# Stanley Kubrick used a NASA-inspired lens to film by candlelight in Barry Lyndon, but what did NASA use it for?

At 07:35 in the CinemaTyler video How Kubrick Achieved the Beautiful Cinematography of Barry Lyndon he says:

As far back as 2001: A Space Odyssey, Kubrick and Alcott had been talking about the idea of shooting night interiors exclusively by candlelight. Kubrick had wanted to shot by candlelight for a film on Napoleon he was researching.

At the time, there wasn’t a lens fast enough to get a decent exposure in such low lighting conditions. The lens they ended up finding for Barry Lyndon was a Zeiss f/0.7 50 mm lens that was developed for NASA to take pictures of the dark side of the Moon (American Cinematographer). I spoke about this lens in another video that I have linked to in the description.

That video is The Kubrick Files Ep. 3 - Kubrick's Cameras. At 07:58 he says:

Perhaps the most famous of Kubrick’s lenses is the Zeiss Planar 50 mm f/0.7 lens by Carl Zeiss.

This lens was used on the interior scenes for Barry Lyndon. And in some scenes the interiors were only lit with candle light. This would have been impossible to capture on motion picture film as there didn’t exist a motion picture lens to capture a proper exposure in such low lighting conditions.

Kubrick, delightfully stubborn as he was, decided to adapt a lens that had only been used by NASA at that point… Kubrick had eh lens “redesigned by Cinema Products” so that it would work with his “Mitchell BNC 35 mm camera” (Kubrick Exhibit).

From Lenses at the Kubrick Exhibit in San Francisco 23 June 2017:

[Left] Zeiss Planar 50mm F0.7 lens by Carl Zeiss

In order to shoot the interior scenes for Barry Lyndon by candle light, Stanley Kubrick had a special lens by Zeiss redesigned and adapted. Initially, such Zeiss f/0.7 lenses had been used during NASA space flights. With a maximum aperture of 0.7 (indicating the relation between the focal length and the diameter of the maximum aperture) it was about two stops faster than the available high speed lenses at that time. This made shooting by candlelight without additional lighting fixtures possible. The lens was mechanically redesigned by Cinema Products, USA, in order to fit Kubrick’s own Mitchell BNC 35mm camera.

• have you tried entering a few search terms into a search engine? The first search result for 'Zeiss f/0.7 NASA' answers your question. – Hobbes Aug 7 '19 at 8:53
• @Hobbes does it? – uhoh Aug 7 '19 at 9:41
• – Hobbes Aug 7 '19 at 9:48
• @uhoh: that Wikipedia article has sources which don't lead to NASA: I think it's more-or-less junk. – user21103 Aug 7 '19 at 14:13
• Not quite enough at this site for an answer, I think, and also I had to rely on machine translation out of Italian, but it's still very interesting, appears to be well-researched, and proposes that "After this experience [Ranger] in mapping the dark areas of the Moon performed with the Angenieux f / 1.0, it is possible that the NASA technicians have found this brightness still insufficient, and in anticipation of the famous Apollo project ...have commissioned ... an even brighter optic , namely the Planar 50mm f / 0.7. " -- from marcocavina.com/omaggio_a_kubrick.htm – Roger Aug 7 '19 at 17:51

This isn't a complete answer, but I think there should be at least some doubt cast on this story: it's certainly not as clear as a lot of people think it is.

The Wikipedia entry for this lens claims that it was developed in 1966.

So we can wonder what it might have been used for. It is generally claimed that it was used (or designed) for pictures of the Moon's night side, and that makes sense to me. If so, it was not going to be useful from LEO, because it's way, way too short. So if indeed it was used for lunar night side photography it must have been used from Lunar orbit (see below for why not on the surface).

The first candidate is the Lunar Orbiter programme, which surveyed the Moon in 1966 & 1967. But:

The camera used two lenses to simultaneously expose a wide-angle and a high-resolution image on the same film. The wide-angle, medium resolution mode used an 80 mm F 2.8 Xenotar lens manufactured by Schneider Kreuznach of West Germany. The high-resolution mode used a 610 mm F 5.6 Panoramic lens manufactured by the Pacific Optical Company.

(From the above Wikipedia page.). So, not Lunar Orbiter. (Incidentally, if you don't know about Lunar Orbiter: it was an amazing thing which processed film in space!)

That leaves, I think, Apollo: and in particular Apollos 8, 10 & 11-17.

This seems to be a good reference for the cameras and lenses used during the Apollo program and I have not been able to find any evidence that this lens was used. If it was used it would have been in one of the survey systems from orbit I think, because there's just no purpose to a lens that fast on the surface (focussing an f/0.7 lens while wearing a spacesuit would be ... interesting, not to mention that they did not land during the lunar night). But some of the links to details of the survey stuff are broken. It's kind of a short lens to be using from orbit though, unless they wanted pretty broad surveys (which perhaps they did).

Another possibility is that it was designed for pictures of Earth's night side. That's possible, and would allow a much broader range of missions (in particular pretty much all manned missions in era, and many unmanned ones as well). I have not investigated this option.

I can find no good evidence of it from any kind of NASA site.

