'Mortal Engines' Visual Effects Team Talks About Solutions To Industry-Wide Problems, Bringing Those Moving Cities To Life, And More [Interview]

Mortal Engines director Christian Rivers is a protege of filmmaker Peter Jackson, who co-wrote and co-produced this adaptation with his The Lord of the Rings collaborators. This is a tight-knit group, so it's no surprise that when they needed a company to tackle the visual challenges of a movie about a city that eats other cities, they turned to New Zealand-based Weta Digital, who have consistently proven to be one of the best in the world.

Last week, I sat down with visual effects supervisor Ken McGaugh and animation supervisor Dennis Yoo to talk about how things have changed since The Lord of the Rings, if they're affected by Peak TV, how to potentially fix one of the biggest problem in the VFX industry, and more.

This interview was lightly edited and condensed for clarity and brevity.

Do technological limitations affect the early look of sets and vehicles, or is all of that designed beforehand and you guys are tasked with a way of bringing it to life?

McGaugh: More the latter. A huge aspect of our job is problem solving. We want to be facilitators for the filmmakers and don't want to become obstacles, so often if there's something that's going to have prohibitive cost implications or be very difficult to achieve, we collaborate with the filmmakers and find alternatives. The general rule of thumb is that if they ask you to do something and we think it can be done differently or think there are going to be issues, we can do both: we say, "Here's what you asked for, here are the issues we think are going to happen moving forward, and here's an alternative that's a possible solution." It ends up being collaborative that way.

I'm sure you guys have concept art to use as a guidepost, but there's no real existing movie template for "cities that roam the wasteland eating other cities." Where do you even start on a project like this?

Yoo: Pre-vis. A lot of that is the artists working directly with [director] Christian [Rivers], who is a pre-vis artist himself, so it's pretty amazing working with someone like him where randomly he goes, "Hey, can you do this?" and he gives you all the notes, but then he's like, "Wait, I'll just do that and send you it." And you're like, "OK!" It's pretty awesome.

There seems like a lot of room to hide little easter eggs in the cities.

Yoo: I think if we had enough time, we would have.

McGaugh: One of the biggest challenges on this film was just the amount of work we had to do in the amount of time we had to do it. There were definite easter eggs – one of my compositing supervisors loves putting this picture of a fat kitten into everything he works on. Didn't get a chance to put that in there. Being London, and he and I both used to live and work there, there's an oddly-named chicken and pizza shop in North London, and he wanted to put the sign in there, but never got a chance to.

Yoo: I have a friend and fellow animator, his name is Jesse Lewis Evans, I always try to implement something of him as an easter egg. I dropped him as an easter egg in The Lovely Bones: there's a magazine called Groovy Teen Magazine, and there's a question there that goes "Is Jesse a Girl's Name?" And he hates me for it, ever since. I'm always trying to do something like that, but unfortunately on this movie, I didn't get a chance to.

What kinds of conversations did you have with Stephen Lang about his character? Dennis, I know you primarily worked on the animation for his character, Shrike.

Yoo: I met him on set, and he gave me some insight into what he thought Shrike would be, and then he gave me this amazing e-mail as well, and I had a big read through about what his intent on Shrike was in the performance. But unfortunately, Shrike was constantly evolving. Even Christian's ideas were evolving in a progressive way that what he ended up being was this mechanical being that I thought we successfully did. Mainly because [nobody] can actually walk or move like that. Having Stephen there to guide that in those initial stages was just understanding his performance on set, grabbing whatever he had as reference as much as you could.

I know you used data from his facial expressions, but was there any motion capture used to achieve his physicality?

Yoo: There was zero motion capture on Shrike. But if we thought he was moving a certain way on set, we'd use that as reference. There's a scene where Lang takes a step back after he realizes Hester is [spoiler], and we took that motion as reference of Stephen doing it similarly, but we needed to enhance it to make it feel more robotic and heavy.

Both of you worked on The Lord of the Rings – what's the biggest difference in effects and animation work now as opposed to what you were doing back then with Gollum?

McGaugh: On my side, a lot. Because I deal with a lot of stuff that technology helps with. Unlike animation, where you're still moving keyframes. (laughs)

Yoo: (laughs) Yeah, on my side, I have to say we're still doing the exact same things artistically. The things that are changing is how we implement them. For example, on our animation submissions [to the filmmakers], we used to get away with murder. Because there were no textures because they were too heavy. If we got a plate, we used to shrink everything down. It was very low-res and pixelated. What ended up happening was, when you saw an animation submission at that time, you'd pass it to the client and they'd have to use their imagination more about how it'd look in the final renders. That always causes a problem because when you render it, they always thought it was going to look a little different. We'd have to go back and there'd be this iteration of rendering it, going back, and fixing it. Lately – and I think the whole industry is changing this way – in animation, we're trying to push the boundaries of what we can show in animation and trying to get it as close to the actual renders as possible. Lighting, texturing, shading, effects.

