How Kubrick Made 2001: A Space Odyssey – Part 6: Jupiter and Beyond the Infinite [A]


Kubrick: “You can say a lot of things about what it should be you know what should be. You know, it should be an interesting story, offer possibilities for cinematic beauty in terms of photography and editing and, it’s in the end, truthful and honest in dealing with the thematic material. Historical films offer possibilities of doing the sort of things that movies do better than anything: else re-creating atmosphere (just as futuristic films do). Description is the most boring thing in a novel whereas in a movie, what you put on the screen is a great effort to the makers of the film. I suppose that, you know, I would always be attracted to something which offered interesting visual possibilities.”

In an interview, Bryan Loftus talks about the day the studio backers came to visit. They were concerned because, despite Stanley Kubrick helming the project, it had been two years and tons of money had been spent and they hadn’t yet seen a foot of film. Loftus said, “Stanley had put together this reel of some of the scenes that equaled about fifteen minutes. It was all spectacular stuff. I was just walking by the preview theater one day and Stanley saw me and he said, ‘Brian! Brian! C’mon on here, and take a look at this.’ He dragged me in and it was just him and I. The thing that impressed me the most was the music he had put to it. I said, “My God Stanley, what is that music?”  He said, ‘It’s Mahler’s Third.’ I’ve never seen that footage again with that particular music to this day, but when the backers saw it, they said, ‘How much more do you need’” It was powerful stuff” (Loftus Interview).

Publicist Ivor Powell sat in while Kubrick showed some footage to the backers. He recounts how captivating it all was. They even took the press on a tour of the sets.

He said, “There was a day when the press came and I remember all of them standing on the pod bay set and [First Assistant Director] Derek Cracknell was standing in the back with a microphone saying, “Okay…Open pod bay door number one please Hal...” Then the door would open up.  It was amazing because it really felt like you were on a ship.  Stanley really spared no expense on the film, and it was the same on all of his films as you could imagine. If something was required to be stainless steel on any of those 2001 sets, then it was, and it wasn’t spray painted silver or anything like that to save a few pounds (£s)” (Powell Interview).

The final sequence of the film begins with a tilt down from the pinpoints of starlight in a vast darkness to Jupiter and it’s moons. A floating monolith becomes visible from light reflecting off of its surface.

The destination of the Discovery One spacecraft was originally Jupiter, but in “early September 1965” Kubrick decided to change the destination to Saturn. This passage is from a biography called Sir Arthur C. Clarke: Odyssey of a Visionary, “Kubrick’s desire for the change was based on aesthetics—the beauty of Saturn’s rings and the many spectacular special effects that could be created as the spaceship flew among them. He asked technical adviser Fred Ordway to research the ringed planet.

‘I prepared the memo,’ says Ordway. ‘Kubrick was delighted, and Clarke backed the change with great enthusiasm. Arthur then brilliantly wrote Jupiter out and Saturn in. Everybody got excited about the change except the special effects people, who went ape.’

Eventually this major change proved to be too much for the special effects group, which was already under tremendous pressure… [Kubrick] made the final decision. New Saturn effects would have presented ‘insurmountable problems’—technical as well as too much additional time and money—for a production already over budget.

For the film, therefore, Jupiter remained the destination planet for spaceship [Discovery]. But Clarke retained Saturn as the destination planet in the novel, and this became one of the major plot differences between the book and the movie” (Odyssey of a Visionary).

In order to create Jupiter, Douglas Trumbull used “a semicircular band that could rotate on a special rig.” Trumbull says, “‘Onto that band, two projectors projected a revolving transparency of some artwork I had painted; and the result was a spherical scan, rather than a linear scan, because the band just rotated on its own axis rather than moving through space. So Jupiter was just shot as a still photograph. I used an 8×19 camera—at f/124—and it took about two hours to make one image. Once we had that, we could composite it with moons and stars and such on the animation stand’” (Cinefex 113).

Douglas Trumbull: “Anyway, you start seeing the painting is now scanned onto this hemispherical surface and it results in Jupiter as one continuous exposure. If you look at this very very closely you’ll see these anomalous scan lines here, which are, in fact, slight changes in exposure from some variation in the generator or whatever, but once I airbrushed this and modified it for the animation camera I could make it into a pretty successful Jupiter.”

