In the winter of 1975, a film reached theaters in which the 18th century smells of wax and smoke, and the camera sees what, before it, only the human eye had seen by candlelight. Behind it was the story of how “space” lenses—built for NASA and wrapped in a legend about filming the dark side of the Moon—ended up in the hands of a director obsessed with the paintings of William Hogarth.
In optics, there is one number that determines everything: the f-number, the ratio of focal length to entrance-pupil diameter. The lower it is, the more light reaches the film. Most cinema lenses in the 1970s worked at f/2 or, at best, f/1.4. A lens with an f/0.7 aperture lets in four times as much light as an f/1.4—the difference between making out the face of the person across from you at a candlelit dinner and seeing only a blurred patch. Carl Zeiss produced a small run of such lenses—the Planar 50 mm f/0.7—and part of the order went to NASA programs. Legend credits them with filming the dark side of the Moon, though Zeiss itself has long clarified that the lunar surface in 1968–1969 was photographed with its Biogon lenses, while the ultra-fast Planars were developed for another purpose—high-speed photography in extremely low light. It was a beautiful story, and it stuck to the glass for good, outliving its era. The lenses were designed for still cameras, for stationary cameras and precisely timed exposures—and then three of them ended up in England, on a soundstage where actors played Irish adventurers and Prussian aristocrats.
How exactly Stanley Kubrick found these lenses was recounted by Jan Harlan himself: Kubrick read a note in American Cinematographer about the 0.7/50 lens and caught fire; Harlan called Zeiss, where Dr. Kemmerer honestly warned him—the glass was unusable on a motion-picture camera, its rear element sitting only five millimeters from the film plane. Kubrick, true to form, refused to take “no” for an answer. Harlan eventually bought one copy and took it to Ed DiGiulio, then, after successful tests, bought two more for conversion to other focal lengths. Kubrick had been a photographer before becoming a director; he understood optics like an engineer and felt them like a painter. He needed candlelight—the real thing, warm, uneven, falling away into darkness in the corners of a room. Ordinary 1970s film equipment could not capture that light without an army of fixtures that would have killed the whole idea. An f/0.7 lens could—but it had been made for an entirely different camera, a different format, a different universe of tasks.
Mounting a NASA photographic lens on a movie camera is rather like transplanting an aircraft engine into a boat: theoretically possible, practically an engineering nightmare. Ed DiGiulio, president of Cinema Products Corporation, took on the job, and what he did to the Mitchell BNC camera deserves a story of its own.
The first problem: the rear element of the Planar lens sat just 4 millimeters from the film plane. Four millimeters is the thickness of two stacked coins. In an ordinary movie camera, between the rear lens and the film sits a reflex shutter, a viewing mechanism, a whole array of moving parts. Here, there was simply no room for them. DiGiulio chose a non-reflex Mitchell BNC camera, whose viewing system worked differently—by racking over—but even there, the housing and the aperture plate had to be bored out from the inside, metal chewed away millimeter by millimeter so the glass would not strike the wall.
The lens itself also had to be rebuilt. The built-in Compur central shutter—standard on the photographic version—was removed. The adjustable shutter vane was taken out. The iris blades were converted to manual control, because the automation of a still-photography lens and the rhythm of a movie camera are incompatible things. In essence, what remained of the original lens was its optical design; all the mechanical hardware around it was rebuilt from scratch.
Focusing at f/0.7 is a trial of its own. Depth of field at such an aperture is vanishingly thin: in a close portrait, the tip of a nose could be in focus while the ears had already slipped into soft blur. For precise focusing, DiGiulio designed a fine-pitch mount, in which moving from infinity to a distance of 5 feet (roughly 1.5 meters) required two full turns of the focus ring. On an ordinary lens, that range is covered in a quarter-turn. The slow thread gave cinematographer John Alcott a chance to catch focus to within fractions of a millimeter—but any movement by an actor in the frame turned it into a race.
And there is one more detail that genuinely amazes me. Because the rear lens hung four millimeters from the film, any careless shift of the camera for viewing could shatter the glass or damage the film gate. DiGiulio designed a protective microswitch: before shifting the camera to check the frame, the operator had to turn the lens forward by nine full revolutions. Nine turns—a ritual, a compulsory prayer before every glance through the viewfinder.
A 50 mm lens on 35 mm motion-picture film gives a field of view close to a moderate telephoto—good for portraits, but too tight for wide shots of a room. Kubrick needed scenes in which the whole hall was visible, with its candelabra, the table of card players, the depth of the space. The solution came in the form of a Kollmorgen projection adapter: an optical attachment that reduced the effective focal length to 36.5 mm while retaining the f/0.7 speed. That was closer to a normal movie lens; the angle of view had room to breathe.
There was also a more radical attempt: the Dimension 150 adapter, capable of compressing the focal length down to 24 mm—a wide angle in which the entire room opens up. But it was rejected because of geometric distortion: straight lines swam, walls bowed, and the 18th century began to look like a reflection in a Christmas ornament. Kubrick, for all his obsession with light, was not prepared to sacrifice the geometry of space. A room had to remain a room.
The film stock also became part of the experiment. They used color negative Eastman Kodak 100T 5254—the standard studio stock of the day, rated at 100 ASA under tungsten lighting. It was pushed one stop in processing, raising the exposure index to EI 200. Push-processing is, roughly speaking, leaving the film in the developer too long: the grain grows larger, contrast increases, shadows become a little noisier. But combined with the f/0.7 lens, even this modest shift was enough—the candles stopped being set dressing and became sources of light.
