A few questions about film projection (and one about digital de-anamorphization)

[cole t parzenn]

Are my eyes playing tricks on me or digital de-anamorphization make the grain look spherical, while optical de-anamorphization makes it look stretched? If the latter, why?

 

I saw film projection for the first time in I-don't-know-how-long, last night (Insomnia, shot in scope by Wally Pfister on Kodak 5246 and 5279, printed on Fuji F-CP 3519D) and it raised a few questions:

 

How much of the grain (there was plenty) comes from the camera stocks, the three generations' removal from the negative, and the small format, respectively?

 

If resolution and graininess can be separated, is the resolution of a print limited by the negative, the printing process, or both? With blowups and reductions, is there a visible difference in resolution, or just visible graininess? Why or why not?

 

In wide shots, a character's corduroy jacket seemed to induce very subtle moire, like there were lines going in both directions. But film doesn't moire, so what was I seeing?

 

Maybe it's the cinematography and not the projection but the close ups looked really good. Better than I've seen from digital projection. I'm not sure that I'd say there was more detail in the actors' faces but there was somehow more nuance. Any idea why this was?

 

Grazie.

[David Mullen ASC]

It's a skinny image exposed onto normal shaped grain and the grain only gets stretched horizontally in order to make the skinny image look normal, so the effect on the grain would be the same whether the image was stretched optically or spherically. But one thing you get with scope print projection is often the mediocre anamorphic lens on the projector, which is an additional factor in softening the image -- many scope projectors cannot hold sharpness from corner to corner. This is one reason why 70mm blow-ups looked so good, you had a larger print area and spherical projector lenses with less image enlargement compared to 35mm projection in the same theater.

 

Negative stock has much larger grains in it because the speeds are so much higher than print and dupe stocks, where are below 10 ASA in speed, so most of the grain you see comes from the original negative, only a little from the dupes and print stock, but they add a little too. There is resolution loss across generations, particularly when making the final release prints from the dupe negative since those don't use pin-registered printers.

 

Moire should be minimal with film but I suppose there is a tiny bit of interaction between the grain structure and detail in the image. But you never know if that wide shot had some digital clean-up work done and went through a D.I. process.

 

Generally most digital projection is steadier and thus sharper than film print projection, but prints have a greater contrast ratio with deeper blacks, so that may be giving you the impression of more "nuance". I've always found that Sony's 4K projector, which uses some sort of liquid crystal technology, to have grayer blacks than DLP projectors, and definitely don't match the black levels of a good Vision Premier print, and you end up with a very sharp image that feels a bit flat compared to the print version.

[Mark Dunn]

 

Moire should be minimal with film but I suppose there is a tiny bit of interaction between the grain structure and detail in the image.

 

 

The grain is random from frame to frame, though, whereas the sensor sites stay put. So you'd that any moiré would only last for one unnoticeable frame.

[David Mullen ASC]

Yes, you're correct, I don't really know why he saw moire unless that shot was digitally manipulated.

[John E Clark]

Yes, you're correct, I don't really know why he saw moire unless that shot was digitally manipulated.

 

Would there ever be a case where during the process, Newton's rings would show up. Or were most prints 'optical' and hence not bring two surfaces in contact... or was a vacuum used for contact printing?

 

In my still printing, some glass negative holders, ostensibly to hold the negative 'flat' could cause Newton ring patterns if 'the right' conditions were met.

[David Mullen ASC]

But a Newton Ring would not repeat itself in the same part of a row of frames in a roll that was continuously contact printed