I may have misspoken or misreported some information in my Death Wish comments, and desire to make amends.
I've taken a longer look, and here's what I'm seeing.
It appears that the film may partially be derived from IP or dupe, but in deference to the studio techs, it's extremely difficult to tell the difference between the three, especially if at some point rolls may have been mis-canned.
A dupe neg in an OCN can? No real way to tell the difference.
An IP in an OCN can? Ditto. Orange-based Kodak stock is all the same. Even if one looks at the perforations for a clue. They're all the same.
Set up an IP on the scanner rather than a negative? No real difference.
It's also possible that mis-canned rolls were sent to a scanning facility which also could not tell the difference between an OCN, a dupe and an IP.
If you recall, I noted that it's extremely difficult to tell the difference.
Without checking timing, I'd say that the first two or three reels are derived from some element other than an OCN, as they have a different appearance on screen, with more movement. It's possible that DP Arthur Ornitz or camera operator Lou Barlia may have been consuming too much caffeine, which affected the stability of the image - but these gentlemen knew their way around a camera, so it's doubtful.
It may be that the scanning facility had an extremely loose transport. Who knows?
Or all of the reels could be from OCN. If that's the case we'll go with the problematic scanner, second rate scanning facility, or caffeine induced camera crew.
As far as grain, it's entirely possible that parts of the film were shot with Kodak's (new at the time - and experimental) "Living Grain" stock, which allowed grain to move around and take different positions on the film, joining friends in granular peace and love.
From a nominal seating distance, everything is fine, and that fact that there is no 4k information is irrelevant to most viewers, who can't tell the difference anyway.