Secrets of the Force

by Edward Gross

  GEORGE LUCAS

  So I was really not a fan of writing, which is ironic, but when it got to that part of it, I knew that I could write what I wanted. I knew still drawings couldn’t tell me what I wanted, because I had to do it in terms of motion. I knew that was the secret behind what I was doing in terms of visual effects by that time. The one thing I grabbed onto was that, as opposed to Stanley Kubrick who had two-minute shots in 2001, and where the thing just sits there and the ships were moving slowly so that you can see every detail and it’s perfect, I said, “I’m not going to do that. What I’m going to do is I want to do shots that are only about twelve frames, thirty-two frames at most, and it’ll move by so fast that you won’t see how it’s put together. It’ll just be a razzle-dazzle of visual fantasy.”

  We had no technology to do what we were doing. At film school, I started as an animator and we came up with this idea of basically taking an animation stand and hooking it on rails, and putting the model there and using it just like an animation stand, but an animation stand relies on persistently machined parts so that everything is exactly the same. It really became a thing of metalworking, and making a precision thing that would move down a track and then we could repeat the same move every time and also link it to computers, which was the first time that was done. They were just starting to put computers on animation cameras, and so we just said, “Well, we’ll just do it that way.” And without that, we could never have done it, because it was literally like eight hundred short shots. It was basically a model on a stick with the camera moving past it and then doing the same thing on the star field and then matching those things together. Even though it’s really simple now, it was a big head-scratcher at the time and everything depended on it.

  American Graffiti was miserable, while THX was fun. But as soon as I started getting bigger and I started pushing the envelope, I started getting myself into real trouble. When you’re in real trouble, then you’re miserable all day. What happened with this one is that it was literally that I came back from shooting, having a miserable experience, and ILM hadn’t done one shot. They had spent half the budget, it was six months later and we only had six months to go before the film was going to be released. We had eight hundred shots and there wasn’t one that I had accepted. I realized that I had painted myself into a corner and didn’t know what to do.

  When Fox accepted the script, I started preproduction with ILM. So they were working when we had to build the whole thing and build the camera from scratch, I’d hired a couple of those guys before that just to start figuring out the technology and how we would do it. I was pushing the envelope with one thing, which was how to make spaceships move fast in the space and pan with them, because I was really obsessed with the kinetic energy of a pan. I said, “If I could just put a pan on a spaceship, it’ll really make this thing take off.”

  * * *

  On June 1, 1975, that special effects company, ILM (Industrial Light & Magic), officially came into existence in Van Nuys, California, in a warehouse that the recently hired John Dykstra had discovered. Other early employees were model builder Grant McCune, Jerry Greenwood, Richard Alexander, Bob Shepherd, Al Miller, Don Trumbull, model builders Bill and Jamie Shourt, and Richard Edlund, who headed up the camera department.

  Their first task was to design and construct a motion control camera, which would allow them to program camera movements and repeat them. In July, ILM would add Robbie Blalack and Adam Beckett as the heads of the optical department. Ben Burtt was hired through USC to head up the sound department, which would, of course, be an integral part of Star Wars.

  RICHARD EDLUND

  (first cameraman: miniature and optical effects unit, Star Wars)

  I think we’re all a product of experiences. I had a lot of varied experience from being a one-man band, where I did everything myself, to working with other people. All of those experiences that led up to this contributed not only in technical prowess, which I needed to do the project, but also to prepare me psychologically for such a project and to have the opportunity to work with such an incredibly talented group of people that we assembled. I was like, one of the first three or four people that were brought in. At a higher level, anyway. It was John Dykstra, Grant McCune, Joe Johnston, and Dick Alexander. Basically, I was at Robert Abel and Associates and I was the hot cameraman there at the time. We who were the monks in photographic technology knew about each other all over town. I knew who John was. I’d never met him, but I knew who he was and he knew who I was. So one day I got a call from John, “Come on out and let’s talk about this sci-fi movie we’re going to do for Fox.” So I jumped in my Volkswagen and headed out to Valjean Avenue in Van Nuys, which was this empty warehouse. He was there with Gary Kurtz and I sat down and talked to them and in about half an hour I had the job. Maybe less. I mean, I had all the chops that were required to be the first camera on the show and basically be responsible for shooting all the elements. It was a great opportunity. I didn’t start right away; I think I gave Abel two months’ notice.

  ERIC TOWNSEND

  (author, The Making of Star Wars Timeline)

  Fox had given Lucas a $1.5 million effects budget, instead of the originally proposed $2.3 million. By contrast, the effects shots for the film 2001 cost producers $6.5 million in 1967. The motion control camera was built at a cost of around $60,000. This technology, which was pioneered by ILM, allowed filmmakers to program camera movements into a computer while filming stationary models, giving the illusion of movement.

