Search form

Wētā FX Fashions Creatures Great and Small for ‘Godzilla x Kong: The New Empire’

VFX Producer Kevin Sherwood and Animation Supervisor Ludovic Chailloleau talk Shimo, Skar King, One-Eye, Suko, and the myriad environments of Hollow Earth created for the latest Warner Bros./Legendary Monsterverse outing.

For a couple of behemoths who first saw the light of day in 1933 and 1954, respectively, (King) Kong and Godzilla are holding up pretty well. Which is a good thing, since in Godzilla x Kong: The New Empire, the sequel to 2021’s Godzilla vs. Kong, and the fifth film in the Warner Bros./Legendary Monsterverse franchise, the big guys have a lot to contend with – i.e., a colossal danger hidden within our world that threatens the existence not only  of their species, but our own.

In creating both the environments of Hollow Earth, where much of the mayhem takes place, and many of the inhabitants wreaking said mayhem, leading New Zealand-based visual effects studio Wētā FX also had a lot to contend with. Their critical contributions included building most of the new main characters from scratch, developing a dozen or so secondary characters, and creating five large-scale, full-CG environments, also from scratch, each with a unique look to help communicate the journey the characters take through Hollow Earth.

We spoke with VFX Producer Kevin Sherwood, who also worked on Godzilla vs. Kong, and Animation Supervisor Ludovic Chailloleau about their monstrous (sorry) undertaking.

Dan Sarto: Wētā FX’s work on the film was in the Hollow Earth segment, which comprises the second act and the beginning of the third act. You created multiple environments, as well as a number of characters and creatures. As I understand it, almost everything you did was fully CG. But, were there any plates or live-action shots to help with lighting or color or other details?

Kevin Sherwood: Not really. Out of 417 shots, I think nine of them were plates and, honestly, all the plates we had were in the zero-gravity fight inside the spaceship looking out at our environment. So basically, just putting stuff outside the windows. They used a lot of really detuned funky lenses on the plate photography. So, we had to take some of that footage and kind of match it in compositing so that we could put that treatment over the top of some of our work – just so that it matched. But as far as the lighting, no, we were pretty much on our own there.

DS: What were some of the highlights of the environment work from your standpoint, from the world-building standpoint and from the standpoint of new challenges? Walk me through some of the highlights.

KS: For me, the nicest thing was having a baseline from the [2021] Godzilla vs. Kong movie that we could riff off of. Because there's so much work. We worked on this movie for about 56 weeks, I think, and about half of that we spent just building assets – the creatures and environments – and getting to the point where we could start doing shots. There's so much to build because there are so many environments. And the nice thing was that we didn’t have to try to figure out what it's supposed to look like, because we already had Godzilla vs. Kong to guide us.

So, it was more like, how can we take that to the next level and what can we do that's cool that fits in that world? Because we've already built that world. And it lets you focus your effort on the cool part of the work and not the building part of the work.

DS: Apart from taking the existing stuff to new places, did you also have to improve the quality because it would be viewed in higher resolution? Did you have to make things richer, or deeper, or more textured? Did the fur have to be furrier?

KS: There was a little bit of that. The resolution has been stealthily creeping up for the last 10 or 15 years... We did the first Avatar at 2K resolution, and this movie we were essentially rendering almost at 5K in order to match the weird lenses that they used on set. They were anamorphic lenses, which pull the corners in, which meant you had to render a lot of extra resolution.

So, there was some of that – just making sure that everything you build is going to hold up at the scale you're going to see it on the screen. I think for us it's more about the volume. You look at that frame and it's like every rock and every pebble – somebody had to make that and somebody had to put it where it's supposed to be and decide how the light needed to be on it and how much fog is in the shot. There's just so much work, and it takes a long time.

DS: Scale is always a significant concern with regard to big creatures and big environments. I know from talking to other folks who do what you do that you have to cheat things, because if the creatures moved the way they really would, no one would be interested in looking at it. What did you do to ensure that the big characters moved and interacted in a way that's entertaining and art-directable?

KS: The filmmakers came to us at the very beginning and Adam Wingard, the director, told us that they wanted to bring the camera up and treat the characters like actors, and allow them to move quickly and have fast, exciting battles, and not worry quite so much about scale and relative speed. Because, as you say, the second you do that, you can't have a fight between Kong and Skar King – it just doesn’t work. So I think they did that a little bit on purpose. They kept all those beats of the kaiju in Hollow Earth – where it’s just the monsters – so that we could speed things up and keep it interesting and fast and not stress the scale so much.

But of course, the second that they step on the ground and kick up dust, or they kick some rocks, or there’s a waterfall or splashes in a lake, it’s a problem. Human beings intuitively understand how the physics of the world work. And that was where we always ran into trouble because that stuff has to work at the right speed and scale or it looks completely wrong. There's definitely an uncanny valley for physical effects.

