The fully CG character and the transformation of a plain backlot, in Masters of the Universe. An excerpt from issue #67 of befores & afters magazine.
Beast Man (voiced by Gary Martin) in Masters of the Universe is one of Skeletor’s henchmen, and is sent to Earth to take out Adam. The character was completely computer-generated by ILM, and shared with other vendors. On Earth, Beast Man takes on Teela and Adam in and around a city roadway in Oklahoma City.
Proof previs’d the sequence. “We kept the Beast Man sequence simple in terms of previs, sticking to previs’ing just the shots with Beast Man and then using Travis’ animatics for the non Beast Man shots,” outlines Proof senior visualization supervisor Adam Coglan. “That was all keyframed. When we moved on to postvis we received a model and rig from ILM that we imported to Unreal. The animation was then a combination of our own keyframe stuff and ILM providing their animation files, to be rendered and comp’ed by us. The environment was a combination of the freeway section that they were building on a backlot, then extending the background to the left using reference of Oklahoma where they were shooting backplates and to the right using the section of Canary Wharf where Adam and Teela would escape. From a high perspective view, the Unreal scene looked quite oddly kitbashed together but once you were on the freeway, it made sense visually.”
During the shoot, stunt performer Clayton Grover wore a shoulder mount and an eyeline rig so that the eyeline would be correct for Beast Man. “About 20% of the shots in the final edit had some sort of physical placeholder stunt performer, but the rest of it was entirely digital,” breaks down Vickery.
The roadway was shot at Bovingdon in the UK on a former RAF runway. “They essentially took the piece of tarmac and marked it out and put up concrete bollards either side of the road, with a dozen vehicles queued up behind,” explains Vickery. “That meant the entire environment of Oklahoma City would be digital, but based on photography and scans of Oklahoma itself, as well as Canary Wharf where we shot a lot of reference of high-rise glass buildings that were appropriate.”
“I actually got a note during post-production asking me what we had done to this environment, but they hadn’t realized it was entirely digital,” laughs Vickery. “It was completely seamless. We did have the benefit of shooting it the right way. We shot it outdoors in natural daylight. The lighting was very natural.”
ILM got to building on its Beast Man asset after earlier concept art and a maquette of the character had been made. “On top of that,” shares ILM’s Ponti, “Travis supplied us with these beautiful storyboards, which were not just simple one-board-per-shot, they were almost like keyframe animation. Every shot had multiple key poses and key steps in it. From there, we also were told that the original cartoons and toys were a guiding principle. Tim Burke had also been very keen to approach this in a traditional manner, in the sense that we didn’t plan for motion capture or performance of any kind.”
Beast Man’s hair and fur, and his armor, were elements that ILM crafted after building up a reference guide. This came from the toys, at first. “We would ask questions like, for the fur around his neck, is that actual fur that’s overgrown, or is it like a piece of armor?” relates Ponti. “Some of the toys over the years have interpreted it one way or the other. We ended up going with some of the more recent toy lines where they made it like an additional piece of hide and a separate piece of armor. There was also his medallion, which we had to work out how it would dangle on him, and how all the straps worked.”All of that design happened in tandem with motion studies, orchestrated by ILM animation supervisor Michael Lum. Ponti mentions there were discussions about leaning towards a gorilla versus a man. “In Travis’ boards, the animation was very much simian.”
For Beast Man’s face, a LiDAR scan of the maquette— which featured the character roaring—was reverse engineered by ILM into a neutral pose. “Our model supervisor, Daniel Brkovic, started doing poses for different shapes,” says Ponti. “This included roaring and being scared. They were shot-related, actions we knew he was going to do. We also had all these character sheets that we would show Travis, firstly to see if we were getting the character he wanted, but as well for us to figure out the anatomy of Beast Man’s face. At one point we also added in the face paint. This is something our lead lookdev technical director Valentina Rosselli did. She actually painted her face and took some photos and then transferred that to Beast Man.”
Ponti adds: “It was all very exciting, for another reason. ILM, of course, is famous for all our acquisition technology like Medusa and Anyma. Well, we did none of that. It was like it was the early 2000s. Beast Man was very handmade.”
That same approach was taken in terms of animation. Beast Man was predominantly achieved via keyframe animation (although Lum and his team did utilize an ILM capture stage for capturing some motion, including capture of performer James Unsworth). “What they had done on set was do a take with Clayton to get the camera moves down,” outlines Ponti. “And then they would do a clean plate, and that’s what we worked on. The performance on set was not something Travis would go back to. You don’t get that kind of freedom that often, to really go from concept, to development, to animation, when you’re making a character.”
Beast Man’s groom required several different types of fur, which were then simulated. “We have fur on his body, on his armor, called his pauldron,” details Ponti. “The fur on his head is quite matted and grimy and greasy because he’s a filthy Beast Man, of course!”
As noted, the Bovingdon airfield location was effectively a concrete backlot in place of Oklahoma City, with a range of vehicles on the set. ILM’s city build involved gathering data from Oklahoma City and Canary Wharf. “Most of the time,” says Ponti, “we would replace all the cars and all the roads, partly for reflection needs. Travis was really happy with it. We would say, ‘You will never tell that we shot this on a grubby backlot somewhere here in the UK.’ In truth, it was quite challenging because it was a very changing season when the shoot happened. There were days when it was pouring down and days where it was really sunny. As a result, we had to do a ton of work to just make it all fit together, and to justify the sun in the background.”
New in-depth issue on Skeletor, Cringer and all the He-Man heroics in ‘Masters of the Universe’
“That was something I loved about working with David Vickery on this,” continues Ponti. “David knew we were doing a good job with the environment itself, and his main focus was about the continuity. He’d say, ‘Maybe we could put a bit of sun coming through there, or add a bit of shadow from a pretend building off-screen here.’ It really helped make it seamless.”
Having escaped Beast Man on Earth, Teela and Adam return to Eternia in Teela’s ship by ‘bending space’. They embark on a hyperspace-tunnel-like journey through a rainbow-colored rift back to their home planet. ILM created this effect, having also generated a similar look for young Adam journeying to Earth years earlier.
“Travis had a very clear idea of what this should be, although it was also a very abstract thing,” advises Ponti. “There was a desire of it being as grounded as possible, while also being very organic. As we developed the transformation and other effects in the movie that were working really well, we started integrating some of those elements back into the rift. This idea of echoes—outlines of the objects inside the rift, such as the spaceship, or Adam, they actually worked really well, so there was this 80s taste to it. And then of course, the saturation, the colorfulness, that’s one of the biggest notes that Travis would give us.”
Some of the earliest looks for the rift were based on slit- scan techniques, inspired by the Star Gate sequence in 2001: A Space Odyssey and the original Doctor Who opening. “That’s why initially we leant heavily into the environment out of the trails of the people inside the environment,” says Ponti. “In the end, we settled on something a little different because we needed the audience to understand what they were looking at. Still, we managed to input a bit more depth in there with some of these trails.”
