I've been reading on using Cineon/DPX Log for archival purposes and there are some points that are not quite clear to me. Maybe there are still some people here who know about these things.
As far as I understand, Cineon was created by Kodak for usage on their Cineon scanner which was more like a densitometer taking lots of density readings for every image. These were then saved as corresponding digital values with a log curve applied to have more values for the darker densities in the negative and therefore preserve the very bright highlights.
How does this work with modern scanners these days?
As far as I know current sensors record the information in a linear way.
So where is the advantage of using DPX Log these days?
Isn't it just a LUT applying the Log curve and an offset for black and white?
Where is the difference in quality compared to gamma corrected linear 10bit DPX?
And how does all of this apply to scanning print film, which has different densities and gamma compared to negative. Is there a point to scanning print as DPX Log (even back then on the Cineon scanner)?
before you go down the route of the log to linear nightmare, do they accept EXR? its much easier to work with(and newer). you can do half float which is lossless 16 bit images which is really nice or full 32bpc that can hold Film's/RAW's latitude. That being said, most DPX stuff is full range unless specifically told otherwise(and its in LOG) and 8 bit log losses quality. You can go 10 bit DPX if you go LOG. You can go linear DPX if it is 16 bit or you lose precision. DPX is old and the production pipeline can get muddled if you don't have a vfx supervisor as all composites have to be converted from log to lin first. This is why most small studios go EXR. It came from ILM and way simpler to use. ie. not mess up as often.