'John Carter': The Original Space Adventure

Go behind the scenes of Andrew Stanton's first live-action feature and long-awaited Edgar Rice Burroughs adaptation.

New York street scene. All images: ©2011 Disney. JOHN CARTER™ ERB, Inc.

Before Star Wars and Avatar and all the rest was Edgar Rice Burroughs' influential Under the Moons of Mars, which later became the more familiar novel, A Princess of Mars, followed by 10 more in the "Barsoom" series. Pixar's Andrew Stanton had waited most of his adult life for someone to make it into a movie, and finally decided to take the plunge himself after Disney purchased the rights at his urging. The result has garnered more attention for its marketing missteps and opening weekend box office disappointment (an estimated $30.6 million) than for its thrilling moviemaking and stellar VFX.

Indeed, Stanton has delivered the mythical goods, using the familiar iconic story and visual shorthand to remind us where it all came from and why John Carter's stranger in a strange land conceit is so timeless. While many have quibbled about the narrative deficiencies (Stanton's original script with Mark Andrews and Michael Chabon was more epic in scope but scaled back for budgetary reasons), the animation by Double Negative and Barsoom environments by Cinesite are praiseworthy. Additionally, Halon's previs (under the supervision of owner Daniel Gregoire) amped up the White Ape gladiatorial fight sequence with a hand-held observational style that suited Stanton's vision.

Dneg owner and production VFX supervisor Peter Chiang says both lead VFX companies raised their game to deliver Stanton's naturalistic vision (they called it "No Country for Old Men with Martians"). MPC and Nvisible also contributed when the workload increased to more than 1,500 shots.

The Princess' wedding procession.

"We introduced Andrew to new tools and he took the stylistic renderings of Pixar to a more photoreal level," Chiang observes. "He's so used to signing off shots early in a grayscale stage or with very simplistic models and approving that animation, knowing that when he sees the renders, he's sure of what he's approved already. And so it meant that the pipeline for us could be far more streamlined. And he would accept signing off on animation before we would start to do the cloth sims and creature effects -- all those additional layers after the principal animation had been signed off. So it made it far more fluid. He dialed into the shots and what changes he wanted made and we would guide him on the reality and getting the lighting in the comps."

This represented the biggest creature work for Double Negative and so they rewrote all of their tools to handle the scope and complexity, right from rigging to muscle systems, eye renders and the complexity of eyes to cloth sims to creature effects. "We had to rebuild everything," Chiang admits. "I started in May 2009 and we did a very simple proof test that ILM's Roger Guyett supervised at a shoot in Vasquez Rocks involving two dozen shots with full creature on 35 mm by cinematographer Dan Mindel. We did lots of cheats for muscles before we developed the full muscle system. There was facial capture for the Tharks so we had to really get into the performance and render it in a very shrewd way. And from that we learned the demands of the film for Andrew. We rebuilt our creature pipeline, and did facial capture on set with two NTSC cameras and translated that into FACS shapes (using Mova capture) and then into the Tharks face. There was no translation to a real-time 3D model because the 3D model was too complicated and it always got filtered, and the throughput of that data didn't work for the Tharks. Therefore, Andrew would sign off on the capture on set by looking at the straight video feeds of the NTSC cameras. Animation software was Maya and rendered in RenderMan and composited in Nuke."

A Helium ship.

A Zodanga ship.

For the crowd system, they relied on Dneg's in-house Mob and wrote tools for muscles, eyes, etc. They also worked very hard to make the green skin subconsciously like human skin so you weren't thrown out of the film. "We went toward human characteristics so the muscle system was based on human muscle but stretched out," Chiang continues. "Legacy designed the creatures with Andrew and provided ZBrush models to us and when we looked at the eyes, Andrew wanted to retain the performance of the actors and went back to white human eyes so when they evoked emotion, it resembled what the actors did. Weight was important. He wanted a lot of human characteristics and that was the challenge. We had to get the lighting and skin renderers and subsurface detail right, including the bump maps, displacement and skin sliding."

The Thern Sanctuary.

Meanwhile, Cinesite completed 831 visual effects shots, which included creating and populating the majority of environments for the film. They also converted 87 minutes of the film into stereo 3-D. Cinesite's senior VFX supervisor Sue Rowe spent several months on set in the UK and Utah; she was assisted by four other Cinesite supervisors. The sequences included Zodanga, a mile-long rusty metal tanker that crawls like a myriapod across the surface of Barsoom (supervised by Jonathan Neill); the beautiful and elegant city of Helium, with a huge glass palace in the middle (supervised by Christian Irles); the Thern sanctuary, a huge underground cave that forms around the characters as self-illuminating blue branches as they walk through it (supervised by Simon Stanley-Clamp); and the huge aerial battle between Zodanga and Helium. Supervised by Ben Shepherd, each side's airships use solar wings to travel on light. Cinesite's team also provided explosions, fire, digital doubles, a CG Thark City environment and set extensions based on photogrammetry.

Zodanga.

For Zodanga (designed by VFX art director Ryan Church), there was a lot of attention devoted to shader resource files, per frame asset visibility and prman XML stats analysis.  One of the major challenges of Zodanga is that it's a city on legs, so the design of the legs, scale, materials and rigging had to match the time period of the story, while the surfaces and weathering had to make it look like they'd seen years of service on the Mars landscape. The textures, surfaces and edges were detailed to give a dirty, industrial feel using a combination of Photoshop, Mari and Mudbox in tandem with in-house shaders and lighting development. Since Zodanga is a very boxy, utilitarian-looking city, Cinesite needed to break up a lot of the straight edges to show wear and tear on the concrete. This was done by modeling and texturing using Mudbox as well as other techniques. Compositing used a template script in Nuke as a starting point for every shot. This was populated by around 60 layers to give compositing a very granular control to be able to tweak the lighting in Nuke.

Helium.

Helium (also designed by Church) proved very time consuming and render heavy to get full 3D renders; due to the sheer volume of assets required, Cinesite developed a proprietary hierarchical caching system, allowing for grouping and duplication of individual models within larger structures. The difficult part was accessing each different stage of this hierarchy, which was made possible by various filtering options. Each asset also had its own lighting and shading file, which was easily adjustable even from the top node of the hierarchy. Cinesite also developed level of detail files for modeling and texturing which could be manually adjusted or calculated automatically through a shot camera. The fully CG Helium environment was a huge challenge as this would be the look of the Helium city that would be reused in a number of other sequences. Both Helium Major and Helium Minor required high-res textures for parts of the city that they would push in on.

In the end, Cinesite achieved for Helium's Palace of Light what Stanton termed "the jewel of the city."

"Andrew was so ready for the CG process that he was ready to take on John Carter," Chiang concludes. "He knew the fundamentals and his expertise was invaluable."

The Palace of Light.

--

Bill Desowitz is former senior editor of AWN and editor of VFXWorld. His blog is Immersed in Movies (www.billdesowitz.com), he's a regular contributor to Thompson on Hollywood at Indiewire and he's the author of the upcoming James Bond Unmasked (Spies), which chronicles the 50-year evolution of the iconic superspy from Connery to Craig.

randomness