8 Movie Special Effects You Didn't Know Weren't CGI: Classic

It was a chase that destroyed the Batmobile, a Joker-themed semi, a dump truck, a paddy wagon, multiple cop cars and God knows how many bystander vehicles. And that shit was worth it.

Two specific points during the insane car chase at the halfway point of The Dark Knight are so over-the-top they seem like they would have had to be computer-generated, if for no other reason than they would have killed the stunt drivers.

The first is when the Batmobile first shows up to take out the Joker's convoy. It speeds in ...

... and offers what equates to a vehicular uppercut to a garbage truck.

The scene ended up in the trailer and inadvertently encouraged a bloat of fanboys to flock to their keyboards and pound out protests against the fake-looking CGI in the movie. But as Christopher Nolan has proved time and again, he doesn't mess with that shit if he doesn't have to.

No, what you are actually seeing there is a complete one-third-scale model of the Batmobile, the garbage truck and a large section of lower Wacker Drive in Chicago.

Even the damn 180-degree move that the Batmobile pulls off at the end was done by a radio-controlled model.

So what about the climactic moment in that scene when they flip the Joker's 18-wheeler after Batman clotheslines it with a grappling hook? If that was a model, it was pretty goddamned convincing.

The flipping of the semi was accomplished with a technique known in Hollywood as flipping a real goddamned semi. To get the mind-boggling amount of upward force needed to lift the big bastard head over heels, the FX crew built a huge steam-piston mechanism in the trailer.

Of course, then the challenge was to make it look like this insane stunt was occurring right in the middle of the banking district in Gotham. So how the hell do you pull that off? Build a miniature city and edit in the truck somehow? Film the truck out on an open course and use CGI to fill in the background? Shit, no! They just went to downtown Chicago, closed off a street and flipped their goddamned semi.

Why? Because that's how Batman would do it.

And even if you watch it today, there's something oddly realistic about it, especially when compared with more recent Roland Emmerich stuff like 2012, which made the destruction of L.A. look like a very expensive video game cut-scene:

The difference, of course, is that the fire in the streets in ID4 is not CGI. It's real fire.

That sort of thing isn't easy to do in real life -- after all, how do you make the fire go sideways? Fire doesn't normally plume horizontally, which is a good thing most of the time, but the whole point of the aliens' city-destroying weapon was the unearthly way the blaze would slowly spill outward and engulf the city.

For the effects team, the solution to this shot was relatively simple.

Can you tell what you're looking at? That's a model city on its side. You'd probably recognize it better like this:

They called it the death chimney. Just turn the city model sideways, put the pyrotechnics at the bottom and put the camera at the top. Then they shot the explosion at a high shutter speed so that when the film was slowed down, they got their horrific, creeping wall of unstoppable fire.

See, this is how destruction used to be filmed back in the day. Someone spent weeks making a detailed model of New York, then you set it on fire and hoped to hell you didn't screw it up so they'd have to build it all over again (because then the model builder would find you in the parking lot and beat your ass).

Not if you don't want it to look like shit. It's one thing if the actor is just standing next to the normal-size characters in a field, but at various points in the trilogy, you see Gandalf grab the tiny Frodo, hug him, ride on the same carriage with him and sit down at the same table. To pull that off Peter Jackson, needed a buffet of effects techniques ranging from simple to insane.

Sometimes it was as easy as using a child in a Frodo wig shot from behind ...

... or just compositing the actors together from different shots, or digitally sticking Frodo's face onto a tiny double. But the coolest effects didn't involve any computers or green-screen trickery at all. It's called "forced perspective."

The idea is that you put one actor really far from the camera and the other one really close to the camera, then shoot at such an angle that it appears they are next to each other and that one of them is really big and the other really small. Which sounds simple, until you realize that you need to build everything on the set so that the actors can interact with it at the same time while hiding the fact that they're far away from each other.

The simplest example is with Gandalf's cart. In the movie, you see them sitting side by side ...

... but the real cart is built so that if the camera is stuck in that spot, it hides the fact that Frodo is actually sitting about four feet behind Gandalf, with Ian McKellen's body hiding where the bench is split:

But the complication comes when you realize that this works only if the camera remains perfectly still. So any shot where the camera moves around has to involve a computer, right? Nope. In scenes like this one, where they share a table ...

... they are actually sitting at two different tables, one human-sized and one hobbit-sized ...

... that are made in such a way that each piece slowly turns with the camera, so that the whole time, they appear to be one simple table, shifting with the perspective of the viewer. This required that the camera be put on a motion-control rig and half of the set be put on another rig that completely counteracted the movement of the camera. So when the shot moved, the set, props and even the actors moved accordingly (yes, while McKellen was trying to stay in character as Gandalf, he was on a stool that was slowly scooting him around the room).

Take a moment to think about the crew that put that together, knowing the final goal was for you to never notice it.

Maybe the best example of the mind-boggling ingenuity that goes into any "I can't believe it's not CGI" scene is one that was cut from this film (a scene which, as we explained in the past, fills in a pretty big plot hole). The scene shows Sarah and John Connor opening up the Terminator's head to extract and reboot his CPU, giving him the ability to learn and adapt easier.

Here's the scene.

In a single take, we see the back of the Terminator's head in the foreground and his face reflected in the mirror in the background, clearly showing Arnold's un-animatronic likeness talking and emoting.

As they unscrew his dome, the camera moves around until the shot finally ends on the top of the non-reflected Terminator head, which is opened up like a damn pickle jar.

Now those of you who are experts in Arnold Schwarzenegger trivia already know that the man does not have a giant hole in his brain. So, what, they just had a fake head in the foreground and used CGI to put a fake reflection in the mirror? Easy!

