When visual effects supervisor James E. Price met with Alexander Payne regarding Downsizing—his first film in which visual effects play a major role—Price expressed his desire to work on the film, calling it a dream job. “Why?” Payne inquired. “It doesn’t have a lot of the traditional things maybe associated with visual effects.” But this is precisely the reason why Price had to be involved.
Having worked on blockbusters like The Lord of the Rings and Pacific Rim, Price was hungry for a challenge, and an experience that would be different. His experience on Downsizing demonstrates that while visual effects are often credited only with their utility for big-budget films, effects are applicable to all kinds of storytelling—even a humorous, character-driven story like the kind Payne tends to make.
Below, Price explains Payne’s desire for “spectacularly banal” effects and the “mind-bending” math that went into Downsizing‘s visuals.
What aspects of Downsizing were compelling to you?
I first read the script actually many years ago. I met with Alexander, and I told him at the time that it really was a dream job for visual effects. It’s an opportunity to weave visual effects into a drama, in a way that we don’t normally get to.
One of things that Alexander was very adamant about from the beginning is that the effects be “spectacularly banal.” What he meant by that is that the story was fundamentally a story of a character trying to live his life, and the visual effects were meant to be in support of that. There were a lot of nuances, in terms of the photography, the production design, etcetera, that visual effects had to work with in order for the movie to be told.
The other thing that was really attractive about it is it was funny. It was an opportunity to use visual effects in a way that was humorous and kind of out of the ordinary from the types of things that we might normally do.
What was it like collaborating with Alexander? Was he at all familiar with the visual effects process?
He’d done a few visual effects in the past, but he hadn’t done anything on this scale before. We spent a lot of time working together to understand the process. One of the things that he said was, “I’d like you to try to fool me into thinking that I’m just making a regular movie.”
We did everything we could to present the situations—whether it was digital environments we were creating, or mixed scale, where we had big people talking to small people—in such a way that he could photograph it just like he would photograph anything else.
For the scene where Jason Sudeikis’ character is sitting on the cracker box talking to Matt Damon’s character in the kitchen, we put a doll on top of a cracker box, and Matt played to the doll. Alexander was able to compose it just as if it was a normal scene, and then we painted out the doll and replaced it with a green screen element later. We used any technique we could that would help ground Alexander and the actors in the environment.
We would build partial sets. When we shot the green screen element of Jason’s side for that scene, we built an oversized cracker box and had him sit on it. We knew we weren’t actually going to use the oversized cracker box because it wouldn’t match the detail of the real, small cracker box, but it gave him something to work on and help get him in the mood. In all the situations, we tried to do as much real as we possibly could and then replace what we needed to later.
There must have been an interesting intersection of cinematography, production design and effects on this project.
There are many decades of movies featuring “big and small,” or scale differences, but we found that a lot of the techniques really weren’t applicable. The oversized props didn’t really seem real. It just didn’t fit in with the overall aesthetic and style of the movie, so we knew that we wanted to use as much real as possible. If it couldn’t be done practically, we would do it digitally. We didn’t want to use any oversized props.
With those kinds of rules in mind, the production designer and the cinematographer and I all met and worked out exactly what we needed to build. “Okay, we’re not going to build an oversized prop here, but we need the right bounce light on the character as if they’re standing on the prop.”
In the case of the cracker box, it’s a facsimile of what’s going to be there. We will light with that in mind, but we’ll ultimately end up replacing it. It was a situation where we all collaborated to see how each of us could contribute to the final look, knowing what our parameters were, and what we needed to make it look real.
Were certain items—like the gigantic rose, or the bottle of vodka—done with oversized props?
In the case of the rose there was a prop rose that was made, and Matt Damon’s character did carry it. But we ultimately replaced it because it didn’t have the right kind of light transmission and it didn’t flex in the right way.
As a prop, it was perfect, in the sense that it helped us compose the shots, it helped Matt have a prop to work with, it did the proper bounce light, but it didn’t have a level of realism, so we replaced it with a digital one. We did that in a lot of cases.