Pretty much everything in these shots is animated except for the monolith itself, which was a 6-foot-long miniature, which “was lit and shot on a stage with a motor revolving it like a barbecue spit” (Cinefex 112).

“The black block was filmed against black—positioned to fit a particular location on the screen—and added into the space shot. The moons in that shot were generated with a slide projector and a small sphere about a foot in diameter. Some colored patterns and textures were loaded into the slide projector, then projected across that sphere, which was photographed with a 35mm still camera” (Cinefex 112).

Colin Cantwell, who worked on special photographic effects, said, “We could shoot moons full-phase, or half- or crescent, just by changing the position of the slide projector. And sizes could be varied simply by setting the camera closer or farther away from the sphere. In the end, we had this nice set of 35mm slides with little moon images on them that ranged from about a quarter of an inch for a large moon down to an extremely small image for a small one. For the big alignment shot, we literally just cut little round holds in black paper and then rephotographed the slides bottom-lit’” (Cinefex 112).

The big wide shots of space and the tiny Discovery One introduce us to the scope of the final sequence, but also instills in us a sense of isolation after everyone on the ship except Dave have been killed—including HAL. There is no sense of normality by way of human relationships or the comforts of the ship—there is only Dave and the unknown.

I’ve always wondered why Dave leaves the ship behind. Perhaps he had seen the monolith floating in space and went out to retrieve it.

Keir Dullea: “Why am I in the pod? You know, you could- perhaps the decision was that that HAL, before he was put away by me, had somehow affected the running of the ship.”

Originally, Dave was going to find a huge monolith-shaped slot in one of Jupiter’s moons that he would fly the EVA pod into. Apparently in Clarke’s original treatment of the story, Frank does not get killed by HAL and the two men discover the gigantic slot in one of Jupiter’s moons. They send probes down into it and one of the men accidentally gets sucked into the slot and the other chases after him (Cinefex 110).

Trumbull stated that the slotted moon effect “just never looked convincing” and Kubrick was unsure about the effect. So, they dropped the idea and, instead, replaced it with a new introduction of the Stargate sequence. This time, Dave would see the monolith floating in space and fly the EVA pod over to it. He would extend the pod’s arms to touch it, but instead of a surface, there would be nothing and Dave would slip into it (Cinefex 110).

They actually filmed this effect by taking a shot of the full-sized pod traveling one way and then they moved the camera and had the pod traveling the other way. Then they combined the two shots, but it “never looked right” (Cinefex 110).

Trumbull said, “Stanley could never fully visualize whether he would like a shot or not unless he saw it on the screen… so we would take still photographs of Jupiter and several moons and some stars, put them in some kind of configuration, and just shoot thirty feet of film so Stanley could look at it in the theater” (Cinefex 110).

Colin Cantwell came up with the alignment of Jupiter and the moons while working on different configurations on the animation stand. Trumbull notes, “It looked terrific—everybody liked it enormously. It was like a major religious alignment of moons. Then someone suggested that we put the monolith up in there, so we did. Suddenly the whole scene started to come together out of these random elements” (Cinefex 112).

It was while they were pursuing these ideas that they began working on the Slitscan technique.

The Slitscan effect in the film was based on the work of an avant-garde animator named John Whitney. You have probably seen his work before. In 1958, Whitney animated the title sequence for Alfred Hitchcock’s Vertigo alongside a graphic designer named Saul Bass (Wiki).

In 1961, Whitney made an experimental short film titled Catalog, which showcased several animation effects including Slitscan. In this animation, we can see in the first few frames how the concept works. Basically, if you are taking a long exposure, you can create lines and shapes with single points of light. When these lights rotate, it becomes a circle. When the lights change, the circle changes.

Douglas Trumbull and Con Pederson had seen a demonstration by Whitney while they worked at Graphic Films—the company that made To the Moon and Beyond. Whitney had also worked on To the Moon and Beyond and had experimented with Slitscan on that film.

On July 23rd, 1965, Pederson wrote Kubrick this letter:

Dear Stanley:

We have more film for you to look at – a couple hundred feet of 35 B&W test made by a man we often work with, John Whitney. He calls this particular technique “slit scanning”. Essentially, it is a method of obtaining smooth and continuous motion variations from a single initial element, in two-or three-dimensional configurations.