McGaugh: So that barrier between what you present for animation and what's final is shifting closer and closer and closer to final.

Yoo: But the actual procedure is quite similar.

Was there one particular scene in this movie that proved to be more difficult than the rest?

McGaugh: On my side, we divided the work up. I handled mostly the first act in post-production, and anything that was set on London. Luke Miller was the visual effects supervisor who did the middle act, including Airhaven, and Kevin Smith did the final act. Third acts are always the hardest to do. It's the hardest thing for filmmakers to find in the edit, to have a good climax and conclusion to a movie, and they often tend to be the most visual effects-heavy as well, so that whole final act was particularly difficult. Not to mention that there's this whole desire to have something that's really exciting, wraps up all the loose ends, tells the story beats it needs to, but doesn't look derivative, either. That final battle had ideas floated around that could have steered it too much into Star Wars territory, for instance. So they were very conscious to focus on the story that needed to be told and try to do it in as exciting a way as possible.

One of the things I noticed early in the movie is in a scene where Hester is running around on the smaller city that's about to be ingested, and as she runs, elements of the city seem to fold in on itself and compact down. How much of that was visual effects, how much of that was practical on the set?

McGaugh: There's only one thing on set which moved and that was a ticket booth that was on a forklift that lifted up. Through the editing process, that moment got diminished quite a bit. So you can see it in the back of a shot, but it's dwarfed by the big CG. Marcos Batoni, our pre-vis and post-vis supervisor, came up to me later and said that a lot of those shots of the stalls folding up were literally tests he had done a couple of years ago, not meant to be finalized, just examples to get things going. Conversation-starters that ended up becoming shots.

That's cool. Was that on purpose, or did those make it in due to time restrictions?

McGaugh: The filmmakers said, "We need something that does this," and they had this whole back catalogue of things that had already been done for pre-vis and post-vis, and it's like, "Oh, that looks like exactly what we want." There were more refined versions [made] – that was a conversation-starter, and then more work was done to build out what this whole trading cluster is, but those little tests just happened to be what filled the gaps perfectly.

Earlier this year, I spoke with Dan DeLeeuw and Kelly Port about Avengers: Infinity War and I asked them the same question I want to ask you guys now: I've heard a lot of horror stories about studios paying less and less to VFX houses for increasing amounts of work. Do you see a solution to that problem?

McGaugh: I think there are behind the scenes efforts from facilities to try to unify milestones so that all visual effects vendors are delivering the same types of milestones in the same stages of work. Or at least communicating with the same language to facilitate the communication aspect so you don't have to start over with every client. At the end of the day, it means most of our internal efforts are towards efficiency. It's trying to become more productive. What costs more than anything is the amount of iterations that the clients ask us to do. So that's the reason why we were able to be very efficient on this movie is our relationship with the clients – with Wingnut, with Christian, because Christian has actually been one of us – and it brought the number of iterations down dramatically. Also Christian, being a visual effects person, just knew what he wanted. He knew what the stages were, and what was important and what wasn't.

Yoo: He also knew that putting time into this one shot and doing a new iteration of it actually takes time away from other shots. So he knew what kind of cost and how that would actually impact him as well. So him seeing this and going, "This isn't too bad. I want you to keep going." That iteration thing, I wish we could drum that in a lot more. Because it's literally doing another shot, especially when they change certain things.

With the rise of the Peak TV era, do you guys feel the increased pressure of that at Weta? Does it feel like you're taking on bigger workloads because there are more shows than ever before?

McGaugh: Every visual effects facility has their capacity, and you can't take on more than your capacity. Where the pressure comes in is that even though the quality of TV and the quality of the visuals is higher than it's ever been, it's still not feature film budgets.

Yoo: And they're expecting to see that.

McGaugh: Yeah, they're expecting to see feature film quality, but the budgets are never there. Again, that's more pressure for efficiency. So we're definitely feeling that.

Yoo: And there are different aspects, too, like subsidies. Just because a film is made at a different vendor doesn't mean we're less efficient. A lot of times, they're just cheaper because of subsidies rather than anything else.


Mortal Engines arrives in theaters on December 14, 2018.