This is where we enter the territory in which cinematic legend slowly starts nibbling away at reality. The line “the film was shot entirely by candlelight” is one of the most beautiful in film history, and it is almost true. Almost.
The key nighttime interior scenes really were filmed by the real light of specially prepared candles, with no added artificial fill light. This is not a publicity stunt—it is a documented fact. The candles were real—they were selected for size and brightness, arranged in the frame with the precision of a lighting technician rather than a set decorator. Every candelabrum performed a double function: it was an 18th-century prop and, at the same time, the only lighting fixture.
But “by candlelight alone” applies to the nighttime interiors, and the film runs for more than three hours. The daytime scenes—and they are the majority—were shot in natural daylight streaming through the windows of historic buildings. Nor was this a complete renunciation of control: windows could be draped, reflectors could redirect the light, and the choice of shooting time determined the character of the illumination. Kubrick did not give up staged control—he gave up electric control in particular scenes. The difference is substantial. The candles were placed deliberately, their number in the frame was regulated, the staging built so that actors’ faces would fall into pools of light. It was the direction of light, with fire performing the role of lighting equipment.
I think it matters not to destroy the legend, but to sharpen it. Shooting by candlelight without a single electric fixture is already a feat. But the feat was not that the lighting was accidental and “documentary”; it was that every candle stood where Kubrick wanted it, and the f/0.7 lens allowed that intention to take shape on film.
A detail that contains all of Kubrick: the candles for the candlelit scenes were custom-made, and a couple of the specimens Harlan gave to the Zeiss museum decades later had three wicks instead of one, so the flame would be just a little brighter than usual. Even the props here were engineering solutions.
Kubrick did not hide his visual sources. The paintings of William Hogarth, Thomas Gainsborough, Michelangelo da Caravaggio, and Rembrandt—artists who knew how to paint candlelight and oil-lamp light so that the viewer can feel warmth at their fingertips. Genre scenes of the 18th century—a card game, a family dinner, a salon conversation—existed in painting long before cinema, and artists of that era were solving the same problem Alcott faced: how to convey volume and mood with a weak, concentrated source of light.
In painting, it works through chiaroscuro—the sharp contrast of light and shadow, when a face is snatched from darkness while everything else sinks into half-light. The f/0.7 lens reproduced that effect optically: shallow depth of field blurred the background, candles cast hard patches of light across faces and hands, shadows fell thick and naturally. Kubrick achieved chiaroscuro through the pure physics of glass.
His ambition was specific: the past had to look not like a theatrical set lit by spotlights, but like a place where people lived in that light. Spotlights give cinema away—they create even, flooding illumination, with soft, predictable shadows. Candles give life away—the light is uneven, flickering; it catches one cheek and leaves the other in darkness. The viewer of “Barry Lyndon” does not watch the 18th century from a theater seat—they seem to be peering into a room through a half-open door, seeing what someone of that time would have seen. That physical tangibility of the past is rare in historical cinema, where costumes are usually too clean, faces too well lit, and candles burn with suspicious brightness.
There is another aspect that is rarely discussed. Eighteenth-century painting is a frozen moment. Cinema is movement. When an actor slightly turns their head beside a candle, the pattern of light on their face changes instantly—as it would in a real room, but never could in a painting. Kubrick captured what painting could only imply: candlelight in motion, its instability, its living breath on skin.
Zeiss Planar f/0.7 lenses did not become a standard cinematic tool. They remained exotic—one-off lenses suited to highly specific tasks, requiring camera modification and inhumanly precise focus. After “Barry Lyndon,” Kubrick kept them, and they were used intermittently, but there was no mass imitation: too complicated, too temperamental, too dependent on the skill of a particular cinematographer.
Yet the consequences proved broader than the technology itself. John Alcott won the Oscar for Best Cinematography for “Barry Lyndon”—an award difficult to separate from DiGiulio’s engineering madness and Kubrick’s stubbornness. The film proved that extreme light sensitivity could work as an artistic solution—the choice of optics shapes aesthetics as much as the choice of actor or music. Historical cinema after 1975 became more attentive to light: directors and cinematographers began asking what kind of light people in the past lived by, rather than what kind of light was convenient for filming them.
The trajectory of these lenses themselves—from Cold War space programs to candlelit interiors—says something about the nature of technology. Optics designed to shoot in light that was barely there turned out to be the perfect tool for recreating a world in which people had lived two hundred years before Apollo flew. The military-space age produced the glass through which audiences saw the card tables of Georgian England and Prussian drawing rooms lit by the same fire that had burned there in the lifetimes of Thackeray’s originals. At the time of its release, it was the fastest lens in the world—its speed even made it into the Guinness Book of Records. All three converted examples have survived: two travel the world as part of the traveling Stanley Kubrick Exhibition (since 2004, it has been seen by more than 1.5 million people), while the third—the one Harlan tucked away “just in case”—came home in 2022: Harlan lent it to the Zeiss optics museum in Oberkochen, together with a pair of those very three-wick candles. The glass that filmed 18th-century candlelit scenes took its place in the display case of the company that had once ground it. The legend, now hard to separate from Ed DiGiulio’s engineering specifications, is displayed with a museum placard.