Its possibilities are unlimited – particularly when we combine it with other components in animation. It can be adapted to “hard” images such as those in the test, or “soft” elements such as mist, light, etc. We want to plan around the technique in displays such as the hologram contour-analysis (which reveals the Stargate on Jupiter V). It may also provide some solutions in the ending.

We have lots of ideas cooking – more on that as soon as I can get a report prepared.

For now, can you please tell us how we can ship you this small roll with best chance of it getting through all the snags? Through MGM here maybe?

Good luck,

Con Pederson


Kubrick responded:

Please airfreight John Whitney film directly to me through Barnett International Los Angeles –stop- Advise Barnett to airfreight via TWA and address as follows Translloyd Air Services LTD for Hawk Films LTD London Regards Stanley Kubrick

Trumbull had said that Kubrick told him vaguely that “he wanted the camera ‘to go through something’” (Keil, Whissel 118). Trumbull used the technique from the Whitney animations, but this time, he put the slit outside the camera. He made a rig with an animation camera on a track that would move toward a sheet of glass “masked in black material apart from a small vertical or horizontal slit” (Keil, Whissel 118). It would work to the same effect as when you take a long exposure of a highway at night and all the cars end up looking like streaks of light. (long exposure cars.jpg) By having the camera track toward the light, the light would look like it was moving toward the camera as the camera got closer to the slit. They could also flicker the light to create spaces between the streaks.

Kubrick liked the tests so much that he “authorized the construction of a dedicated Slitscan machine in an upstairs room of the animation building” (Cinefex 110). The machine was described as “a six-foot-tall rotatable slit cut into a sheet of metal and positioned directly in front of a mechanized twelve-foot-long glass panel onto which backlit artwork was affixed.” “The room was painted totally black,” recalled Trumbull. “We erected our slit with a light source directly behind it. Then we built a track leading up to the slit and mounted a 65mm camera on it with a shutter that could stay open on one frame of film. To modulate the lighting coming through the slit, we used high-contrast negative transparencies that would slide behind the slit as the camera moved toward it. Some of them were very tightly controlled patterns that we photographed from books of optical art—elaborate patterns of circular lines or straight outs ended up in there. Then we’d lay in different colors with gels. So, by having this artwork move behind the slit, backlit, we were able to effectively turn light on and off. As we did that, we moved the camera along fourteen feet of track toward the slit—a full fourteen feet for each exposure. It took about forty-five seconds to a minute per exposure, and each frame was made up of two exposures” (Cinefex 110).

Douglas Trumbull: “So basically, you could make an exposure from 15 feet to 1 inch in one minute, so the focus is constantly shifting during that entire exposure time so you have literally an infinity of focal distances.”

With both the slot and mirror effects scrapped, they had to figure out another way to begin the Stargate sequence. Trumbull said, “’We had a technique we used all the time… of putting two pieces of film on separate projectors and projecting them onto the screen at once. That way we could see a spaceship with a star background, or any other combination of elements. We tried this using the slitscan material coming out of nowhere—right out of the stars—and it looked terrific. I had filmed one slitscan shot in particular that was different from all the others in that it started from total black” (Cinefex 112).

Trumbull also had a small accelerator motor so that the speed of the lights would constantly increase (Cinefex 112).

Trumbull goes on to say that, “the Camera was a big, heavy device that traveled along this track pretty fast… until it got about two inches from this giant twelve-foot pane of glass and had to stop dead. It had a special electric brake and clutch and method for reversing all the motors, and it ran automatically. There was always a certain amount of wear during a shot, though; and after a while, the brake would start to slip, and the whole thing would slide a little bit closer to the glass each time. So if someone wasn’t there to make the necessary adjustments, sooner or later it would go crashing through the glass. But it was basically automatic; and oftentimes that camera would run for thirty-six hours, continuously, to shoot one take. All together, we spent at least six months working the slitscan footage” (Cinefex 111).

I had a chance to visit the Stanley Kubrick exhibit when it was in San Francisco and they had a booklet that featured Trumbull’s detailed description of the Slit Scan Camera Mechanism. I’ve put a link to it in the description if you would like to read it.

