Visual Effects 1: Classic Visual Effects

Films have always relied on visual magic, on camera and laboratory tricks, never more than today. Special photographic visual effects existed for a century before the advent of computer-generated imagery, and often play a part in our fondest memories of favorite movies. There were few electronically created effects of any kind before the Eighties, and we’ll eventually get to the CGI era, but first, a guide to the classic processes and trade secrets that made the magic that most of us loved, from Metropolis through 2001, from Inside the Third Reich through Back to the Future. It’s the story of a distinctive twentieth century craft that still has relevance today.

Every history of special effects starts with George Melies, who was a fairground illusionist who brought trick photography to audiences in Paris. Other early films mostly ignored effects, except for a perennial fantasy favorite, ghosts, easy to do with a double exposure. Silent films began using glass shots: painting an elaborate setting on a sheet of glass and filming through it. Simple, but if you’ve ever seen the YouTube clip of Charlie Chaplin roller-skating in a department store, getting “dangerously” close to a “sheer drop”, you’ve seen how good it could look, even back then, if you lined it up right.

The Germans elaborated on the idea with the Schuftan process. Imagine a mirror on front of the camera, at (let’s say) a 45 degree angle. It reflects a detailed miniature set, the size of a model train layout and just off to the side of the camera, of a city in the distant future, with tiny blinking lights and a moving monorail. An expert technician carefully scrapes a rectangular hole in the mirror’s silvering. The camera will film costumed actors straight through that opening. If the exposure and the focus is balanced between the actors on set in front of the camera, and the scale model off to the side of it, you have a finished shot of night club guests seen through the window of a penthouse on the 200^th floor of a 25^th century building. No further work needed; it was all done in the camera. In 1926.

Note that in those days, and for most of film history, there wasn’t a hard-and-fast distinction between in-camera photographic tricks and what we’d now called mechanical, or practical special effects. Many big “effects scenes” combined the approaches for whatever worked on the screen. It’s also useful to remember that for decades, color films were rare. It was easier to do effects in black and white, where you could do simple things like use colored filters to darken or lighten the sky. To this day, Europeans call filming in daytime using blue filters to make it look like nighttime “American Night” (“La Nuit Americaine.”)

When you’re dealing with models of battleships, surfaced submarines, three-masted schooners, just about anything on water that you want to film in a studio water tank, you have a special problem: water doesn’t “miniaturize”. In business-speak, it doesn’t “scale”. You can make a perfect small model of the HMS Bounty, but you can’t make the waves in the water of the tank behave like the ones in the open sea. They’ve tried, using slow motion and mixing the water with gelatin, but it still looks fake, even when they spent the time and money to try to get it right: Tora, Tora, Tora couldn’t make it look good even for $20 million in 1969 dollars.

Alfred Hitchcock used a lot of ingenious tricks and model work in his films (one could even say he had an unfortunate penchant for falling in love with his models), as well as glass shots and rear projection. In the climax of Foreign Correspondent (1940), in the first days of WWII, a German U-boat shoots down a British airliner over the Atlantic. The scene begins with a detailed model of the flying boat-style plane. It dissolves to the inside of the cabin. The set of the cockpit had “windows” made of paper, with back projection of ocean waves racing up to them. At the moment on that film when the wounded craft hits the ocean, Hitch released 500 gallons of on-set water through the “windows”, instantly flooding the plane. It’s a scarily convincing effect even 83 years later.

Two of the most famous effects scenes of Fifties movies also involved water, lots and lots of it. The subject matter of When Worlds Collide is about as grim as it gets: the literal end of the world. But days before rogue planet Bellus slams into our globe, its wayward twin, planet Zyra, will swing by close enough to cause catastrophic worldwide destruction of every major city. Arrivederci, Roma. Big Ben tolls for thee, London. Sayonara, Tokyo. (Indignant Japanese people, accustomed to monster movies: “What? Us? Again??”) But the crowning scene was the impressively realistic tidal wave hitting Times Square—yeah, go laugh it up, you anti-NYC fuzzballs!—which startled 1951 audiences, because that model of midtown Manhattan is just large enough to allow realistic looking flooding.

Flooding, in fact, is the essence of the next special effects triumph of Hollywood in the Fifties, the parting of the Red Sea in The Ten Commandments (1956). For many years, this was regarded the greatest movie magic of all time. The buildup uses every visual tool of golden age Hollywood, Before the big moment even happens, you see the skies darken and swirl. Nothing less than a miracle is about to take place right here on Earth, in front of your eyes. Moses raises his arms, and God causes the parting of the sea, allowing the people of Moses to escape. The memory of that awesome moment echoes a generation later in the climactic minutes of Steven Spielberg’s Close Encounters of the Third Kind as well as his and George Lucas’s Raiders of the Lost Ark.

The basic idea of what made that stunning image in The Ten Commandments was actually pretty simple. Imagine three pools of water side by side, each with their own drains and water pumps, a long skinny one flanked by two large squarish ones. Leave the center one empty. Now, while “overcranking” the camera (filming at high speed so it will be in slow motion when played at normal speed), flood the side pools so the water spills over the walls into the center pool until it’s full. Now run that film in reverse, and what do you get? Walls of water rushing upwards out of the middle until you see the bottom of the trench.

How do you get that shot into reverse? A tool called the optical printer. It looks like a bench lathe. It’s basically a workbench with a movie camera and a small projector facing each other. You’d run the parting of the Red Sea through the projector backwards, and film it forwards. You can make the image bigger or smaller by moving the two closer or farther away. You can use the optical printer to do all sorts of tricks.

