Industrial Light & Magic Artists creates the virtual reality world the OASIS for director Steven Spielberg’s mind-bending joyride.
‘Ready Player One’ © 2018 Warner Bros. Entertainment Inc., Village Roadshow Films North America Inc. and Ratpac-Dune Entertainment LLC.
Steven Spielberg’s Ready Player One sends movie audiences into a dismal future. But who really cares when it’s possible to escape a dreary existence anytime by entering a virtual playground called the OASIS? In the OASIS you can become anyone you want in a world you create. The movie takes off when OASIS creator Halliday (played by Mark Rylance) dies leaving behind a challenge: Find his Easter Egg and win control of his fortune and the OASIS. And thus, a massive treasure hunt begins. We follow young Wade Watts (Tye Sheridan), who is known as Parzival when in VR. Parzival and a group of his friends band together against an aggressive team of competitors led by Sorrento (Ben Mendelsohn), the head of an evil corporation.
Rolling Stone film critic Peter Travers calls the film, “…a mindbending joyride that jacks you into a fantasia bursting with CGI wonders, dazzling cyberscapes mixed with live action, hidden Easter eggs and infinite pop-culture shout outs to the 1980s.” Not quite three weeks into its release, the Warner Bros. film had earned more than $400 million worldwide at the box office.
“I think it’s a mistake that some motion capture projects are made by getting all the data and figuring out where the camera is later on. That’s not cinema. Cinema is when you have a point of view from where the camera is and everything else is staged to the camera.”
Janusz Kaminski was the director of photography for Ready Player One, working alongside production designer Adam Stockhausen. Artists at Industrial Light & Magic created the virtual reality world in the OASIS, while Digital Domain artists created visual effects for the real world outside the OASIS, and organized the performance capture sessions and virtual cinematography. ILM’s Alex Jaeger was the virtual production concept design supervisor, Roger Guyett served as visual effects supervisor, and David Shirk was the animation supervisor, with Grady Cofer supervising artists in ILM’s London studio. Matthew Butler supervised the visual effects artists at Digital Domain, and Gary Roberts was virtual production supervisor.
Approximately 90 minutes of the film takes place in the all-digital OASIS, a highly detailed environment filled with nods to the real world and pop culture populated with avatars based on actors who also had roles in the real world.
“Our goal wasn’t to create the real world,” Guyett says of the OASIS. “It was to create a world as complex as the real world, an alternate world that was grounded enough to feel there were real world stakes. ILM has undertaken films with a lot of digital work -- Rango, Warcraft, even Star Trek had 50 minutes of completely digital work in it. But, the amount of effort to wrangle all the characters, literally hundreds of them, and all the environments was quite an operation.”
Guyett cites the New York race as an example of the challenges the visual effects artists faced while creating an exciting, high stakes, rich, textured, virtual world. Dozens, perhaps hundreds, of cars race toward a finish line leading to one of the three keys that will unlock Halliday’s Easter Egg. The roadway takes them through a cityscape. Cars slam into the rails, fly off bridges, crash into each other. The visuals are dense and the action is nonstop. T. rex makes an appearance. So does King Kong.
“We wanted to make it feel gritty and dirty, a kind of French Connection version of New York,” Guyett says. “But things happen that are more game-based. We change the time of day. The bridge unravels into a hot wheels-like track.” The two race sequences were tackled by ILM London under the supervision of Cofer, VFX supervisor Daniele Bigi, and animation supervisor Mathieu Vig.
Because the OASIS is digital, when avatars are present in the virtual environment -- which is most of the time -- those scenes were shot on two greenscreen performance capture stages in London. Actors wore motion capture suits and facial capture rigs to provide data for the ILM animators. Witness cameras trained on the actors’ faces also provided reference for animators.
“We were on the stages for six weeks or so,” Guyett says. “It was like a normal shoot in the sense it was run by production. But, it was different in that we went out of our way to allow Steven to understand the spaces he was shooting in. He could put on goggles, look around the [virtual] set to scout the environments, and work as a filmmaker does.”
For the New York race, actors had proxy representations of the cars on the motion capture stage, but in his headset, Spielberg could see those actors as their avatars in a game-resolution virtual environment. A Unity game engine provided the real-time graphics. Spielberg also had a separate virtual camera rig that looked something like a cross between an iPad and a gaming console. This virtual camera mimicked the real world cameras so he could frame shots of the actors as they performed on stage but appeared as their avatars in the virtual world on his screen.