Uh, no. There is no mirror. It's a window. On opposite sides are two John and Sarah Connors, and two Terminators miming each other's movements exactly so that they would appear to be reflected in a mirror.

But wait -- something doesn't quite add up. That's clearly Sarah Connor in both the foreground and the reflection ... what gives? It's not like there are two Linda Hamiltons out there.

Meet Leslie Hamilton, Linda's identical twin sister. She came in quite handy during the making of this film, especially when the T-1000 took on Sarah Connor's form.

Yep, for all that you heard about the groundbreaking CGI, it all came down to identical-twin shenanigans. Basically, a special effects version of The Parent Trap.

The big-budget project was to have a stylized, surreal look in every frame. The studio, going through a CG craze, dropped its top visual effects artists in the director's lap so they could paint everything in pixels and Phantom Menace that shit.

In response, Coppola fired every one of them and replaced them with his 29-year-old son, Roman. The result is a movie with effects that were 100 percent done "in camera." That is, what you see is what they shot. It doesn't sound that impressive at first, but then you start looking at the kind of shots they needed to get. What on the page was to be as simple as Reeves taking a train ride, wound up looking like this in the movie:

That's a shot out the window of the train where we see Gary Oldman's stare, hovering in the clouds for some reason. To get that, they actually filmed a model landscape moving with the eyes projected on it, then projected the whole thing in back of Reeves sitting in the train. It's a projection of a projection on a projection. And that was a piece of cake compared with this:

Normally, this book-train montage shot would be a cinch: You shoot your train, shoot your book and put it all together in the computer.

Fuck that! To get that shot, they actually built a model train and a gigantic book. Then they filmed it. What you see is what was actually there.

For example, at one point Carrey's character, Joel, is having a fight with his girlfriend (Kate Winslet), who walks angrily into the bathroom only to completely disappear and transport into the kitchen, then transport to the front door before leaving -- all within the same shot.

Later, Joel walks in on ... himself, talking with the doctor who is later responsible for the memory wipe.

In a blink, the shot then turns from Joel to the doctor ...

... and then to another Joel, the Joel who is in the memory itself.

This keeps happening, back and forth, as the scene unfolds.

Once more, this doesn't seem like a big deal if you can just make a real Jim Carrey interact with a CGI Jim Carrey. This method is probably how the shots would have been done for this scene if Eternal Sunshine of the Spotless Mind had been directed by someone less insane than Michel Gondry.

Instead, Gondry relied on his theater background and insisted on pulling off the illusion using a technique known as "making his actors run and change costume really, really fast."

No, really. For that scene where Kate Winslet seems to inexplicably appear in two separate rooms, they put a trap door in the bathroom and had her book it over to the kitchen before the camera got there -- then used a double as she left through the door.

The double Jim Carrey shot was actually much more excruciating. To create two Joels, Carrey would change his wardrobe and demeanor whenever the camera panned away from him and run to the other side of the set to play the other part as quickly as possible. It took so many takes to accomplish that Gondry and Carrey actually had an on-set argument about it because the actor didn't think it could be physically done. The strange thing is, we're pretty sure we still agree with Carrey there.

Making an actor fly or float or do Matrix-style kung fu is the oldest Hollywood trick in the book. Step 1, hang the actor from cables. Step 2, remove the cables from the finished shot, which these days can be done digitally.

So when they needed the actors to float around in the zero gravity of space in Apollo 13, it seemed pretty simple. Either do the wire trick, or hell, just composite in the actors with CGI entirely.

But NASA stepped in and said, "Look, we put dudes on the moon. Did you read about that? We can do anything. And what you need is to just get rid of this whole gravity thing."

Enter a special craft they affectionately call "The Vomit Comet." It's NASA's own KC-135 airplane designed to do one thing and one thing alone: Create a zero-G environment right here on Earth.

To accomplish this, the plane does a series of parabolic arcs, which is a fancy way of saying that it goes up and down really fast.

This action causes a brief window of complete weightlessness for anybody lucky enough to be on board. It's used as a training program for astronauts, but for Apollo 13, it was turned into the soundstage.

It took a mind-numbing 600 or so arcs to complete all the shooting. So when you watch that film again, concentrate on the faces of the actors during those shots of weightlessness and note that their looks of pants-shitting excitement are completely genuine.

But before you go envying them too much, remember the plane's "Vomit Comet" nickname. Look at the high-speed dives and climbs of the aircraft and imagine you're inside it, and three or four burritos are inside you. It ain't pretty.

You're probably wondering how anything from a film made in 1981 could possibly be confused as CGI. Well, they did have computer-generated images back then; they just looked terrible. Like this shot, which is supposed to be from Snake's glider's computer as he is descending on the futuristic apocalyptic cityscape that is 1997 New York City:

See? 80s graphics, man.

Believe it or not, what you see there was extremely cutting-edge for the time. In fact, it was so cutting-edge that it was completely out of the question given the film's budget. But John Carpenter wanted this high-tech graphic to appear in the film -- after all, it's supposed to be 1997! So they had to find a way to do that shot of computer graphics without using computer graphics.

So they grabbed their model of New York, which had been used for various other shots, and bought a roll of green tape and a black light. That's it -- this cutting-edge effect was done with five bucks and a trip to the hardware store.

David Bell is a freelance writer and video editor. You can read some of his work here.

For more on special effects, check out The 5 Miserable VFX Jobs That Make Movies Possible and 6 Hilarious Special Effects From Turkish Cinema.

And stop by Linkstorm because it will help you forget that today is Monday.

And don't forget to follow us on Facebook and Twitter to get sexy, sexy jokes sent straight to your news feed.