  JOHN DYKSTRA

  Jim Nelson was hired by the production to be the producer of the visual effects. Bill Short, Dick Alexander, Grant McCune were all guys who had been working with Doug Trumbull. Doug’s dad, Don Trumbull, who did a lot of the optical and camera design work for us, obviously worked at his son’s place, too, so I brought those guys in. I brought some people from my educational background into the environment. Then I had worked at Berkeley and that was where Al Miller came from to do the electronic stuff and Richard Edlund came from Bob Abel and he was somebody that I knew from doing commercial work. Robbie Blalack had been operating an optical printer setup and he came in. We were all young. Everybody was in their twenties.

  RICHARD EDLUND

  Basically, you have to know that you can work with other people and it was a beautiful group of people that we assembled in the first few months there. For example, Robbie Blalack was a talented young guy, just out of Cal Arts, and he had made some kind of a deal with Larry Butler [visual effects artist] over at Columbia who had built a special optical printer for Marooned [1969]. So Robbie had that printer and he brought it out to Cal Arts, but it was a 35mm printer and he didn’t have the kind of money to do any work in 35mm at that time. So basically, he wound up taking that printer to the fledgling ILM and Dick Alexander had to spend many months converting it, and our electronic genius came up with a drive system for it and then we got a whole bunch of VistaVision movements from Howard Anderson, Jr.

  * * *

  Howard Anderson, Jr., was a cinematographer specializing in visual effects. He co-owned and operated the Howard Anderson Company with his brother, Darrell. The company, which was started by their father, Howard Anderson, Sr., was an optical effects house that worked on many films and television series, including the original Star Trek.

  RICHARD EDLUND

  They had an insert stage where they would shoot stuff and they had optical printers and things like that that were in 35mm. When I was at Abel’s I discovered that he had a VistaVision printer that had been sitting in a room for years and it hadn’t been used and I was trying to talk Abel into buying it. I think we actually wound up buying that printer for Star Wars from Howard for $14,000. I mean, what a deal that was and it was a VistaVision printer!

  JOHN DYKSTRA

  We rented a warehouse next to the Van Nuys airport. It was, I don’t know, four thousand or seven thousand square feet—I can’t remember which—and we set about putting together the sta
ff to make the movie. There were a lot of questions to be answered. The story of Star Wars as presented to me included spaceships having dogfights, which was the most exciting thing, and when I went to talk to George about it, we did a lot of “hands-flying.” If you have ever seen people who are pilots talk maneuvering airplanes there is hand-flying and we talked about what existing technology was available to actually execute work like that and what our options were.

  All of those effects were to occur while being viewed from a camera platform that itself needed the fluidity and freedom of motion of a camera plane. This visual concept was a far cry from the locked-off camera approach to spacecraft miniature photography seen in the space classics of the past. This was a challenge, to say the least. As the meetings and story breakdown continued, it became clear that this film would not showcase twenty or thirty special photographic effects shots, but would use spaceships, miniatures, and all manner of photographic effects, as you would use automobiles in a film of contemporary time setting. In the entire film, there are some 365 miniature and photographic effects shots. The challenge, therefore, became a task of mammoth proportions. In order to produce the quantity and quality of special photographic effects shots called for in Star Wars, a complete in-house system would have to be developed. This system would include miniature design and construction facilities; the design and fabrication of a camera motion control system, electronics, and mechanical facets; and a complete optical house and animation department. I felt that the in-house system would be the only way that consistency of quality and control over each of the separate operations could be maintained.

  I had an overall view of what I thought needed to happen and it was based on my experience with Doug Trumbull and his visual effects facility Future General and my experience at Berkeley using computer-controlled cameras for the National Science Foundation. It involved the creation of a complete studio because we needed to do the construction of the miniatures, we needed to construct the camera systems, and we had to develop or construct the optical department because we were going to use a larger film format than conventional. One of the problems with film is that when you duplicate it, which is what you do when you do optical printing, you lose quality. It picks up contrast and the grain gets worse and the resolution gets worse.

  So I wanted to do a large negative in order to reduce the quality loss that came as a result of that optical printing step. So we were going to use VistaVision and the reason that the optical printer was purchased was because it was a VistaVision printer. That was our format of choice and of course that required that we build new operating systems for the printers because they were all old-fashioned systems and [we needed] new optics because the quality of the glass had a lot to do with the quality of the image. So we designed and built new lenses and purchased some new lenses for the optical systems and designed a complete photographic system meaning cameras and film-handling equipment around the VistaVision format.

  * * *

  VistaVision was a higher-resolution, widescreen variant of the 35mm motion picture film format. Created by engineers at Paramount Pictures in 1954, the process required new cameras through which the 35mm negative traveled horizontally rather than vertically. The result was a negative image with an area nearly three times that of a standard negative image. IMAX, also a horizontal film system, is a descendant of VistaVision.

