Info: The process of actual FILM editing - negatives, interpositives etc.

Reading a couple of restoration based threads, I noticed that I probably STILL don’t 100% understand the process. How EXACTLY do you get from shooting a scene to a finished film print? This is something what I “assume”

1. Original camera negatives - basically the film which is inside the camera while shooting the scene
2. Interpositive - basically a negative of the negative from 1)

Now here goes my question - what is used for editing? Are edits done directly on the negatives or on the IPs? I googled a lot and some sources use the term “negative cutter” which would IMHO mean the cuts are done on the negatives itself, however some sources say that the cutting is done on a positive. What is the truth?

3. Internegatives - so if I got it correct, this is again a negative of version 2), correct
4. Positive - release print made out of 3)

Is that correct? In which point is the actual editing being done? I assume it’s either after step 1 or step 2, correct?

And now let’s go even more complicated - what about shots which include optical effects and/or composites? Since this is a Star Wars site, let’s say a Star Wars example - the shot where our heroes look through the window on the cockpit of the Millenium Falcon and see the flying fighter and the Death Star. I remember that this specific scene was explained in some of the documentaries, not sure which one, but from what I remember they combined them using the optical printer, in other words magic. But what did go inside the optical printer? The original negative? The internegative? The interpositive? The “answer positive”? In other words - how many generations away from the negative can we get in such a complex shot as the one I mentioned from Star Wars 77?

And another question 😃 From what I though, original negatives are the film reels which came out of the camera equipment, correct? But many people all over the internet (I think even here on ot.com) use the term to describe a finished shot with integrated optical effects, or is it just me?

And the last question, I promise - what would be the correct term to describe the finished film, properly edited, with all effects and the finished sound track, which would be the “master” for creating duplicates? At least I hope that they’re just doing one “master” from which are theatrical release prints being made.

Sorry for so many questions all at once, I think I have a vague understanding of this topic but I’d really love to be 100% sure I correctly understand what are other people talking about 😃

Thanks

Fascinating stuff, thanks a lot.
So now a quick summary, if I understood it all correctly. Let’s do it on an example, e.g. Star Wars.

It has scenes of 3 types :

I) scenes consisting only of live action shots
II) scenes containing some simple visual effects like fade-ins, fade-outs, dissolves etc.
III) scenes containing composite shots created using live action shots and visual effects

If I was the Star Wars editor, I would do …

I) these scenes would be done by cutting and joining the original camera negative
II) these scenes would be created using negative A and B rolls
III) these scenes would be a combination of edited negatives and various optical elements joined together using an optical printer

So we would have properly edited scenes of 3 types, and they would still be negatives - correct? Then we would join these scenes together, and interpositives would be made? Then from these interpositives new internegatives would be printed, and these IN would be the source for theatrical prints?

Correct or complete bull…?

Regarding whether the negative or a printed positive is used in the optical printer, the original “Star Wars” used black-and-white YCM separations-- also worth noting the YCMs were made on Estar base instead of standard acetate to maximize stability. “Empire” used YCMs as well, but accompanied by a substantially enhanced optical printer with, if I’m interpreting the article correctly, a new lens that finished the composite directly to 4-perf anamorphic rather than on VistaVision where it would then need to be reduction-printed.

The printing tech changed substantially less on “Jedi”, but they made crude black-and-white composites of every VFX shot first (“Empire” only did this for problem shots), which from an editorial standpoint was extremely helpful for the sound team as they could work on these shots before they were finished.

Info comes from https://www.theasc.com/magazine/starwars/ (I’d have preferred direct links but this page doesn’t let me. The articles sourced are Episode IV “Composite and Photographic Optical Effects”, Episode V “Composite Optical Photography”, and Episode VI “Optical Effects”)

Here’s a question: were the original '77 release prints of “Star Wars” made from internegatives? I do wonder how the OCN could be in such notoriously deplorable condition if the numerous prints weren’t made from it (not including the color fading or '97 recuts).

