Cineform Intermdiate from AVCHD / HDV

djkarn105

djkarn105

Well-known member
I have a question. On Cineform's site it says this:

When Neo Scene detects 4:2:0 chroma it properly interpolates the source chroma to 4:2:2 for more accurate color processing during editing and effects work. And if you ever "key" your material, CineForm’s chroma interpolation will substantially improve your resulting visual fidelity.
Click to expand...

now this sounds magic-tastic but it seems too good to be true. How can it create information that wasn't there to begin with?
How good is the interpolation really?

On the same note, does Apple Pro res do the same thing? Any opinions on which does it better?
 
It doesn't create something that wasn't there. It interpolates, as it says. Same thing as if you take a photoshop picture and blow it up -- it will average the surrounding pixels together to "guess" what the missing information would have been.

It's not doing anything "magical", it's doing what any native 4:2:2 codec would have to do, which is convert the 4:2:0 data, and invent the missing chroma information. As to *how* it does it, well, that's the question. It could be something as simple as duplication, or it could be a sophisticated 9's average or ... well... obviously some will do better than others, and there may be instances where a particular method works better for some things and not so well for others...
 
Barry_Green said:
It doesn't create something that wasn't there. It interpolates, as it says. Same thing as if you take a photoshop picture and blow it up -- it will average the surrounding pixels together to "guess" what the missing information would have been.
Click to expand...


when I say "not there" I mean not in the source image. Side by side, 4:2:0 footage transcoded to 4:2:2 would have less accurate color information that footage captured at 4:2:2 no?

As to *how* it does it, well, that's the question. It could be something as simple as duplication, or it could be a sophisticated 9's average or ... well... obviously some will do better than others, and there may be instances where a particular method works better for some things and not so well for others...
Click to expand...

That's where the magic (proprietary) part comes in I guess.

Seems like, bottom line, if I'm grading HDV / AVCHD footage it's a good idea to transcode to another format.
 
Last edited:
You're correct, it isn't there in the source image, so -- well, I guess the proper way to say it would be: it's not "real" 4:2:2, as it would be had it been captured in a 4:2:2 format in the first place.

It's an up-rezzed approximation of what it thinks that 4:2:2 might have looked like.
 
You must log in or register to reply here.
Back
Top