The vodka bottle, that was only a trailer shot. It didn’t make it into the movie. It was the same kind of thing. We had special effects build a rig that would deliver the water that looked like vodka. They built a metal frame, so the actors and the extras would know not to walk through the space where the bottle was going to be.
Then, we ended up replacing the bottle. In the case of the trailer shot, to be honest, we had a bottle of Absolut. We held it up in front of this camera and lined it up with the rig on the set and took a quick reference photo of it. That ended up being the element that we used for the trailer.
From your perspective, when are in-camera forced perspective tricks the way to go?
In our case, we knew that everything needed to be totally real. In the case of the actors, we knew we had to do forced perspective. Our characters are five inches high. They’re 1/14th scale, so we would shoot the big side first and then take all the measurements, scale them 14 times bigger, and then shoot a green screen of the person. Because the camera was much farther away, they looked small.
We knew we were going to do that because we weren’t going to be using any digital doubles. Then, our challenge became about matching eye lines and pacing. We did audio playback and placed large facsimiles of the faces, so the actors would have an eye line to look toward. The decision in terms of the actors was very clear.
In other cases, whether it was props or the environments, it was dictated by what we felt would make it most realistic. We weren’t going for anything fanciful or stylized. We wanted reality. Whether it was the cracker box or the rose, we knew that we were going to either photograph a real one and compile it under the green screen element, or, when we couldn’t photograph it, because it was moving too much or the camera was moving, we knew we would do it digitally.
What was the approach when it came to “big-small” scenes? Was it different depending on which perspective you were with at the time?
In looking at other movies and in talking about the aesthetic of Downsizing, Alexander likes wide lenses; he likes to see things in the frame clearly. One of the hallmarks of shooting small things is a shallow depth of field that you get with macro photography.
We wanted to be able to control that so that we could make the characters look small, but we could also control the aesthetic of it, so that the focus didn’t get too shallow, and things would get too hard to understand.
We created this idea of what we called the “big camera” and the “small camera.” A big camera would be a normal movie camera shooting normal-size people. When it was shooting something small, it would have a bit of a macro look. The idea of the small camera was to imagine, when we’re with the small characters, that we were a small movie crew. Therefore, when we were shooting them, they looked normal, but when we’re shooting a big person, they looked huge.
Our idea was: the size of the camera would be dictated by whomever we were emotionally or narratively with. For instance, at the beginning of the movie, when Matt Damon and Kristen Wiig are on the bus and the camera pans over, and we see all of the small people in the travel compartments, that’s a big-camera scene.
The camera is the same size as Matt Damon, and the people in the box just look small. Yet later, when we’re on the other side and he’s downsized, and we’re with the orderlies and looking up at the big nurse, that’s a small camera [shot]. She looks big.
It’s a little mind-bending. From a technical perspective, whenever we had to shoot something small with the big camera—for instance, the shots of Jason Sudeikis’ character on the cracker box—we would shoot a series of focus passes, and also a series of f-stops, so that we had choices for depth of field in post, so that we could aesthetically choose what we wanted those shots to look like.
Given the way in which visual effects have progressed in recent years, would the making of this film have been a much different process a decade ago?
I think the answer to that is yes and no. A lot of the advances that we have made recently with digital characters really wouldn’t apply because that’s not ultimately the solution we wanted. We knew that we weren’t going to use digital characters because we were interested in having the real performances.
One area where I do think that it gives a lot of freedom and flexibility is in the detail level of the environments. There are a lot of techniques that we could use now to combine real photography and base the effects in reality, and then also combine that with purely digital work. We knew that Leisureland was going to be a massive build. With the way that the technology’s advanced, we knew we would be able to handle the level of detail required. I think that was very important.
That being said, we used a lot of old school techniques. Alexander’s style really worked well with these types of visual effects. He doesn’t move the camera a lot. He doesn’t cut a lot, so we were able to recreate these setups. We had locked off cameras a lot of times for the green screen work, which is very helpful. But the margin of error was incredibly low. We didn’t have the luxury of knowing we could recreate a character digitally in post, or do a CG shot for coverage that we might decide we need, so we knew we had to get everything right on set.