So that was where all the cheating happened. We lifted all the water tech from Avatar, but we still had to dial gravity down because, as soon as you get Kong and that monster fighting, all the velocities and the forces are so insanely crazy that the water never does anything you want and it's not art-directable. So, it was like we definitely got in there and tinkered around with the physics of the universe for those shots, and tried to slow it all down to where it still looks interesting, but it’s at least a little controllable.

DS: Moving to the animation side of things, you guys keep improving the technology to do a better and better job of capturing performance. It's still, in essence, hand animated – you're animating from a performance, you're not just capturing something and sticking it in. How did you approach the animation? How much of your existing pipeline were you able to use, and how much did you have to change – if anything – for a show like this?

Ludovic Chailloleau: In terms of tools, it was pretty much the same tools that we usually have. We didn't want to go with the whole tech from Avatar because it would require too much preparation. So we went straight to the more solid pipeline that we have for animation. And the main thing for me was to really work out with the actors a key performance. I went through the whole movie and defined some key performances to direct them in a way that would give consistency to each character in the movie.

That being said, for the apes, we went with normal scales. Apes can have very sharp movements with their arms as they're moving. And that sharpness is not something that works at a large size. So, we were very cautious about the physics, and we slowed it down just a touch, just enough to lose all the sharp angles. The capture gave a very, very good base. Then we went in and key framed what had to be key framed, taking off the things that were killing the weight. And the final thing was to make sure that the camera always kept some parallax.  

For me, those were the key things on this show. I think this was the only time I really worked with a system for the camera that was the same for every single shot. So once we dived into CG shots, there is a little bit of this handheld feeling on the camera, which gives parallax, which helps to sell the distance, so it's not looking like 2D. The camera is never pivoting with no motion. There is always something which brings some foreground, mid-ground and backgrounds to help to break up with the scale of the textures, which is there to sell the distances.

DS: You did quite a bit of design work on your scenes. Often you guys will get assets that come from the director's team directly, or maybe from another visual effects house, and you have to make that work. Is it easier doing your own design on the sequences you are then going to create?

KS: Yes, 100%. When you’re getting previs from a vendor, no matter how much time the client has spent on it, you still have to figure out how to undo a lot of the "mistakes" that they've made or it won’t be workable. It might be concept art of a creature that you have to pull apart and figure out where the biomechanics are wrong so that you can give something to Animation that's going to be animatable and give you the performance you want. We were lucky to already have a relationship with the director from doing Godzilla vs. Kong, so there was some trust there for sure. They were willing to let us help them out creatively and do some heavy lifting. They didn't feel that they needed to give us some super-polished previs or super-exact ideas about motion. They knew they could give us general design ideas for creatures and that we would take that and make it work.

LC: We had to modify a few designs to help the story of the shots. We just had be aware of what they were doing in terms of motion. For example, we had flying creatures that looked like whales where we had to move the fin, change the proportion of some things, so that it would work a little bit better with the scale and what they had to do in the shot.

KS: It's nice to have that relationship with a client where they can send you something and you can turn around and go, dude, this is not going to work. It’s why I love working here. You can just go to the client and say, this isn't going to work, but here's how we can change it to make it work.

DS: You mentioned that one of the biggest challenges on this show was just the enormity of the scale and the density of stuff you had to create. In thinking back over the whole production, were there any specific tasks that stand out for their complexity or for presenting unexpected problems?

KS: For me, the biggest thing was a subset of building the assets. We were in charge of building a bunch of the new characters, like Shimo and Skar King and One-Eye and Suko. And when we first got the art for Shimo, it was like, oh, that's scary, with the crystals and the breath and the iridescent skin. And you've got hard deadlines to get that work to other vendors because they need to finalize their shots. So, if you're late, you’re screwing up everybody on the whole show that also needs that character. And we kind of went around and around and it wasn't right and it wasn't right. And I was getting worried that I was going to start getting angry calls from other vendors like, where's the character? We need to do these shots.

And then it was one of those things where you just wake up one day and suddenly it's working, and it's like, oh my god. It wasn't this linear process. We basically flatlined and then suddenly one day we're at like 99%. It was the biggest relief in the whole show.

LC: I had the opposite experience because I anticipated that the biggest challenges would be in the differences in scale we had to deal with on this show – Suko is quite small compared to Kong. It's a significant difference. And when you have actors on stage who are basically the same shape and size, it quickly becomes a problem when they have to interact. It was important to keep a consistency in their eyelines, but I didn't want them looking at tennis balls or that kind of thing because it takes away from the natural aspects of the performance.

But it turned out to be less of a problem than I thought. We had to carefully think about the way we would cut that to capture things in a way that was natural for the actors, but it was much easier than I expected, which was a nice surprise.

Dan Sarto's picture

Dan Sarto is Publisher and Editor-in-Chief of Animation World Network.