Keir Dullea: “The set was not shaking. I did this with my body. I tensed my body up and the cameras that close, you tense your whole head up and eventually you can make yourself shake and that’s what I’m doing there… and he played an excerpt—if anyone wants to go out and get it and listen to it, you’ll see how beautifully it fit that sequence. He played one of the movements from the Antarctica Suite by Vaughan Williams.”

The still shots of Dave’s face frozen in time were likely necessary to bridge the different sections of the Slitscan footage. We can see how it is able to change from a vertical split to a horizontal split and it is only Dave’s face that is jarring and not necessarily the transition. It is the same with the shots of Dave’s eye.

Eyes on Cinema found a rare trailer for 2001 that includes some test footage of the Stargate sequence that doesn’t appear in the film. And I’ve put the link in the description if you’d like to check it out.

One interesting shot shows patterned shapes hovering over a cosmic sea.

Douglas Trumbull: “The diamond shapes floating in this weird kind of almost liquidy landscape, were taking 35-millimeter prints of the Stargate sequence and projecting them onto little triangle shaped facets of these diamond shape tetrahedrons. So each of those little- each facet of the tetrahedron is actually a movie being projected onto a white plastic tetrahedron. So it would be one pass, camera gets one back, the projector gets moved to the other side, another exposure, on and on and on over multiple tetrahedrons and it just took hours and hours and days of photography to transpose those movies onto all those facets.”

As for the surface below the shapes, Trumbull explains, “The thing I did was to make a drum, about six feet long and one foot in diameter, and cover it with crinkled aluminized mylar. Then I placed a little piece of black paper at an angle in front of the slit so the camera [would] only see the reflection from the slit on the lumpy surface of the drum. So now, instead of a flat plane of light, I had one that was randomly modulated; and by shooting the drum horizontally, I created something that looked like a rolling sea of light, changing color and shape as it came toward the camera” (Cinefex 113).

The Slitscan effect has been used many times since including the titles for Superman and for the wormhole in Star Trek: The Motion Picture (Keil, Whissel 118).

There are many shots in the Stargate sequence that weren’t solely the product of the Slitscan rig like in these shots.

These abstract shots were the first that Kubrick shot. They were done so early that Kubrick shot them in a small darkroom in New York City before the production left for England. In order to make these massive cosmic events, he dripped small amounts of lacquer and paints on water. They were shot with a macro lens and none of these were larger than a pack of cigarettes (Making Stargate Sequence).

John Alcott says that they used “a big black tin, about the size of a table, filled with black ink and banana oil” (Cinefex 113).

“We had so much light coming in onto it,” says Alcott, “and the intensity was so great that we had to shoot it quickly because the radiant heat would boil the ink. We would take a matchstick or a toothpick and dip it into this white paint and let it drop into the center of the tin. It was like dropping paint into thinners—as you dropped it into the black ink, it would shoot off into different areas. We photographed this at high speed, at 96 frames per second, in reverse, so that instead of its dispersing away from camera, it would look as if it were coming toward the camera. It was simple, but very effective because you could put a white spot and then little red on top of the white and get a buildup of different colors’” (Cinefex 113).

Christiane Kubrick: “ I remember these large trays with black water, smelly, and Stanley leaning over them and dribbling inks and oily things and lacquers onto the surface and they exploded the most beautiful way and he loved these shapes. They were very organic, in fact, they followed the same physical law as, presumably, the cosmos does and so he was very excited.”

What follows are several sweeping landscape shots over an alien planet. Most of this footage was shot in Scotland in late October 1966. If you recall, I spoke earlier about the YCM process for Technicolor. There would be a strip of film in yellow, a strip in cyan, a strip in magenta, and a base strip in black and white. When put together, you get a normal looking color image. 2001 was shot on a single strip of 65mm color negative film, but you could still apply the YCM process to it. While making the prints, they were able to separate the film into strips of yellow, cyan, and magenta, but instead of putting them back together normally, they messed with the f/stop of each strip (Birkin Interview).

