Can you give me an impression of the procedures and tools that you, Russell Carpenter and Jimmy Muro used while shooting on the ship set?
Cameron: Because of the scale of that big ship set, with the lighting instruments it required, every scene had to be very carefully worked out. The Vid Stick was very handy in terms of pre-visualizing shots, but it just wasn't a situation where we could come in and say, 'Hmmm, I think it would be nice if we did it this way,' and then turn the shot over to Russell to light it. He would have had 10 hours of lighting to do. We had to anticipate what we wanted to do well in advance. Shooting on the ship went very slowly at first, but after a few days we learned the ground rules of that set and how we had to place the big lighting instruments like the Musco, the Night Sun and the balloons so they didn't reflect in the water and create dappled lighting all over the sides of the ship. The end of every shooting day would be a briefing on the next lighting positions for the big, hard-to-move fixtures. This is a standard process.
A Musco can move pretty fast, and on a standard shoot you can change its position in 20 minutes. But if we had to move balloons, towers, Muscos and a Night Sun, we could be looking at four to five hours of work, and that could kill us. Fortunately, we had spent a lot of time previsualizing the situation and had actually designed the tank facility with the aid of a scale tabletop model, with a sheet of Mylar standing in as the water surface. We created a berm around the perimeter of the tank model, and Russell cooked up what was essentially a miniature Musco that was in scale to the ship in terms of height and intensity. From there, we figured out what the lighting would look like, and picked out specific places on the berm from which we could light things.
Using a bounce light hanging from a tower crane was an idea that came directly from those tests on the model. I knew that if we did not have complete flexibility in the lighting, and if I couldn't literally walk from one end of the ship to the other and do a scene along 400' of deck in the latter half of the day, there was no way that I was going to be able complete the film on schedule. That resulted in the development of a furling system on the crane arm.
As the shoot evolved, the lighting balloons we used for this purpose turned out to be a nicer light that withstood wind conditions better although they were harder and slower to move while the tower crane turned out to be an ever-more-powerful tool for placing the camera. Working with a set of that size and that height proved to be remarkably difficult. We knew that going in, but it took us a surprisingly long time to come up with the tower crane as a solution. We had first turned to the Akela crane. We went out and did a test with an Akela at one of the studios in Culver City, and it filled the street. We were like, 'Wow, this should do it!' But when we got it out next to the ship set, it was a joke. We then thought we could put the Akela on a 30' platform built on barges, which we could float right up to the set. That worked, but the real breakthrough in getting angles on the ship didn't happen until we figured out that we could hang a gyrostabilized camera from the basket on the crane arm, which allowed us to put a camera anywhere on the entire set in a matter of minutes. We couldn't have shot the picture without it.
Our approach to the film was to try to present, in human terms, an event that is far outside human scale. Therefore, there were only a few times when I wanted to do the big wide shots of crowds on the decks to show the greater story that was happening. I instead tried to start on a wide shot, come sweeping in on a recognizable character, and then punch in closer. That way, we'd always have a human connection. I also thought it was important to know where people were geographically, so a lot of the wide shots evolved into closer shots. This, of course, was a photographic approach that hadn't been used on any previous Titanic films, because they didn't have a big enough set or the visual effects to back them up. The shots that interested me the most were the ones where you saw the ship and understood its size and then came closer for a specific piece of dramatic action.
This idea was used most importantly for a scene in which Jack and Rose are embracing at the bow of the ship. They start out as two tiny dots only a few pixels high, relative to this monstrous ship coming toward the camera, and we sweep right up on them into a pretty tight shot.
Can you describe the early screen tests you did with Russell Carpenter during preproduction to develop the period atmosphere?
Cameron: We did some tests, and they were very promising, but they were really ridden in on the back of auditioning some of the actors. They weren't formal photographic tests. Part of the testing was my insecurity over not having done a period film before, and wanting to see if I could work with a given actor to create a sense of the time period. I wanted to test that on film.
We tried flashing and overt diffusion and all of the other classic techniques for suggesting period, but they all struck me as barriers that would stop the audience from directly experiencing the events in the story. We really got into our true testing phase for this during the two weeks before starting principal photography [at Baja Studios]. During that time, we started experimenting with more subtle techniques, like light filtration. A half ProMist ended up being the heaviest filter we ever used; a quarter was more standard. We also used very light smoke, just to soften the backgrounds and the lighting. We wanted rich color saturations and strong color separation between the warm and cold tones. That way, if we wanted to, we could warm all of the lights by adding a couple of points of yellow and some red in the printing; the blues don't disappear. We coordinated these later camera tests again with wardrobe tests, which was also a sneaky way of figuring out how to light the actors.
Another thing Russ came up with was the idea of filling at a slightly cooler temperature than the key. You can't have warm without cool your eye just gets acclimated to a lot of yellow and doesn't appreciate it as warm anymore. Then you have to dump in even more warmth to make it stand out, and all of the beautiful colors in your wardrobe disappear. Russ was filling with quarter-blue or even half-blue; that way, we could time the whole thing over so the fill would no longer be perceived as blue, and the highlights would suddenly get this beautiful golden feel. I think it looks amazingly rich and has a lot of depth.
On our last couple projects, Russ and I have really done a lot with controlled overexposing to enrich the blacks. Doing this is like 'safe sex' for digital effects work you never know when a piece of negative will become a piece of digital internegative. These days, you should be prepared to do effects work on every single shot. The more information you have on the negative, the better.
This leads directly into the subject of Super 35. You're certainly one of the format's strongest advocates.
Cameron: It's a God-given format, and now that we're in the digital age although not everybody is yet you can reposition a shot, axially stabilize a shot if it has a little bit of roll, and so on. It's like shooting in VistaVision while being able to use lightweight sound cameras. Granted, VistaVision simply gives you a lot more negative area, but most of the problems with framing in the 'scope aspect ratio are vertical. Having additional vertical frame material available means having the option to reposition shots and even do moves within shots. Sometime an actor's head dips down in frame slightly, but with Super 35 you can chase him and reframe the shot, either optically or digitally. We probably have 100 shots in Titanic that were trimmed vertically after the fact.
Any argument against Super 35 went away a few years ago when Kodak came our with their T-grain intermediate stocks. I defy people to compare it to anamorphic, which has its own problems. In anamorphic, you're going through an anamorphic lens when you're shooting and then going back out an anamorphic lens when you're projecting. Each anamorphic stage causes lateral smearing of the image, which leaves lighting artifacts that I don't like. Whenever you have lens flares in anamorphic, you get these straight lines all the way across the frame. Some people like those, but I think they're ridiculous and don't correspond to anything in the way the human eye sees things. People associate that effect with 'big movies,' but that's just because for a long time big movies were all shot in anamorphic. So some people think a picture has that 'big movie' look because the headlights on a car create these white lines that go from one edge of the frame to the other. That doesn't correspond to anything I see in my daily life.
Anyway, that's a stylistic choice. The compelling factor is that the intermediate stocks are so good today that the loss of O-neg area is relatively inconsequential. In Super 35, you are losing negative height vertically, but you're gaining horizontally. You're not gaining as much in anamorphic as most people think. It's not the difference between half-aperture and full-aperture, which is an oversimplification. I don't know the exact figures on [image area loss], but if you follow the rules of good, clean hygienic photography and printing in Super 35, you'll end up with something that in the mind of the audience is indistinguishable from anamorphic or better. Super 35 just lacks certain anamorphic artifacts that I think are detriments, and I much prefer that look.
[ continued on page 3 ]