  RICHARD EDLUND

  I remember when we went to see the printer, we had to climb up about three flights of stairs along the side of the stage to get up to the top area where this printer was. Then we go into the room and it had these switches that you turned. They weren’t toggle switches. You turned them to turn the light on and everything was kind of battleship green and there’s the printer standing there. It was not dusty because the room was closed and actually the last camera report has been written up on The Ten Commandments [1956] and was sitting on the write-up table. They basically left the room and never came back and so twenty years later I guess it would have been, we wound up buying the printer.

  Howard also threw in two roto benches that were these big contraptions with mirrors and large format. So we could roto at twenty-two-inch field, which is what I decided on and so we had twenty-four-inch cells and these Bell & Howell 2709 cameras, which were very steady cameras that had blanking shutters in them so it would pull twice to make eight—that’s tech knowledge, but anyway, we had two of those roto machines and one or two extra 2709 cameras, and then a box of movements. These were VistaVision movements and in compositing, you have to have a positive and a negative movement for each head and the big pin is in a different position for each movement …

  The issue in compositing for motion pictures is you’re lining up with a composite image that’s less than an inch wide and that’s being blown up to fifty feet. So a thousandth of an inch in a one-inch image is like 3/4 of an inch or an inch when you blow it up. So the kind of precision that’s necessary and all that was uncanny and it was miraculous that we were able to put 365 composites together in under two years for Star Wars. I mean, we had to actually build the system to start with.

  JOHN DYKSTRA

  We had to start designing equipment the day we walked in and assembling the team that was going to do it. So we had to have machinists, cameramen, model builders, you know, lighting and grip, all of the things that are required to make a studio that will operate. So we’re putting together our team while we were designing the cameras. We were building the cameras while the miniatures were being constructed. We began photography when the first of the cameras and the first of the miniatures reached completion and we had the wherewithal within this studio to start lighting and shooting stuff. It wasn’t as quick as the studio would have liked, but we were building cameras from scratch and designing and building a complete numeric control camera system including building the processors that ran it. There were no laptops or desktop computers at that point. The computers that controlled the movement of the camera were individually designed and wire-wrapped. That stuff was all designed and built essentially in the first year. The second year was spent finishing the systems, redesigning and reconstructing some of the systems to meet the new and special requirements of shots that were specified for the movie and then completing the work.

  RICHARD EDLUND

  When I got there, there was a big empty stage that was like a hundred feet across and fifty or sixty feet wide. That would be the area that I would have for the boom camera that we built and in the beginning, there was nothing on the stage but a card table with a phone on it. And so that’s what I started out with and we had to build the Dykstraflex, which had a forty-two-foot track and a boom that went up and down about ten feet, maybe a little more; and it went sideways, it rotated, camera pan, tilt, and roll; that was the camera.

  * * *

  One of the most important early developments was the creation of Dykstraflex—designed by Dykstra, Al Miller, and Jerry Jeffress—which was the first digital motion control photography camera system. Created specifically for Star Wars, the system used old VistaVision cameras for their higher resolution and allowed for seven axes of digitally controlled movement, which could be duplicated for multiple takes. The trio would ultimately win an Academy Award in 1978 for their creation of the system.

  JOHN DYKSTRA

  I had the idea for years. This is just one permutation of the concept. It’s not a new concept. It’s very old, like blue screen or front projection. It’s been around for years, just waiting to be perfected or improved to suit the needs of a particular situation. Al Miller is an electronics designer. We sat down with two bottles of wine during the old Future General days and figured out what the machine ought to do, what it ought to look like, and how to program it. Basically, it is just a combination of all the techniques that have been used for years, combined into a sophisticated device capable of manufacturing techniques.

  It started out with the premise that we wanted to have the flexibility to move the camera in free s
pace and have multiple moving subjects. Traditional techniques to do that where you wanted to move the camera and the subject and do it non-real-time, meaning not at twenty-four frames per second, was stop-motion photography and that was the Ray Harryhausen process. One of the problems with that process was the camera was still during the exposure so the subject and the camera move itself had no motion blur.

  RICHARD EDLUND

  So there’s like five or six channels of motion control required there and then you have the model mover which is another track that was perpendicular to the main track and that one went from right to left on the tracking. It didn’t have a boom, but it rotated and you could rotate the model. So that was the limitation that we had and we built this very complicated contraption that I compare to a violin that we then had to learn how to play. So we built this very complicated machine and then I had to learn how to play it and then program shots one axis at a time. I usually wound up doing the track first to get the size change and then I’d do various pans, tilts, rolls, and I’d have to position the model right in the right position. We were flying right by the model often with fairly wide lenses to get a real motion dynamic going and every once in a while I’d crash into one. I’d scrape the model and then I’d have to call Grant [McCune] and he’d have to come out and fix it. It was quite a deal.