In an interview with the TV Store Online, Andrew Birkin explains:

“What Bryan [Loftus] discovered was that if you put the strips together using different f/stops it produced this weird effect with the colors. You didn’t get the same wash pending the density of what was photographed. So reds could turn blue, and greens could turn purple. Bryan showed me what he had discovered and I suggested that he should take it to Stanley. Bryan was very shy and reserved.

I said, “Okay, well, would you mind, then, if I took this to Stanley?

He said, “Well, what exactly would you suggest to him?

I said, “I don’t know, but I’m just thinking about how this might be able to stand-in for the landscapes that [Special Photographic Effects Supervisor] Con Pederson had been painting but were months away from realization.”

Stanley said, “Well, what are you suggesting?”

I said, “How about mounting a camera on a plane or helicopter and then let Bryan do his magic.”

He said,”Hmm…that could be interesting. Okay, go and do it.”

I said, “Who? Me?” (Birkin Interview).

Special effects supervisor Wally Gentleman had been discussing the alien landscapes as early as when the production was still in New York City. Gentleman had worked with Technicolor cameras a lot and knew that botching one of the strips would produce bizarre colors. It appears at this point, that Gentleman had already brought up the concept of messing with the Technicolor process with Kubrick (Cinefex 113).

Birkin continues: “So we did some tests to make sure that the idea would work. The best we could do was to make the camera a part of the aircraft, by bolting it to the floor. I went to work with surveying maps. I just looked at places in Scotland that I’d always wanted to go. I began to plot out various runs against the times of day when there would be too much shadow. I had twenty pages of notes which I ran by Stanley. He sent me to Scotland with a Camera Operator” (Birkin Interview).

Birkin goes on to say that when he was in Scotland, the camera operator became so nervous about the conditions that he decided not to go through with it and left Birkin in charge of the camera operating. Birkin was quite inexperienced, but had worked as a loader on a commercial. For a lot of these shots, it was only Birkin and the pilot in the helicopter. They shot [quote] “miles and miles of film”—about 600 feet for every foot used (Birkin Interview).

Kubrick liked the shots so much that he sent another crew out to shoot landscape shots of Monument Valley, which you might recognize from any number of Westerns. These shots were done by second unit director Bob Gaffney and they had to fly dangerously low to capture them.

Back in the studio, Bryan Loftus was hard at work converting the landscape shots into bizarre alien colors. Loftus said: “You had three color strips, and you took those and made three high contrast versions off of the negative, and then three low contrast versions. Effectively you would be creating twelve different strips of film for every shot in the Stargate off of the positive and negative film. So by effectively combining those the wrong way you would end up getting lots and lots of different colors on the film.  You couldn’t predict the color combination and how it would come out.  So we just had to keep going and going through all of the combinations.  Stanley finally said, “Well, I think I’ve seen everything that we can do. Can we get other combinations?” So, I came up with the idea of random combinations” (Loftus Interview).

He rigged up three spinning wheels that would give them a random setting to use for the aperture, color filters, and so on. Kubrick’s goal was to come up with color combinations that no one had seen before.

One major part of the Stargate sequence that didn’t make the final cut was a section called the “City of Lights.” Trumbull explains,

“I made a series of panels with hundreds of light bulbs all over them in patterns—hexagonal patterns, rectangular patterns, circular patterns—in varying diameters. For example, there would be a rectangle inside a rectangle inside a rectangle, and each could be turned on and off independently. Then I rigged up a thing that allowed me to turn the camera up on its side; so now, instead of moving in and out toward a piece of light, it was moving relatively up and down. With this configuration, I could shoot the pieces of artwork – these panels of light that I could switch on and off during the exposure – and what I would get would be volumetric scanned exposures of building-like shapes. I made a bunch of cam-operated sequencers that would change the diameter of the shapes as they scanned vertically in space so the shapes would taper off as they got taller, like the Empire State Building” (Cinefex 113).

Thanks for watching! The next and final part is already written and partially recorded, so it shouldn’t take too long to complete. If you enjoy the videos I make, please help me make more by donating to my Patreon. For a dollar, you’ll get early access to videos and for five dollars, you’ll get all of my unedited notes and transcripts and all of the photos and PDFs I use to make my videos. And if you’re new here, please hit that subscribe button now because there are plenty more videos on the way for cinephiles like you!

Thanks again for watching!