Debunking the color of Vaders Lenses for his ANH Helmet

[mverta]

That's only part of it, as film stock and many other factors contribute, but all of it can be compensated for, within a margin of error without absolute color reference (a chart), but still quite closely.

For my restoration, I determined the "neutral" state of the Tantive footage by analyzing the white first - there is a wide spectrum of white tones in the various plastic pieces - from blues to pinks to greens, often within the same frame. But I had some additional references for R2's blue, and some fleshtone references for various actors. By combining what I knew as likely targets for these references, I found a grading point which indeed satisfied all targets (again within a margin) consistently. That is, once applied, all similar features (specific panels on the walls, the floor, costume colors, specific actor fleshtones) all lined up consistently in all shots and with known references, which suggests that I was at, or very nearly at, the true coloration. It's a comparative analysis, but it can be very effective, and is done all the time.

The lenses in the helmet are transmissive objects - and whether their coloration comes from particulate matter suspended in the medium or some sort of external coating doesn't matter. We know a range of likely materials they were made from, and can determine attenuation of the material farily closely. The other part of their coloration comes from what's behind them (David's skin tone) which again, we can establish a fairly narrow spectrum for. It wouldn't be difficult to simulate these various factors digitally and allow a physical simulation of the light propagation to determine a likely range of potential colors for the lenses. Again, unless we hear some overriding physical phenomenon which would suddenly render this otherwise straightforward process invalid.

_Mike

 

[defstartrooper]

That's only part of it, as film stock and many other factors contribute, but all of it can be compensated for, within a margin of error without absolute color reference (a chart), but still quite closely.

For my restoration, I determined the "neutral" state of the Tantive footage by analyzing the white first - there is a wide spectrum of white tones in the various plastic pieces - from blues to pinks to greens, often within the same frame. But I had some additional references for R2's blue, and some fleshtone references for various actors. By combining what I knew as likely targets for these references, I found a grading point which indeed satisfied all targets (again within a margin) consistently. That is, once applied, all similar features (specific panels on the walls, the floor, costume colors, specific actor fleshtones) all lined up consistently in all shots and with known references, which suggests that I was at, or very nearly at, the true coloration. It's a comparative analysis, but it can be very effective, and is done all the time.

The lenses in the helmet are transmissive objects - and whether their coloration comes from particulate matter suspended in the medium or some sort of external coating doesn't matter. We know a range of likely materials they were made from, and can determine attenuation of the material farily closely. The other part of their coloration comes from what's behind them (David's skin tone) which again, we can establish a fairly narrow spectrum for. It wouldn't be difficult to simulate these various factors digitally and allow a physical simulation of the light propagation to determine a likely range of potential colors for the lenses. Again, unless we hear some overriding physical phenomenon which would suddenly render this otherwise straightforward process invalid.

_Mike

So the colour is ?

 

[mverta]

As I said, I'd have to simulate it to find out. Don't you people read posts all the way through? If there's sufficient interest, I'd be happy to. I happen to have the corridor modeled anyway, so the lighting's already done.

_Mike

 

[defstartrooper]

As I said, I'd have to simulate it to find out. Don't you people read posts all the way through? If there's sufficient interest, I'd be happy to. I happen to have the corridor modeled anyway, so the lighting's already done.

_Mike

Er yeah i did read it, i read it as you had already done so

For my restoration, I determined the "neutral" state of the Tantive footage by analyzing the white first - there is a wide spectrum of white tones in the various plastic pieces - from blues to pinks to greens, often within the same frame. But I had some additional references for R2's blue, and some fleshtone references for various actors. By combining what I knew as likely targets for these references, I found a grading point which indeed satisfied all targets (again within a margin) consistently. That is, once applied, all similar features (specific panels on the walls, the floor, costume colors, specific actor fleshtones) all lined up consistently in all shots and with known references, which suggests that I was at, or very nearly at, the true coloration. It's a comparative analysis, but it can be very effective, and is done all the time

Anyway please do whatever it is that needs doing it's all greek to me i'm afraid, i apologise if i didn't read correctly :thumbsup

 

[mverta]

My restoration only determines what the "final" color is - that is, the lenses normal color, plus the contribution of Prowse's skin coloring. We'd have to factor the skin tone contribution out to get a better idea of the actual lens coloring, so we'd have to run a simulation - it's do-able, but I didn't want to go forward if NHM had some knowledge of a physical material property or photographic anomaly which would render that invalid (though I can't think of one which would apply in this context).