There’s a pro wrestling term, “kayfabe.” “Staying in kayfabe” is sticking to your story, not revealing inside secrets to the audience. For a long time, Hollywood avoided disclosing how effects were brought off. As Bagehot once said about monarchy, too close an examination of the realities would be like “allowing sunlight to intrude on magic”, which would have been a vivid and dread metaphor for a studio camera crew, zealously devoted to protecting the Kodak contents of the black “Mickey Mouse ears” on top of a 35mm movie camera from the slightest touch of sunlight.

Then the most influential special effects movie ever made went into production, and after it opened, director Stanley Kubrick allowed American Cinematographer magazine to document how it was made, as the ASC said, “holding back nothing”. It seemed ironic or surprising that of all people, Kubrick, popularly seen as a super-secretive recluse, would be the film magician who finally broke kayfabe.

Why did he do that? My guesses are: professional pride in raising the level of the craft, and confidence that his shots hold up even after you know how they were done.

Thomas Carlyle wrote “Genius is an infinite capacity for taking pains.” Kubrick had that quality in spades. He methodically listed the unique challenges of space movies, and researched previous attempts to solve them on film.

Contrary to what most people thought in 1968 and in the years since, 2001: A Space Odyssey didn’t depend on inventing new processes, but on carefully studying and improving old, even forgotten ones, to the point where Stanley’s way of doing things became the basis for the next twenty-plus years of big-time, special effects-laden movies.

One requirement of a space movie (usually) is being able to depict weightlessness. Despite Kubrick’s dismissing Destination Moon, an Arthur C. Clarke favorite, he didn’t come up with anything new: he still used “flying” spacemen on wire rigs, sets that rotated so it appeared that actors were standing on the ceiling, and the easiest dodge of all, grip shoes (magnetic ones in Destination Moon; Velcro ones in 2001) to establish that there was no need for the actors to float around in that scene.

Two of 2001’s major sets, the airline terminal-like Space Station One, and the centrifuge-like living area of the Jupiter-bound Discovery, are round or are segments of curves that are (supposedly) rotating fast enough to give them a semblance of “gravity.” This was one of the eminently reasonable, but to this date, at least, missed guesses by Kubrick and Clarke. People have stayed in space for more than a year in zero-G; few proposals for long-term spaceflight have suggested anything like the elaborate carousels of 2001.

Kubrick was determined not to have cheesy-looking spacecraft that looked like toys hanging by strings. His MGM film would be a co-release of Cinerama, the IMAX of its day, and 2001 would be showing on some of the largest movie screens in the world. The space ships were going to have to stand up to the closest scrutiny. Kubrick made them among the largest scale models ever made for a film. They were incredibly detailed, using the contents of many plastic model kits to give his ships a look of engineered reality. He’d used a couple of English special effects men on airplane shots for Dr. Strangelove and brought them back, with some new reinforcements, to film the various spacecraft.

One reason that scale models in films look fake is depth of field; large objects that are distant from you look sharp in a way that small ones that are close to you don’t. One way around the problem is “stopping down the lens.” If you want to make something look sharper, squint. That’s what setting a lens at f/96 does. It makes everything supernaturally sharp. Now, a double-decker space station that’s no more than eight feet in diameter can stand up to a super-closeup on a hundred-foot screen.

Of course, that causes another problem: getting a decent exposure at f/96. There’s only so much light you can throw on the model spaceship before it melts, so you use another way. You “undercrank” the camera, so instead of filming 24 frames a second, you’re only filming (say) four frames a second. You can do that, if everything moves in extremely slow motion so when it’s speeded up six times, it looks normal.

And here’s one of the cases where 2001’s refreshed Old Tech paved the way for Star Wars and Close Encounters, Blue Thunder and Back to the Future: motion control. It means being able to automatically repeat every move of a shot exactly, so you can go over the same shot multiple times, adding things in each pass, without blurring the image. It’s what made the famous, iconic opening shot of Star Wars possible.

Motion control existed before Stanley Kubrick rediscovered it, but 2001 set the pattern to this day. Machinists familiar with NC, numerically controlled machine tools, will understand the concept: a long screw drive drags a camera along a track, while a smaller, sideways screw drive mounted under the camera moves it from side to side. Each drive, both forward and sideways, had a Veeder-Root style counter to confirm that the camera motion had reached the end of its cycle. Primitive Seventies-era automation reset the track after each take and used phone landlines to send out a page to a beeper when it was done.

This meant that the camera could make one pass of the “target”, one of the film’s main spaceships, lit to make the image as perfect as possible, then reset the undeveloped film to the first frame to film a second identical pass on the same strip of film, this time only to get the small running lights and activity at the tiny windows. Then the ship model is covered in black tape and unlit as the camera makes yet one more identical pass, this time to get just the subject’s outline so the shot can be combined with others, later.

George Lucas loved 2001 and when it came time to make his own rather substantial mark on the art of special effects movies, as he said in a Rolling Stone interview, when humanity heads for the stars, “We will go in Stanley’s ships, but hopefully with my laser sword at our side”. In the same interview he also conceded, good-naturedly, “If you put his shots and my shots on a light box, yeah, his are better.”

But Lucas was being too modest. He’d learned a lot from Kubrick, and as we all came to know, his own shots were special enough to enchant the world. Large screen film format, large, detailed models, motion-controlled camerawork, deep wraparound sound mix; the whole Space Odyssey visual effects package. Plus Lucas had one crucial tool at his disposal that Kubrick didn’t: thrilling action. You don’t have time to notice the seams here and there in an action movie. You don’t care. To get those exciting shots, your viewpoint has to be more mobile, more flexible. In the final twenty years of film-based movie special effects, the craft would reach the peak of its pre-digital brilliance.

Related articles:

• VFX 2: A Golden Age
• VFX 3: Morphing to Digital