So my best guess is: NASA commissioned this lens but never used it. But that's just a guess. What is clear is that evidence for the general consensus that it was developed for (and perhaps used by) Apollo is at least sparse.

I would be delighted to be shown to be wrong based on non-apocryphal evidence!

• I see what you mean. The night side of the Moon would be "photographable" from lunar orbit only when lit by Earthshine, otherwise it would only be illuminated by starlight or the planets and therefore too dark. Only around new moon would you have half illuminated by the Sun and half by the Earth, it seems like it would be a real niche application. The only possibility I can think of is if the target were not the Moon's surface, but some kind of dust cloud or electrostatically levitated particulate halo that would be very dim against the black of space when illuminated by the Sun. – uhoh Aug 7 '19 at 9:49
• I've added a long comment – uhoh Aug 7 '19 at 23:48

This is too long for a comment.

tl;dr: I can't conclude that the lens wasn't used (as @tfb's excellent answer doesn't yet conclude either) but it does seem reasonable that NASA would have gotten a hold of several of the fastest lenses in the world just in case they might have come in handy.

## Rangers VII, VIII and IX

Ranger is mentioned in @Roger's comment

From the various documents available at https://www.lpi.usra.edu/resources/ranger/ the Camera "A" series used f/1.0 lenses and exposure times of 5 ms (1/200 sec) and 2 ms (1/500 sec). The image sensor was a special vidicon tube that retained the charge image for at least several seconds to allow for a slow-scan read out direct to broadcast to Earth via a low bandwidth transmission, where it was recored directly to film.

The short exposure times were needed because the Rangers were on an impact trajectory and needed to record at very close distances.

There was no attempt to photograph the "dark side" or night side of the Moon using Earthshine for illumination.

One-inch-diameter vidicons are used for image sensing. Electromagnetically driven slit-type shutters expose the vidicons. The image is focused on the vidicon target through the shutter, which is placed slightly in front of the focal plane. The vidicon target is made up of a layer of photoconductive material, initially charged by scanning with an electron beam. The image formed on the photoconductive surface causes variations in resistance across the surface which are a function of the image brightness. These variations allow a redistribution of the charge which remains after exposure. In the Ranger cameras, the charge pattern formed by the image on the photoconductor remains much longer than in commercial systems, so that the pictures may be taken more slowly. By slowing down the picture-taking rate, it is possible to use a narrow electrical bandwidth, which simplifies the communications problem in transmission of the signal to Earth. After the image has been formed on the photoconductor by operation of the shutter, an electron beam scans the surface and recharges the photoconductor. The variation in charge current is the video signal, which is then amplified several thousand times and sent to the transmitter, where the amplitude variations are converted to frequency variations. The frequency-modulated signal is amplified, and the signals from the two channels are combined and transmitted to Earth through the spacecraft high-gain antenna.

## Lunar Orbiter

As discussed in @tfb's excellent answer the Lunar Orbiters did not require absurdly fast lenses, most likely because they were in nice orbits around the Moon rather than being in the process of crashing into it.

## Apollo Missions

Orbits by Apollo spacecraft provided two big advantages over the automated Ranger and Orbiter programs;

1. Cameras could be operated by astronauts who could discuss with scientists on the ground if necessary
2. Film was returned to Earth physically, so no need to develop it in space or transmit via low-bandwidth connections.

Using the same data and Python script used in this answer for at least Apollo 12 and 14 it looks like the angle between the Sun and Earth was about 50 degrees, which means only 50 of 360 degrees of longitude would be night time but illuminated by Earthshine.

Bringing a special camera lens to photograph it and holding the camera steady enough to take scientifically meaningful photographs illuminated by Earthshine seems like an idea that would be worth thinking about and possibly ordering some lenses, but probably not following through with.

Its story is a fascinating read, in the form of a longist Italian article. Translated by Google here.

The plot is full of twists:

• Core based on the 1874 computation of the double-Gauss-type optical cell
• Ideas drafted pre-WW-II in 1928, and again in 1937 by Kodak
• Funded by the German Nazis in 1941 to guide weapons at night, 70mm f/1 produced
• Project revived by NASA in 1966 to photography the moon in shadow (not satisfied w/ Angenieux 100mm f/1)
• 50mm f/0.7 project completed with 10 copies of the lens made, 6 sold to NASA, 1 kept by Zeiss
• Other 3 lenses were bought by Stanley Kubrick who made a movie with scenes lit only by candlelight with it (Barry Lyndon)

The engineering looks eloquent to me. The key idea is to make a supersized 70mm f/1 that lights a much larger image circle, and then design a “condenser” that brute-forces your way to 50mm f/0.7 by shortening the focal length and condensing the light. Basically it’s adding a 0.7x teleconverter that gives 1 f-stop. (Boy, I really wish Nikon/Canon made these for FF lenses on crop bodies.)

In the 50mm f/0.7 optical design you can see the distinct pieces, the double-gauss (1-6) and the 0.7x condenser (7-8):

## Conclusions?

I can't conclude that the lens wasn't used (as @tfb's excellent answer doesn't yet conclude either) but it does seem reasonable that NASA would have gotten a hold of several of the fastest lenses in the world just in case they might come in handy.