_Mike

P.S. Just a cursory examination reveals a lot more information in the Red channel than Green and Blue, which are nearly equivalent, suggesting the lenses are in the red (ruby) family. Neutral gray, for example, shows equal power in the R,G, and B channels. To illustrate:

In fact, the grading adjustments required to actually bring the lenses' color into the amber/yellow spectrum are so great, the entire rest of the spectrum biases green to the point of everyone looking like Kermit the frog. Similarly, the amount of red amplitude reduction necessary to make the lenses gray (neutral) is likewise so severe as to render fleshtones black and white. Checking against other "colored lenses" shots ("We count 30 rebel ships, Lord Vader...") the lens tones exhibit not only the same higher Red channel power, but the SAME power relative to the other channels. So unless there is a severe photographic anomaly present in the registering of this very narrow spectrum of light being introduced by the filmstock (like you'd get with X-Ray film), it is almost impossible the lenses AREN'T red.

 

[Too Much Garlic]

I'm curious what photographic phenomenon would render this useless?

Taking it as it is and making conclusions from it, knowing that the image has been doctored. What you are doing is not what I was arguing about, but actually the opposite - you are trying to remove what color grading was done to the picture.

Would just like to know if you are using the SE versions or the regular versions to do these tests?

 

[mverta]

I have a whole host of sources - prints public and non-public, if you will. There is very little deviation at a core level on all prints prior to the SE, and post-SE, they are more or less uniform changes with the exception of a few region-controlled fx shots.

The color grading done at the time of the original prints and subsequent releases (timing, actually) was not easily capable of such specific region-based corrections. The lenses could not have been timed separately from the rest of the helmet without leaving serious matte lines and requiring multiple exposures, and none of this was done. The color timing present affects all elements in the frame uniformly, and is easy to compensate for. As I said, even if the neutral "de-grading" isn't 100% accurate to what reality is, the fact remains that is impossible to uniformly correct the lens colors to anything OTHER than red family without completely screwing up the rest of the spectrum, and since the lenses couldn't have been timed differently, then they're red. Conflicting accounts aside, we do have the physics of light propagation on our side here, as well as very quantifiable limitations of what was possible at the time.

So to recap - there is no "doctoring" of the image at any transfer stage. The original film timing - which remains more or less consistent across releases prior to the SE did not feature "spot coloring/grading" on the lenses - they were timed along with everything else in the frame. It is not possible to retime any release (SE or otherwise) such that the lenses are in anything other than the red family IN THE AFOREMENTIONED SHOTS. There are certain shots where the lenses are either underlit or obscured such that it can't be determined they're the same lenses. We are simply confirming that in shots where the transmitivity of the lenses is apparent, they are unquestionably exuding greater power in the red portion of the spectrum, mutually exclusive to presence as either neutral or yellow/amber.


_Mike

 

[Too Much Garlic]

Ah... thanks for the clarification.

We are simply confirming that in shots where the transmitivity of the lenses is apparent, they are unquestionably exuding greater power in the red portion of the spectrum, mutually exclusive to presence as either neutral or yellow/amber.

You're mentioning yellow/amber... how about brown material with Prowse's skin painted black around the eyes. How would something like that show in those tests?

Though, interesting that it's in the red... the question then is: what kind of red?

 

[Too Much Garlic]

I recently checked the lenses in photo reference to see if they were changed from the Tantive IV scenes to the Elstree photo and it appears they were not. But I noticed wrinkles on the inside of the left lens...and they became more noticable as time went on...corroborating the suggestion that there was possibly a tinted film of some kind inside the lens.

I'm not sure I'm reading this right. You're talking about the Elstree face mask next to the Indy whip?

 

[mverta]

Ah... thanks for the clarification.


You're mentioning yellow/amber... how about brown material with Prowse's skin painted black around the eyes. How would something like that show in those tests?

Though, interesting that it's in the red... the question then is: what kind of red?

Transmitivity of a material means how much light it reflects versus how much passes through it - in a sense, transparency. Think about a black surface behind completely transparent glass: it will appear black. Any particles suspended in the glass, or coatings on the glass will reduce its transmitivity, making the black behind the glass even darker. Imagine David's white eyes at the center of that black now. All we would ever see, really, even with the completely transparent glass would be the whites of his eyes. Now, with every bit of reduction in transmitivity, like if the lenses were any shade of gray, his eye whites would get darker and darker, but still, the black around his eyes would render the surface of his skin black.

But this isn't what we see in these shots. We can make out the surface of his skin. Since the lenses certainly aren't clear, then the skin certainly isn't black or we'd not be able to see the skin at all. In fact, to be visible at all, with this level of transmitivity in the lenses, it is highly unlikely he has any make-up on at all. Especially if you factor out the red tint from the lenses and increase luminance to essentially "dial out" the lenses all together, you see a natural exposure for eyes/skin.

Brown is really in the yellow family, just darker. So if they were brown lenses, we'd still see higher power in both Red and Green channels, relative to Blue. But again, we're seeing nearly identical power in the Green and Blue channels, with an increase in Red power, so it's in the red family.

