With 35mm Dead, Will Classic Movies Ever Look The Same Again?

The content owners better archive all of their film libraries properly before the films are not in good enough condition to archive in some form of electronic storage.

Then again, the electronic capture and storage schemes keep changing (2K, 4K, etc).

However, will the old (and new) "trash" be kept up to date?

 

Interesting (and a bit ironic, to be "storing" digital content on film). Part of the high cost factor in digital archiving is undoubtedly due to the linear encoding which isn't as efficient for storage purposes as typical nonlinear home video. I can see how the bits could add up significantly with that, especially on 8k masters. Digital storage is continually dropping in price though, and becoming more "solid state", so I suspect that problem will be licked in the relatively near future. Using a "cloud" approach might also be feasible, especially for content in the public domain.

 

There was some movement towards nonlinear archiving (based on human perception and L*ab color, rather than physical intensity) for archiving stills in museums. Not sure where that's at now.  

By "linear" do you mean "uncompressed"? The term "linear" when used for motion pictures is usually associated with non-random access media such as tape or film. In that sense, both home and commercially-stored digital video is non-linear because it's on random-access media.

Howdy Sean,

 

I was referring to the probable method of converting the film's different density (or intensity) levels to digital RGB tristimulus levels, which I assume is probably linear (since CCD sensors are linear).

 

To make matters worse, you also have to contend with the film's own nonlinearity, which is roughly the inverse of human lightness perception and the encoding gamma used for home video content. If memory serves, motion picture and slide film is usually printed with about 2.0 gamma, which is roughly the inverse of the ~.50 gamma in Rec. 709 video encoding. You'll need even more bits in the initial scan to comfortably convert an ~2.0 gamma film image to an ~.50 gamma video image, and make up the bit-deficit near black.... maybe as much as 4x as many bits as the final video image. That's (potentially) alotta bits.

 

I'm guessin on a lot of this though, so the situation may not be as dire as I'm makin it out.