Jamie Shourt and McCune; in the background is the “spray booth”
The nearly complete model.
“In many cases you are using all kinds of flimsy stuff that doesn’t necessarily want to repeat its motions exactly,” Dykstra explains. “So if you shoot a very complex move and send it in, you have to cross your fingers and hope everything will work right. You can’t tear that rig down and go on because if it comes back from the lab and it’s bad, you can’t duplicate that rig again. You can proceed to shoot the next element only after each of the elements had been viewed.”
* * *
FANTASTIC MOVING MACHINES OF THE FUTURE
The motion-control camera would solve many of these special effects puzzles, because it could be programmed to repeat its actions exactly, an idea that had been around for several years. In fact, the camera would be the combination of three machines that already existed in different forms: a specialized camera, a specialized mechanical rig, and a specialized electronic-control device, all of which came to be called the “Dykstraflex.”
Technical drawings indicate how the motion-control camera would work and how it would photograph the models, which would be placed on blue pylons to hide the armatures holding them up—one of Dykstra’s innovations.
Joe Johnston, Steve Gawley, and others upstairs in the bare-bones ILM art department.
A blueprint for the motion-control camera, or “Dykstraflex,” is marked “tilt plate assys” (courtesy of Pete Vilmur).
“All three of the aspects of that camera system were developed at some point, partially, in the past,” Dykstra says. “Doug Trumbull deserves a lot of credit, because a lot of the people who worked on Star Wars were people that he nurtured. The name Dykstraflex was really a joke. The Dykstraflex is not my camera; it’s not a new concept. It’s been around for years, from the days of the RKO General Pictures [in the 1930s]. The name of the camera is not indicative of its designer so much as it’s indicative of the joke that was going around. There was a Trumbullflex, and the same people that made the Dykstraflex had made the Trumbullflex.”
These same people were Dykstra, Don Trumbull, Richard Alexander, and Al Miller. “When we worked with Doug [at Future General], we had a camera that was similar but not quite as sophisticated,” Dykstra continues. “We had an electronic-control system that was far more sophisticated, but not nearly as useful. We had a mechanical drive system that was also more sophisticated but not nearly as versatile. Those elements were designed by somebody else; we went in and tried to figure out how to use them. Star Wars was our first opportunity to take those devices and make a device that incorporated all of their best aspects, plus any good aspects that we could come up with, and eliminate the problems.
“Al Miller and I sat on the floor in my house in Marina Del Rey, drank two bottles of wine, and came up with the outline for what the functions of the electronics were. At the same time, we figured out how the mechanics would have to work and a lot of the basic parameters for the camera. Then we went into the drawing stage and Bill Shourt and Richard Edlund and Dick Alexander and Don Trumbull—everybody had some input to that device.”
“The entire camera system was designed for Star Wars,” Edlund says. “Any trade-offs that we had to make were in relation to the film that George gave us—the 16mm edit of World War II clips that showed all of the dynamics, the cutting sequence, and the way the ships would move. We knew the kind of shots that he wanted; it was a tremendous help in preventing what could have been an arbitrary mishmash.
“John had been working with Doug Freeze [an engineer in Hollywood] and Don Trumbull a little bit before I got involved,” Edlund adds. “My main desire was to keep the camera as small as possible so that it could get as close as possible to whatever it was photographing, without depth of field becoming a problem—depth of field being the worst problem in miniatures, because if anything is at all soft, it ruins the shot. Everything was going to have to be absolutely crisp. One fuzzy shot in the whole movie and you’ve lost your credibility. So I redesigned all the lenses to some degree, adding extra f-stops to the lenses and things like that. Also, we made the camera head narrower than the camera, so that the models could fly above to the right, above to the left, and you could just miss them on the bottom; the fact that you could do that makes the shots much more exciting. It gives you the illusion that the camera is much smaller than it is—the smaller the camera, the greater the illusion.”
Most of the machine work was done in the warehouse on Van Nuys. They welded the tracks, which took about a month, bought war-surplus drives and stepping motors, and converted equipment. The hardware cost around $60,000.
“I’d say that the major construction took about six months [June to December 1975],” Dykstra says. “We had portions of it functioning before that, but there was a lot of lag time because of the unavailability of electronic parts. It’s a totally modular system, compatible with every other device in the shop. We used consistent plugs and connectors. We used the same power-drive systems. All of the track systems use pipe clamps, which means you can bolt pipes together to make whatever fantastic shape you need, and then disassemble them and re-form them into some other shape. They are very sturdy and give you supports for the ships or for the cameras or any oddball thing, which is real useful. It’s a workhorse device; it’s not a computer and that’s really important. It’s a calculator. It’s nowhere near as sophisticated as we could have made it with microcomputers.”
