Difference between Field Operation "None" and "Force Film"

[ByteTree]

What is the difference between setting Field operation to "None" vs. "Force Film". From what I can tell, the only difference is in the .d2v file where Field_Operation and Frame_Rate changes. Will the rest of the .d2v file always stay the same? Will demuxed audio change?

[neuron2]

Originally posted by ByteTree
What is the difference between setting Field operation to "None" vs. "Force Film". From what I can tell, the only difference is in the .d2v file where Field_Operation and Frame_Rate changes. Will the rest of the .d2v file always stay the same? Will demuxed audio change? Did you read the second and fourth stickies in this forum? They both discuss Force Film. Read that material first.

Yes, the only difference as far as DGIndex is concerned is what it writes in the D2V file as you say. The rest of the D2V file stays the same and the audio is not changed. The changes in the D2V file do, however, radically affect how DGDecode decodes and displays the video! After you have read the afore-mentioned material (especially the fourth sticky), it should be clear what is done differently.

[Xesdeeni]

I have a few comments on the contents of the above deep-linked article (I would have posted it in the sticky above, but the thread is closed), specifically in relation to the SmoothDeinterlacer bullet:

* KernelDeint has replaced SmoothDeinterlacer as my favorite deinterlacer (in spite of my having ported SmoothDeinterlacer to AVISynth :) ), and there is an MMX variant (LeakKernelDeint) that is very fast.

* You note that SmoothDeinterlacer "- creates large files ":

Can you please be more explicit in this. In the domain in which SmoothDeinterlacer operates, each frame is the same size, and all the deinterlacers will produce the same size data. I assume what you mean is that lossy encoding (e.g. MPEG) produces larger files, but I think this is misleading. This is listed as a con, but the encodes are only larger because more data is recovered. That is a Good Thing. You could blur the image and make the encodes smaller, but that's not the point of deinterlacing. The point is to recover as much of the original (as if the source were captured progressively) as possible. But more detail means larger encodes.

Xesdeeni

[neuron2]

Is this in reference to something in the FAQ?

Cyberia is the FAQ maintainer, so please PM him about it. Thank you.