distilled a bit, I suppose!). Check out this article at Adam Wilt's website for more information, as well as some graphics that illustrate what's going on in 24p and 24pA recording.