Hm, that’s actually a good question. I remember that in some of the 1997 promo “documentaries” they stated that the original negatives were in such a poor condition BECAUSE they were brought from the archives so many times…

Multiple INs would be a given for such a popular film (and I do recall “The Godfather” OCN becoming so worn that the normally photochemical-only Robert A. Harris didn’t want to risk running it through a gear-based mechanism again), but “Empire” and “Jedi” (and “The Godfather Part II” for that matter) weren’t known for having their negatives in such terrible condition… and later on, the extremely popular “Titanic” only needed new INs made. So I guess a couple more questions:

Is there a rough idea for when the OCN-IP-IN-Print order started? And in the case of an optically blown-up Super 35 film, were the anamorphic internegatives straight from the S35 IP or were two more steps added like OCN-IP-Blowup-Positive-IN-Print?

EJones216 said:
“Empire” used YCMs as well, but accompanied by a substantially enhanced optical printer with, if I’m interpreting the article correctly, a new lens that finished the composite directly to 4-perf anamorphic rather than on VistaVision where it would then need to be reduction-printed.

Info comes from https://www.theasc.com/magazine/starwars

Looked at another article (Episode IV "Miniature and Mechanical Special Effects, Page 4) and realized I indeed misinterpreted this part of the “Empire” article-- the original’s opticals were also directly squeezed to 4-perf anamorphic during the composite, which is worded by its writer (John Dykstra) as making them “one dupe generation from the original photography”.

In summary the print chain could be described Negative-YCM-OCN.

This is quite funny and informative at the same time : https://www.youtube.com/watch?v=vq1DE_W86XE

6 month later and I’m back with another question, this time about optical compositing 😃
EDIT - OK it’s better to have a special thread for it : http://originaltrilogy.com/topic/Optical-compositing/id/53617

As someone who actually works in film editing and has actually worked on projects that were not cut digitally, I will explain.

First of all, the first post that shows A/B rolls. That is only 16mm. Almost never used professionally for material shot on 35mm except in rare circumstances.

When they shoot a film, what is in the camera is the negative. Also called “Original Camera Negative” or OCN for short. Did you ever take still photos on a 35mm camera and when you get them back from the drug store there are “negatives” with your print? This is the pretty much the exact same thing that comes out of the 35mm film Panavision camera. The color values are reversed and the stock has an orange base.

What Richard, Marcia, Paul, Sean, and Duwayne actually “cut” on Moviolas and flatbeds are what is called a “one light” workprint. Meaning a color timer looks at the negative and strikes a print from it. He takes a best guess at what it should look like. Sometimes there are notes on the camera reports from the DP.

The editors cut away on these “one light dailies.” They try things: trimming, extending, swapping out takes. Over the course of the editing process, it gets pretty beat up. Occasionally the film will get massively scratched or shredded. A new print is struck of that shot (called a re-print).

Once the film is finished and they decide they aren’t going to change the picture cut anymore, the film editor’s “workprint” goes to the negative cutter. He takes all the OCN and carefully matches it to the reels they cut from the “one light dailies.” This is a meticulous process. Negatives are always handled very carefully. The “cuts” the picture editors make are done with scotch tape. The negative cutter does “hot splices.” which basically fuses the two pieces of film together. Obviously if the editor cut a scene, and then decided to extend a shot (which happens a lot) the workprint will have a splice in the middle of a shot. The negative cutter will not replicate that cut. He matches the cut perfectly because all film negative has “key numbers” on the edge of the film. He matches everything by numbers.

Once you have the master negative reels, this is where IPs and INs come into play. Not really before. From the cut OCN reels, several Interpositives are made. This is basically making a positive print (proper color values) but it is on orange negative stock. From each of those IP’s several Internegatives (INs) are made. INs are basically copies of the Cut OCN master reels, but they have the final color timing baked in. And there are no splices. From each of them several “release prints” are made. Ultimately thousands of them. How many IPs and INs are struck depends on the size of the release. A studio usually tried to get at least 200 “Release Prints” out of a single IN. For large “premiere” venues in major cities they may strike what are called “EK Prints” directly from the Cut negative. “EK” stands for “Eastman Kodak.”

Now in terms of VFX, that is another can of worms. When you cut VFX movies, typically you want to get VFX heavy scenes done first. It is very expensive to have to do VFX over. Nor do you want to shoot a 10 second shot and then scrap half of it because a scene got trimmed. That’s money wasted. And the VFX guys need as much time as possible to do their work. That still applies today with CGI VFX. But in the photochemical world, let’s say you have as shot with two elements. You will strike IPs for both those shots. The IPs will only be the section used in the cut with some “handle.” In film it would be 1-5 feet. (3 feet = 2 seconds).

