It matters because Hawkins' claims are anachronistic and simplistic.
First, "chromakey" (chrominance keying) is a television-only process. It is an electronic means of on-the-fly compositing. It cannot be used for film. Keying in television has been around for a very long time, but originally in luminance form. You can develop even analogue equipment that drops out signals below a certain threshold and replaces them with a background plate. These days chrominance keying is a fine art that does everything from the standard superimposition of the weatherman over his maps to the indication of first-down lines in televised football games.
So Hawkins' reference to "chromakey" is the first indication that he doesn't know what he's talking about. His "proof" photo clearly depicts motion picture film cameras, not television cameras. His "evidence" and his claims are mismatched.
Second, he specifically mentions blue and green screens. This is anachronistic, as someone in the mid to late 1960s wouldn't yet be using the blue-screen method and definitely not the green-screen method. The blue-screen method works because it corresponds to a primary element in the film, and also because it is the perfect complement to skin tones. This gives better reproduction of skin tones in the foreground elements, where the (missing) blue information is approximated by printing the green element on black-and-white film and then printing through a blue filter. The downfall of this method is the number of generations required, leading to a loss of contrast. Eastman developed the 5497 (I think that's the right number) emulsion as a high-contrast film for photo compositing. It was sold as consumer Kodak VR Gold film.
The advantage of the sodium screen process (e.g., Mary Poppins) was that the holdout matte was generated in the camera. Fewer generations in the process leads to preserving grades and contrast and a higher-quality picture. But this is done by using the old three-strip cameras (i.e., the first Technicolor cameras) and running a special emulsion on the third strip. The entire scene was evenly illuminated by bright sodium lights that emitted only in a very narrow spectrum. If you replace the filter on the third strip with a narrow-pass filter tuned to the sodium wavelength, and you film your actors against a white screen lit by that light, you get a crisp holdout matte on the third strip. The first strip is regular color film with the foreground against a white background.
So to composite, you print the background plate with the holdout matte generated from the sodium-sensitive stock. Then you rewind and print the foreground, which is self-matting. That's why the opticals in Mary Poppins hold up very well even 40 years later. You get no generation penalty over normal printing. The blue-screen matte-extraction process is extremely heinous.
Green is only useful for digital compositing, chosen because digital imaging systems are usually more sensitive in the green band and because green paint is more reflective. This has practical advantages for filming the foreground elements because you can get by with less and sloppier light on the screen. My compositing system allows me to choose any hue I want for the key color, but green works best.
Black-screen and white-screen compositing work on film if your foreground and background densities are such that rudimentary photochemical luminance keying will work. I.e., you create the mattes by overexposing or underexposing your foreground in the duplication process beyond the latitude of the print film. If I photograph you against a perfectly black screen, I can overexpose the dupes to my heart's content without ever getting anything from the background.
So I overexpose your face in order to get the silhouette, and that can be printed on reversal film to generate the burn-in and holdout mattes. This has the advantage of preserving all the chroma information, but has a disadvantage in that high contrast in the foreground may not be fully obliterated by overexposure, requiring garbage mattes and rotoscoping. So you can go to white-screen photography, where the background saturates the film and you can underexpose the foreground to get a natural holdout matte. Conversely, highlights are a problem in the foreground.
Douglas Trumbull made a limited use of this process in Kubrick's 2001, but preferred in-camera compositing. "Star Trek" the TV show also used black-screen compositing.
Blue-screening was available in 1969, but it was not widely used because of its complexity and generally low quality. It wasn't really used seriously until 1977 for Star Wars, and from then was widely used until the 1990s. Finally, the blue-screen process requires a particular wavelength of blue reflecting at a certain luminosity, so that the density in the blue layer of the film is carefully controlled. The colors in the photo are just wrong.
Hawkins apparently has no real knowledge of the techniques widely used in the 1960s for film compositing. He is attempting to project modern techniques and related (but inapplicable) technologies onto his theory. This has led him to fabricate the wrong kind of "evidence" for his theory. In other words, his photo and explanation show what would be done if the footage were faked in 1999, but is utterly wrong for a composite shot of 1969.
|