Jay Redd Talks Men in Black 3 and Looney Tunes 3-D
DS: You mentioned previs. What type of previs was done on the movie, what were you involved with, and how did it end up supporting the final film?
JR: Early on in previs, Spencer Cook, who is our animation director, went out to New York and started doing some exploration on what we termed the “Fish Fight,” which is this large alien fish at Woo’s Chinese Restaurant. [There was] a lot of early character exploration there. As the movie started to blossom, and we started getting whole cut scenes of the script storyboarded, we teamed up with The Third Floor, and Chris Edwards’ team there, to really previs big sequences including the “Monocycle Chase,” and the “Time Jump” off the Chrysler building. There’s a sequence at Shea Stadium and there’s a sequence at the end of the film at Cape Canaveral. It was really helpful for Barry, and for all of us, to help determine the pacing, and [get] an early view in editorial of how we were going to shoot all this stuff. It was key to say which shots we cared about, which shots we didn’t care about, things we thought would last all the way through the picture, other shots, just angles to explore. It really helped inform us of our workflow as well.
DS: Can you tell us about some of the tools that you used in the film, in regards to lighting, compositing, some of the technologies in your pipeline?
JR: While we’re on set, to get the best version of lighting - at least, that was shot – we used a mix of proprietary technology and a tool called the Sphereon. It’s a 360 degree, high dynamic range, [almost a] slit-scan photographic system that captures up to 11 stops of dynamic range. For example, in this room [indicates room of interview], if this thing were moving, [it would capture] very long exposures at high resolution. It gives us a dynamic range of where all the lights are in a scene. So we can take that into our image-based lighting system at Imageworks, render it using Arnold, our ray-tracer, and that allows us to get our digital characters in the scene rapidly. We can look at it and say, “Wow, that matches, the highlights are all working, [etc.]” Then from that stage we can get into the dramatic lighting on top of that. We did that for every major sequence on the film.
Once we’re at Imageworks, we’re using Arnold to render. Most of our animation is done inside Maya. We also use Houdini for our effects simulations, and Katana, which is our compositing and shot management software, to bring all the pieces together.
DS: Were there any technological advances that came out of your work on this film that made the project any easier, or any more efficient, that you potentially will use moving forward?
JR: That’s an interesting question because on this Men in Black film, again, we want to stay true to the first two. There have been a lot of advances in technology [since the last MiB film]. The way that we capture lighting information on set was really the big one for us. Everywhere we went we [used the] Sphereon [to capture] all the lighting information from scene to scene to drop in our characters pretty quickly. But [there was] a lot of mixing of old school and new school techniques. When I say “new school,” I mean modern, like image-based lighting and high-dynamic-range rendering. But a lot of stuff we were doing was still rotoscoping and match moves. This [movie] was shot on film, so it’s a really interesting balance of old school and new school techniques to make it feel like it’s in the Men in Black world. As far as huge technological advancements, I would say, really, the way we treated our image-based lighting.
DS: Let’s take a step back with regards to the process of making this film and let’s talk a little bit about the character design and the visual look. Certainly, the Men in Black films are known for their interesting characters and creatures, the morphing from human to creature, and back. Tell us a little about the process of visual development on the film.