These will be run through an optical printer. Which is basically a camera mounted against a projector. The two elements run through the projector side optical printer - one at a time. One one piece of new unexposed 35mm negative will be run two times through the camera side. This will create a new piece of negative with new key numbers which is often called “Dupe Neg” or “Optical Neg.” And like a xerox copy or a dub of a cassette, there will some quality loss. That is why VFX elements are often shot on large format film. It mitigates quality loss. ILM used “VistaVision.” When Richard Edlund left and started Boss Films, they did everything on “65mm” film.

A one light print is struck from this new negative. And assuming everything went right, it goes to the cutting room and they “eye match” it into the workprint. Basically matching the action by looking at individual frames. I massively simplified the process of photochemical VFX.

And almost everything I have described - is a relic of the past. Very few movies are shot on film. And almost none are finished on film anymore. But this is how it was done on the original trilogy.

I realize this thread is several years old. But hopefully someone is out there listening…

Wonderful, thank you very much. I almost forgot I started this thread 6 years ago, and you decide to revive it on my birthday 😃 Great information, thank you. Could you maybe be go to more details with the optical printer part? I got information from several different websites, and they usually say mutually exclusive things. I assume different companies have used different processes or different processes were used in different eras?

Let’s examine a case I mentioned here 6 years ago, feel free to correct me (PLEASE do 😃 )
It’s 1976 and we want to film a shot in which we see our heroes in the cockpit of the Millennium Falcon watching outside, where they see space, the Death Star and the Tie Fighter.

So we need these elements

• shot of the Falcon interior, in front of a blue screen

• shot of the star field

• shot of the Death Star, assuming again in front of a blue screen

• shot of the Tie Fighter in front of a blue screen (I assume)

• correct?
  Now we need 2 mattes for every one of these elements, correct? How do we get them? I read something about rephotographing the bluescreen footage over a red filter onto a high contrast black and white film, and making negative copies of this again on high contrast black and white film, until we get 2 mattes - one has white parts where the original object was and black where the blue screen was, and the other is the negative of that, meaning it has black over the original object (e.g. ship) and everything else is white. Is that correct? Also should they be “white” or “transparent”? Or does white color become transparent during the compositing?

You will have to forgive me. I work in Film Editing, not VFX. Also, as I mentioned, none of these techniques are used anymore. CGI was already around when I started in the business. Digital Intermediates started replacing cut negative and photochemical Opticals about 6 years after I started.

Hi-Con film is a type of B/W where there is no Gray. Stuff comes out either as completely Black or as completely White. They use this film when shooting title cards as well.

Blue was used because of the three color layers (Red, Blue, and Green) on film it had the finest grain structure. Green is used for CGI because it has the highest luminance and can be the easiest for computers to see.

Have you ever seen an Anaglyph 3D image or movie? If you look at the raw image, there is a blue and/or red off set. When you look at it through glasses where one eye is blue and the other is red, The eye with the blue lens does not see the blue offset. The other does. And same for the red offset in the other eye. Your brain then fuses them together.

It is basically the same idea here. If you use color filters when running a Blue Screen element through the optical onto Hi-Con film, you can get it so the camera only sees the Blue. The light shines through anything that isn’t bright blue and hits the negative turning the silver emulsion black. Anything that was blue is masked. No light hits the negative and it stays white. Because it is Hi-Con film, anything gets even a little bit of light turns to pure black.

Now that you have a hold out matte where the actors are black and the blue screen is white, that can be run through an optical printer to make the reverse matte where the actors are white and the blue screen is black.

The one where the actors are black will be used when running their element through the optical printer. The one there the background is black will be used when running the death star element. Same would go for the the Tie Fighter element (with its own matte). You use the mattes to prevent double imaging. Otherwise you would see a little bit of the background superimposed over the foreground elements.

Blue screen and B/W mattes are still used in CGI. But created in a totally different way.

Nice, thanks. Yes, I know that most of these techniques are a thing of the past, that’s one of the reasons I am interested in them. It’s all turning into lost art, so the least we can do is to preserve the knowledge about how things were done back in the analogue ages, hopefully it makes sense to anybody other than me.