Recalling a technique he had employed while shooting
second unit for the hoverboard sequence in Back to the Future
II, Burgess had key grip Michael Coo mount a trailer hitch to
the back of a Titan crane, to which they attached a long low-boy
process trailer. An extension arm added to the Titan brought the
crane end out about 20' beyond the chassis and square out over the
trailer. A mock-up hook was attached to the end of the Titan arm
extension, and Schwarzenegger was wired onto that apparatus, with
the low-boy trailer just a few feet below him. Cameras were placed
on the trailer, along with the necessary crew and lighting, and the
Titan crane drove down the street, pulling the trailer and actor
to get the shots of Schwarzenegger hanging on for a wild ride. "It's
really just a matter of experience - somewhere down the road you've
done something similar, and you can apply that experience to this
new problem and see if it works [again]," the cinematographer
observes.
During prep, Burgess "pushed the company to test as many of
the wardrobe, hairstyles and props as they could possibly get ready.
We must have tested 25 different motorcycle helmets alone, just to
find the right one. You have to see that stuff on film in order to
evaluate it properly. I always like to get the actors standing next
to each other on film and test the wardrobe, makeup and hair all
at the same time. That way, you get everybody thinking ahead of time.
If you wait for principal photography to make those decisions, you'll
wind up behind the 8-ball.
"We tested some rear-screen projection using high-definition
video playback," he continues, "and although it worked
fairly well, it wasn't up to the level I wanted those shots to have,
so we rejected the idea. Throughout prep, I carry my Palm-Pilot with
me, and every time something new comes up that I think we should
test, I jot it down. When the test day arrives, I try to knock off
as many things as possible in one session and nail down as many ideas
as possible before we have to begin committing resources."
In addition to choreographing the details of tricky action sequences,
early decisions had to be made about the details of CG imagery that
wouldn't be completed until after principal photography had wrapped. "If
there's going to be a CG element that creates some form of lighting,
it's best to have that interactive light source actually photographed
with the actors, instead of trying to create it all later in post," says
Burgess. "Using a practical interactive source gives the actors
a much better feel for how to react to what isn't really there. It
also helps the director better visualize the timing of the CG element,
and it helps the visual-effects artists know what the director and
the cinematographer decided on during production. It takes a conscious
effort to say in prep, 'Okay, when Kristanna's arm changes into this
device, it's going to light up. What color will it be? How bright
will it be? Will it pulse or flash?' You have to ask and answer those
questions before principal photography begins. I've found that the
more interactive light we can actually shoot, the better off we're
going to be. That way we know immediately whether it's right, and
we can give the CG guys a real thread to work from when they add
their magic."
Although Burgess is a fan of anamorphic cinematography, he decided
to shoot T3 in Super 35mm and do the anamorphic blowup via
a digital intermediate (DI). "Because this was a big action
picture that would require lots of cameras and lenses, it didn't
make sense to go anamorphic," he explains. "This is the
first time I've shot a film that was set to go through a complete
DI, and it's been a very interesting process. The DI certainly gives
you more control over the image and more ability to tie all the elements
together, because you have specific control over any area of the
frame. The flip side, however, is that you spend a lot more time
doing it. It takes triple or quadruple the amount of time to supervise
the process, so it's a bit of a double-edged sword. There also seem
to be more people with their hands in the pie; it's not just you
and the colorist, and that can get tricky. You have more toys to
work with to refine the picture, but you're spending a lot more time
and working with a lot more voices to get it done."
Burgess filmed most of T3 on Kodak Vision 200T 5274 - even
some night exteriors. "I like that stock a lot. It has a great
grain structure, very good blacks and great contrast. For a film
like T3, contrast is very important. I used 5274 at night
to ensure that we'd have good, rich blacks, and it forced me to light
more dramatically than I would have if I'd used [Vision 500T] 5279.
The 500 sees more into the shadows and is a lot grainer than the
200; it doesn't reproduce the colors as sharply and crisply. The
200 gave me a much stronger, bolder look that better served the material."
For one night exterior, Burgess used Vision 250D 5246. "During
the opening of the film, we see Connor living life beneath society's
radar, sleeping along the riverways of Los Angeles. I wanted the
foreground to have a warm, fiery feel and the background to be colder,
but not extremely cold. The 250 is a contrasty stock with good blacks,
and it allowed me to use uncorrected HMI sources in the background,
and get a warm look in the foreground with a mix of corrected and
uncorrected tungsten sources."
Burgess shoots most of his nighttime photography at around T2.8. "It
gives me good depth of field, especially with the wider lenses I
tend to use, and it gives me good resolution in the highlights. A
T2.8 at 200 [ISO] is a much different look than, say, a T1.4 at 500
[ISO]; that bolder, more contrasty look is what I was going for on
this film."
Page
2
|