Motion Picture Films

[edgar_njari]

Here is an example of the loss in optical copying process in motion pictures.

 

This is a comparison test at 4K resolution (that's more than 4200dpi) between an original 5245 negative and an IP copy of the same frame.

 

This scanner is not as good as the first example I posted. This was a

cheaper Cintel scan, while the first one is a high-end Kodak cineon scan.

But it is good enough to tell the difference.

 

Notice how the IP is almost grainless at 4200+dpi.

That's how the new Kodak intermediate stock it, it is very smooth at large magnifications with no grain to speak of, but it does have some of that pepper grain, which is not visible in this case though.

But the copying blurred the negative grain a bit.

 

You can clearly see how much sharpness is lost in this first generation IP copy. IP's are done under almost perfect copying conditions, unlike release prints.

Now imagine getting one more such pass, for IN, and then passing it through a high-speed printing process. The loss in resolution and sharpness is huge.

 

http://free-os.t-com.hr/redmist/4Ktest.jpg

 

Once again, the 50D stock is performing just as well as a still stock grain wise at such high scanning resolution

[edgar_njari]

Just noticed, it's not the same frame, but its from the same sequence. I wish this forum had editing, so I didn't have to post so much

[francis_t]

Not to go off on a tangent here, but in regards to the comments about release prints being 4th gen copies, aren't most of the newer RPs being made from digital internegs. So shouldn't this reduce the loss of quality compared to the older optical method?

[edgar_njari]

Its not really standard practice, though more and more films get a digital intermediate process every year.

 

The answer to your question is yes, and no.

Yes, because, as you assumed, there is less copying going on if you don't count scanning and recording (though you should really because it not perfect). So the image is one step less faded, has one step more shadow and highlights detail etc.

 

And NO, because DI process has its own problems.

Usually a recorded positive doesn't really look like the original. And doesn't even look like an optical copy either. It has a look of its own.

The tones across the scale look a bit distorted.

Best example I can think of is Hitchhikers guide to the galaxy.

That specific look is even visible on the DVD. Because the DVD is transfered from recorded film, not the original negatives.

Without going into fancy words, you can tell its a DI, the blacks are usually a telltale sign, you have to observe it to see what I'm talking about.

And of course most DI jobs are done at 2K resolution. So the recorded film doesn't hold much information from the original frame.

So you have yourself a slightly blurry "original" to go from, which when copied further, sort of negates the advantage you had by skipping one step.

 

It works best with cartoons, animation and HD movies, in which case there are no disadvantages. But it has mixed advantages/disadvantages when you are working with film as your original material.

Its a mixed blessing.

 

It is a great tool of expression if you want some specific look or something, but its not worth it if you just want to skip a step in printing. After 2K processing, the resulting print is usually reported to be even softer than a good 4 generation optical. Sometimes it is so, sometimes not, depending on the skill of the people preparing those materials.

 

A 4K DI job would pretty much solve the problem, and show clear advantage over optical printing, as many tests have shows so far, but nobody cares enough yet to do it.

Well some films like Stuart Little 2 did it all in 4K, and some other features.

[rowland_mowrey]

Motion picture negative film is built to have a contrast of about 0.5 while still film is built to a contrast of about 0.6. This difference allows multiple generations to be printed with good latitude. It also makes printing on conventional color papers to be a bit of a problem.

 

The print film for motion picture has a contrast of over 2.5 and a dmax of nearly 4 or 5, so that a print from these will have a contrast near that of a normal slide but a higher dmax and longer tone scale.

 

The grain size of the negative original is over 2 microns, and the grain size of the print film is about 0.2 microns. Both films have heavy interimage and correction along with edge effets. Both films are very thin.

 

The result is very good overall sharpness and grain. I have a single frame from a Star Wars print and it is remarkably sharp and grain free.

 

The Name of the game is to have a picture look sharp and grain free at normal viewing distances. Do you walk up to a 16x20 print from a 35mm negative? You can, but you normally view it at some distance.

 

The motion picture print fulfills its goal and makes a transparency that is far better than can be achieved by an E6 type film. This is due to the long latitude, high dmax and color correction of both films involved in the neg-pos system.

 

Ron Mowrey

[edgar_njari]

Ron

 

That SW frame you have, is not a very good example of motion pictures.

I assume the one you are talking about is Episode 1, from Mos Espa, right?

Anything coming from George Lucas is not a good example of imagery.

That frame has been butchered by low res scanning recording, and then on top of that printed from a thin internegative.

Those people working on Episode 1 have done such a poor job, that it's almost as if they advertised how crappy film can be before switching to video shooting on episode2.

 

If it looks good, great, but, my advice is get another frame to use as an example, one that trully represents the potential of a modern Kodak movie stock.

 

Considering that the plate for that shot was done on Eastman 5245, which is supose to be colorful and punchy, that frame ended up looking subdued in color and had sickly-looking contrast.