“He could say, ‘It would great if we could move this car slightly,’” Guyett recounts. “And, we could adjust the world for the camera just as we would for live-action -- reorganizing and art directing the space based on the camera. I think it’s a mistake that some motion capture projects are made by getting all the data and figuring out where the camera is later on. That’s not cinema. Cinema is when you have a point of view from where the camera is and everything else is staged to the camera.”
All told, the crew captured around 1,000 key character facial performances from the actors. Motion editors translated and transferred that data to the digital avatars for ILM animators to refine into final performances.
“It was orders of magnitude beyond any project of this type that we’ve ever done before,” Guyett says. “The very first motion capture delivery was twice as big as our previous largest project, which was Warcraft. The interesting thing was that people could be anything they want to be in the OASIS. They could invent characters that represented themselves or maybe didn’t. The big thing on our agenda was making sure the actors’ performances translated to their avatars.”
Artists at ILM’s London studio under the Cofer’s supervision created about half the shots in the OASIS -- the aforementioned New York race, the dance in the nightclub where Parzival falls in love with Art3mis (Olivia Cooke), and much of the third act battle outside the castle. ILM’s Singapore studio and VFX supervisor Dave Dally and animation supervisor Kim Ooi took on the second largest body of OASIS work, while San Francisco and Vancouver contributed to the rest. To help manage the complexity, ILM’s tools group developed some new systems and techniques.
“If someone gets killed or taken out of the game in OASIS, their inventory stays in the game just like in the real gaming world,” Guyett says. “And, other people can grab it. So, we had to carry around all that information, the character’s inventory. We had to ratchet up our tools and build AI into our crowd system.”
In addition to educating the Houdini-based crowd system, the software and technology group at ILM worked on new rendering and lighting tools. The goal was to render together inside Renderman everything in what is essentially a highly detailed, complex photorealistic animated feature.
“Quite often when we do visual effects shots in movies, there’s a more piecemeal approach,” Guyett notes. “We have a set extension, or we add a creature in the foreground. We can approach those shots based on layers. But with this film, everything is connected, and more often than not we needed to have the ability to render everything together. So, we solidified our approaches and rendering techniques. Barry Williams, our global lead in charge of digital environments did a great job making sure our process had integrity.”
One of those environments takes the players into a recreation of the Overlook Hotel and scenes from Stanley Kubrick’s movie The Shining. We see blood pouring out of an elevator and Aech (Lena Waithe), one of the Wade’s friends, is swept up in that river of blood. In another shot, a woman rises up from a bathtub, her skin falls off, and she attacks Aech with a hatchet.
“When Steven suggested making The Shining as one of the challenges, we were a little lost at first about what it would be,” Guyett says. “But it’s one of the best sequences in the movie. It’s such a film geek thing to do. And, I got such a kick out of doing it, creating all the detail.”
The team used elements from footage in Kubrick’s film to help recreate environments from the movie digitally. And, The Shining shots were among the few in which real humans -- the twins, the woman in the bathtub -- rather than avatars appeared in the OASIS. Another shot with humans happens near the end of the film.
“There is one moment when Parzival is kind of in a real world with Halliday,” Guyett says. “It’s ambiguous. Is he a digital character in the real world? Or, is he in the OASIS? We used lighting to give each place in the film a different quality. Steven was acutely aware of lighting.” ILM provided the render of Parzival to Digital Domain who created the volumetric projection effect.
“Steven was really interesting to work with,” Guyett adds. “He didn’t do that thing you so often find with directors -- worrying more about the details than the big picture. He’d tell me if he didn’t like something, or didn’t believe a shot, but he didn’t critique it to death. He just got on with the thing that was important -- whether or not you were emotionally attracted to the characters and whether characters were appropriate. But at the same time, he was appreciative of the gargantuan process of pulling this massively complicated show together. Of people working hard and contributing so much.”
Indeed, anyone who sees this film has to come away from it amazed at the extraordinary work by the visual effects artists who created the stunning visuals.
“There were moments,” Guyett says, “when I’d be standing there thinking, ‘Wow. This is quite an undertaking.”