Sandcrawler
Circa April 5, 1975, McQuarrie revised Cantwell’s sandcrawler design and illustrated the escape of the robots from the vehicle as described in the second draft. The sandcrawler has stalled because of falling rocks from the cliffs, and the Jawas are trying to figure out how to put the track back on the wheels. The vehicle the Jawas use for storing their wares had originally been inspired by photographs of a NASA-designed rover built for exploring the terrain of alien planets (perhaps more like Cantwell’s model). The Jawas themselves are outgrowths of the “shell-dwellers” in Lucas’s THX 1138.
McQuarrie’s illustration with revised sandcrawler.
Cantwell’s model of the sandcrawler.
“It was supposed to be a very large, old, rusty, and tracked vehicle,” McQuarrie says. “You had to climb up all the stairs to get in, and there were a lot of rooms in it, storage places. I envisioned this thing on the front with teeth to be part of a scoop that comes down with a hydraulic arrangement to pick up things, like a garbage truck.”
“Ralph had done a painting of the sandcrawler,” Johnston says. “He had done side views, and it was real long and streamlined. I thought that it could be higher and more awkward and rustier, kind of clumsy looking, so that influenced my redesign of same.”
Johnston’s redesign of the sandcrawler.
* * *
X-Wing
“The X-wing had a box inside of it, with gears made of rubber bands, pipes, and motors, and it was continually evolving,” McCune says. “I wanted a dragster,” Lucas says, “with a long narrow front and a guy sitting on the back. Then Colin came up with the split-wing thing.”
X-wing with surgical tubes for circulating air to cool the engines.
McCune holds an X-wing model by its armature.
“Out in space it had to be able to draw its guns like in a Western,” Cantwell says. “So that was the equivalent.”
Cantwell’s X-wing prototype concept models.
“Cantwell’s was a little bit too sleek and slender,” Johnston says. “In fact he used a dragster body from one of Revel’s 1/16-scale model kits; he just took that and added some engines and a little cockpit over it. Fortunately, Grant, John, and I were able to keep an eye on the model-builders and supervise them all the way through so that if something started getting out of hand, as it did several times on the X-wing, we could step in and handle it. The X-wing went through six or seven body changes. We had a lot of trouble with the nose. From the front, you need that little lip on there; w
e worked with that a long time, getting it into a shape that didn’t look obscene.”
Dave Jones in the model shop.
X-wing redesigns.
Grant McCune masking the in-progress X-wing model in ILM’s spray booth.
* * *
VISTAVISION STRIKES BACK
The VistaVision system had been developed in the 1950s by Paramount and Technicolor, but was short-lived. The idea behind it was to give the public an image whose size and quality absolutely dwarfed the black-and-white gizmo in the living room. “When television came out, the producers scrambled to come up with some way of enticing the people back into the theaters,” Edlund says. “As it turned out, VistaVision got shelved after a while. But a tremendous amount of money had been spent on cameras and equipment, which was just sitting around. A movement [inner-camera mechanism] that had perhaps cost $25,000 or $30,000 to build, we could get for a thousand bucks. So we amassed all of this stuff.”
Not only was the high-end equipment a bargain, VistaVision film was also the ideal format for special effects that were going to involve a considerable amount of duping. Instead of vertical 4-perforation film frames, VistaVision used horizontal 8-perforation frames, literally twice the real estate as standard 35mm film, which would still be used for all the regular live-action photography. “It gave us double the quality up front,” Edlund says.
“It’s like anything else,” Dykstra says. “The more area you use to record the information, the more information is recorded, the more detail. What that boils down to in motion picture terms is: You start with a large 70mm negative, and when you reduce it for 35mm projection, even though some elements have been duped, you end up with very nearly the original quality. You can go through one generation and still come out right about the same place. Plus, VistaVision allowed us to use every single film stock known to Kodak; everybody processes it, so it’s just easier to work with.”
While the motion-control camera was being built and VistaVision had been decided upon, the next step was to find the right optical printer, the job of Robbie Blalack. Having started doing theater at Pomona College, then migrating to nonlinear film and editing at Cal Arts in 1970, Blalack had founded his own two-person optical printing company in a basement after buying a 35mm machine from his school. “I brought my own optical printer to ILM,” Blalack says, “and we went out and found a second printer at Howard Anderson’s. We went to Anderson because nobody was manufacturing the movements, but he had the equipment and he had no use for it. He’s not into VistaVision.”
Anderson’s optical printer hadn’t been used since 1963, according to Blalack, having been built originally for The Ten Commandments in 1956. “That printer got totally reworked from the ground up, really,” he says. “After we bought it, we began a very long, difficult, and interesting period of updating VistaVision as a printing format. We used the existing movements and the existing bases, and built the take-ups; electronic motors were put on it, along with new optics, new lamphouses, new lenses—the whole works. That project lasted a good ten months, possibly even a year. We started with myself and Paul Roth, a friend of mine from Cal Arts, and we just kept adding people as we needed them and testing. There were no experts to help us, so we had the fun—or the nightmare—of doing that.”