As for what kind of red, it's harder to determine, but we can get within a decent range. One of the first things we need to factor is the transmitivity of the material they're made up of, so we know how much coloring they're actually adding to the mix versus his skin tone. I can tell you from experience with this stuff they're fairly saturated - they're not fighting his skin tones much at all. But if we knew what material they were made of, I can plug that in. In the meantime, by simulating the light propagation of the scene, we can derive a fairly accurate coloring for the lenses.


One more illustration:

_Mike

 

[gonk27]

Now, with every bit of reduction in transmitivity, like if the lenses were any shade of gray, his eye whites would get darker and darker, but still, the black around his eyes would render the surface of his skin black.

But this isn't what we see in these shots. We can make out the surface of his skin. Since the lenses certainly aren't clear, then the skin certainly isn't black or we'd not be able to see the skin at all. In fact, to be visible at all, with this level of transmitivity in the lenses, it is highly unlikely he has any make-up on at all. Especially if you factor out the red tint from the lenses and increase luminance to essentially "dial out" the lenses all together, you see a natural exposure for eyes/skin.

Interesting stuff. Prowse's face certainly did not have black make-up applied as a matter of course through filming, there are enough photos out there showing him in costume without the helmet proving that.

Mike, what would your analysis say using a shot from the Death Star control room as Vader turns into the light on the line "her resistance to the mind probe is considerable.." it looks pretty orange at that point, although I think that's the only time they do.

Jeremy

 

[Westies14]

haha I'm sorry to go off-topic again for a moment, but I have to say I love when Mike posts about his color-correcting of Star Wars. I do color correcting on photographs and video as a very small aspect of my work, but the degrees to which he takes it and the color sensitivity and knowledge he exhibits is baffling to me. Mike, I know you never intend to distribute your corrected version of Star Wars, but I sincerely hope that fate leads me to a viewing of it somehow one day after it's completed. I can't even imagine how good it's going to look!

 

[propcollector]

ok here is that pic

 

[rickrickrick]

OK, I am not an expert on Vader or color correction or transmissiveness or whatever. I can only speak of my own experience.

The much-maligned Rubies Supreme Vader, which I am guilty of owning and wearing on Halloween, has thin grey tinted lenses. Yet in flash pictures, you can easily see the reddish tone of my pasty white guy skin beneath. It LOOKS like red lenses. (See attached.)

On film sets, the lighting is generally very high key - especially for older films when they used slower speed film stock. And Vader's lenses don't always look reddish, only when light is shining directly into them.

So isn't it possible that they were fairly THIN grey lenses and the reddish tone we're seeing is always Prowse's skin? And that for the sequels they had thicker or darker grey lenses and/or were more careful about lighting them?

Just asking...

Rick

 

[Jumpin Jax]

That shot is looking you right in the eye. Vader is practically in profile most of the time, so it seems unlikely.

 

[Too Much Garlic]

Rick, if you could take more pictures from different angles and perhaps a little larger that could be interesting.

 

[artoo77]

From experience with the actual masks for personal appearances and filmed spots (and I don't want to get into a debate regarding the differences in helmets from film to film...that just exhausts me reading about it...), the lenses are smoked grey plex but lean a bit to the brown/amber range. When doing close-ups with the helmet, we would take gel (Neutral Density) and tape them in to the lens so that we don't see the performer's eyes. For wider shots, we'd remove them for better visibility. On one occasion, we did blacken the actors eyes (with make-up, not a swift right hook).

I always wondered about the "red" look, and from what I personally witnessed, it seems to me to be the tint of the lens and Prowse's face--I think you can actual see his eyes in some shots. To the best of my recollection, all the lenses were more or less the same color. I doubt that they would have introduced any color purposely, as the art direction and thought into the first film was not that sophisticated--and I don't mean that to be a slam to the production people...it was just their approach to a "B" movie. (As an example, they didn't introduce the two-tone paint job, the sweeting of the way the cape lays over the shoulder pads, etc. until ESB). I'm sure the color they used was what was laying around, and the intention was for it to appear to be all black. Any photographic anomalies were discovered too late for ANH--hence the two-tone paint job for ESB, which I'm sure was added to make the mask photograph better then an all black paint job.

If you want this mystery solved, I'd let my esteemed colleague Mr. Verta resolve it, as his skills, knowledge and experience, particularly in this arena, are unmatched. You'd be a fool to doubt his conclusions.

--Don (mverta's brown-nosing agent) Bies

 

[defstartrooper]

Any photographic anomalies were discovered too late for ANH--hence the two-tone paint job for ESB,

The ANH mask was two tone though :confused

 

[Too Much Garlic]

Any photographic anomalies were discovered too late for ANH--hence the two-tone paint job for ESB, which I'm sure was added to make the mask photograph better then an all black paint job.

The two-tone paint scheme was in ANH as well.



EDIT: Oh... got beat to it.

 

[artoo77]

Hmmm...

Oh well, I'll stick to R2.

The ANH mask was two tone though :confused