Cinecompression is yet another result of the
shotgun wedding between
film and
video. It is the process of removing the extra
fields of video created in the process of
transferring 24-frame-per-second film to 29.97-frame-per-second NTSC video.
If you're totally lost at this point, follow the above pipelink for gratification and edification.
Ah! I see you've returned, armed with a complete understanding of 3:2 pulldown! Excellent. Let us continue.
So here we are, with our video transfer of our film footage, and it has all these still frames every fifth frame. (note... this is only how it appears... what you really have is one field of video repeating from the frame previous, and one field of video anticipating the next frame along, combined in an unholy evocation of a single video frame)
This state of affairs is an opportunity on the one hand, and a potential problem on the other; it all depends on what you need to do with you video footage. I shall address each potentiality.
Opportunity- you are cutting a thirty-second television commerical, and it's durn hard to work all the story elements you need (or, more likely, product shots) into thirty seconds. How to use more footage in less time? Speed it up. You have the option of using some kind of speed effect on your video, but chances are it's going to look exactly like you used some kind of cheap speed effect. However, by removing the redundant pulldown frames from a segment of your video footage, you have decreased the runtime with an almost imperceptible shift in the quality of the image. Almost imperceptible. It is sometimes obvious that you have "sped up" the video... but, on the plus side, it's very clean-looking and artifact-free.
This trick cannot be used for any footage that has sync audio, unfortunately; if you start yanking frames out of your video, you completely lose touch with the rate at which your audio was recorded at the shoot. However, dialogue can sometimes be cheated by clever sound designers to appear in sync.
Potential problem- Your commercial needs effects! What would modern advertising be without heavy composite work, titling, or other eye candy? Bland! So you're gonna do it, and you're gonna pay through the nose for it, by the hour. Effects work, which is nothing more or less than animation, is done frame by frame. Sure, computers are doing it, so it's done relatively quickly. But it's still frame by frame. Hmm... how to speed up this process?
Yes, yes, you learn so quickly! Remove those redundant pulldown frames, and you've cut your workload by one frame in five. Over time, that adds up considerably.
The really neat trick, though, is that once the effects work is done, those pulldown frames must be added back so as to avoid the problem of having your effected footage fall short, timewise. So, during the render process, your $100,000 piece of software creates those new frames for you. Isn't that nice?
Speaking of software, the only way to do this is with a program that can seek out the frames of video with combined fields, remove the offending information, and interpolate a new frame from the result. combustion and its more expensive cousins, flame and inferno, manage this quite handily.
The whole issue of cinecompression, not to mention 3:2 pulldown, would be eliminated if cinematographers chose to use the feature on their cameras which allowed them to shoot film at 29.97fps. However, the runtime opportunities provided by cinecompression, as outlined above, have become such an integral part of the commercial post production process, that this simple solution is unlikely to be implemented.