From Anderson they also purchased a matte camera, whose parts were eventually recycled into a contact printer by Rich Peccarella. Another Blalack buy was a densitometer, to measure brightness, among other things. Many of the cameras put on the compositor and optical printer were old Bell & Howells from 1925, a testament to their durability. The equipment and upgrades for the optical department were essential in order to process the VistaVision film in an assembly-line fashion.
The last element in the VistaVision equation was a specially built editing table, or “motion-viewing system,” as Dykstra calls it. “We didn’t have the right editing device for VistaVision, so we had to make our own. We figured out how to do it so we could run six pieces of film in it—but it took us six to eight months to generate the technology.”
In the summer and fall of 1975—apart from building the printers, motion-control camera, models, and other complex equipment—ILM personnel also had to negotiate with the unions of Hollywood. “ILM was easy to set up,” Tom Pollock says. “We just rented the building. The tough part was the union problems.”
“The primary conflict was that the unions, when we first started on the project, were upset because we were bringing in people from the outside to do the special effects,” Dykstra says. “And their primary bone of contention [as they described it] was they had people who had been in the industry for years and years and years, who knew what they’re doing—and if we couldn’t find someone amongst their roster of people to satisfy our needs, then we wouldn’t find them anywhere. But I had to find people who were well versed in a variety of things. I interviewed their people and, quite frankly, didn’t find anybody that I wanted. So we continued to develop a shop without the union’s approval—in fact, negotiations were broken off by the union, not by us.”
“Fox was real scared,” Pollock says, “because they don’t like to cause any waves. They wanted us to use their sixty-five-year-old matte painters and miniature people who were over on the lot. But George was absolutely adamant that we wanted to set up our own shop with our own people. That was one of the control things that we’d been fighting about from the beginning.”
Part of Lucas’s independent streak came from necessity, but another part came from experience. “George learned a lot from Francis Ford Coppola,” says Matthew Robbins. “In his histrionic and flamboyant way, Francis would go out and hurl himself against the realities of filmmaking and film distribution, while George very quietly absorbed these lessons and built his own version of an alternate Hollywood. Not only was he very apart from the system, George was a rebel and was grateful to not be in Southern California. The whole culture of the studios and business managers, agents and lawyers, to him, was enormously distasteful and threatening. He felt that the Hollywood enterprise was corrupt. That it wasn’t really about creativity, it was about making money. I think American Zoetrope and a great deal of Lucasfilm was in reaction to Hollywood.”
“This is a business where hustlers come to town to make the quick buck,” says Charles Lippincott. “You lease the car, you lease that house and everything else possible, up to that mistress you may be living with; you make token payments to writers and pay everybody else with promises, and you sit out that year and see if you can make that one big killing. There’s a lotta guys that die here. It’s the valley of the quick kill.”
“George was shrewd. He understood Hollywood immediately,” Willard Huyck observes. “He was the same way in film school. After about two weeks, he said, ‘Oh, I see how it works.’ He figured out how to do his own thing, and it was the same thing in Hollywood. He understood it right away and said, ‘Yeah, I don’t like the way it works, I’m going to move to San Francisco and do my own thing from there.’ ”
When negotiations broke off between the unions and ILM, both parties were content to see how things developed.
Star Destroyer
“Colin Cantwell’s Star Destroyer had antennas and guns sticking out the sides, which wouldn’t have matted well,” Johnston says. So ILM gave it a lower profile and built it about twice as high. “We made it look like a fortress, while his looked more like a naval ship or an aircraft carrier.” After Johnston sketched it and Steve Gawley did orthographic drawings, Dave Beasley and Gawley started on the actual model.
Cantwell’s model of the Star Destroyer.
Gawley’s orthographic drawings.
Beasley and Gawley working on the model.
“Dreadnoughts were sent out [circa] 1907 by Teddy Roosevelt to announce to the rest of the world that this new country was a military power,” Cantwell says. “So you have this fleet of painted white battleships going around. So the shape then came partly from the battleships—and the shape of a pape
r airplane.”
Though shown here, the Star Destroyer model build can be dated to early-spring 1976, based on the fact that images of the full-sized Millennium Falcon, which are pinned to the wall, were taken after its construction at Elstree Studios.
Model maker Dave Beasley.
* * *
Y-Wing
A Joe Johnston revised sketch, which was then sculpted by the model shop, including Dave Beasley, and painted by Johnston. “I described it and we worked on that together,” Lucas says, “but I would say that’s about 70 or 80 percent Ralph’s design.”
Johnston’s revised sketch of the Y-wing.
Dave Beasley
Joe Johnston
“It was supposed to seem like an old, stock space fighter that was just barely flying,” Johnston says. “They’d taken everything off of it.”
The Making of Star Wars (Enhanced Edition) Page 16