Well, it looks like we have an audience. Who would have thought that a half maroon TIE fighter would go viral? Yeah, it's out there in multiple places. Gotta love the George-bashing comments (not)! Look, when you're trying to figure out how to do what seemingly is impossible, you try some wild ideas and see where they lead. I'll try to get some additional info about the "what the hell were they thinking" aspect of this and get back to you. Meanwhile:
... looks like they've got some yellow panels on the ball there near the window.............
Good catch. Didn't notice it. Some of the wedges are damaged/smudged and that affects the color. But I looked at the two-toned TIE wedge again and scanned a different strip. Yep, there is definitely some form of paneling going on there -
Full res, cropped -
A different frame -
So it's definitely on the model, not the film.
As an aside, these wedges were shot on a filmstock called 5247. While the original camera negative holds up well if treated properly, the corresponding workprint stock does not. It fades rather quickly (within about a year) with the cyan portion of the record going first. This is what the frame above looked like before any color correction -
Talk about purple. Sheesh. Thankfully, the scanner can correct most of that. Yeah, you're relying on an algorithm to correct the color and it may not be perfect. But relatively speaking, it looks good to the eye.
Anyways, after the workprint comes back from the lab, the strips get cut up and taped to a sheet of acetate -
When viewed on a light table, you can see the exposure changes at a glance because you can see multiple frames at once. Like I said before, information is usually scribed on the film as a reminder of how the wedge was shot.
All of these seem to be exposure wedges as they (or portions like the engines) just seem to be getting lighter or darker.
That's wedging 101.
Gene