Riding the mobile London city with Weta Digital.
This latest Weta Digital offering is set over 1,700 years in the future.
**SPOILERS**
Christian Rivers, Fran Walsh and Peter Jackson’s Mortal Engines, from the first of Philip Reeve’s book series, is set well after a monstrous nuclear war has wiped out almost all human life and the tendrils of civilization has rebuilt itself. Few cities remain, and those that do are set upon vast networks of huge iron wheels, chasing and devouring smaller ‘trader towns’.
The visuals are astonishing. Gigantic predator cities barging along the desolate muddy countryside, leaving deep tracks and destruction behind. Such a powerful image, and an astonishing premise for a VFX-heavy movie story. Visual effects are created by a Weta Digital team led by Ken McGaugh, Kevin Smith, Luke Millar and Animation Supervisor Dennis Yoo.
I spoke with the Weta Digital VFX Supervisor Ken McGaugh, who has a long association with the company. As far back as Lord of the Rings: Two Towers in 2002, he has been supervising the creation of CG models, landscapes and directing the effects of fantastical sequences. The work by Weta Digital for Mortal Engines is definitely something never seen on screen ever before.
ROLLING CITY
London is the biggest, lumbering tiered city on these wheels. The buildings move around on the city, in different planes. The planes move around on the base of the vehicle’s chassis. The many vast wheels that propel the city along the ground at speed, rolling the whole city through the landscape, kicking up not just dust and mud, but huge tracts of land, trees and boulders included. The city layout is always moving. The description alone hints at what kind of an animation job this was for the crew.
During the first demonstration of the mobility and prowess of the city in Great Hunting Grounds, those characters in charge of its trajectory, order the chasing down of a small trading town the spy in the middle distance. There is no hope for the smaller city and inevitably this 50-metre tall city is speared and pulled up the front ramp aboard the London ship, inhabitants included.
Weta Digital has an inhouse animation tool they call Gumby which looks after a string of complex collision and deformation tasks. This allowed the team to concentrate on primary animation jobs. “There was a lot of testing of the effects on the ground of such a heavy, lumbering vehicle travelling across it at speed,” explains McGaugh.
“At one stage we had the vehicle travel at 100 kilometres per hour (kmph), and it was too slow. We cranked it to 1,000 kmph but the physics just looked wrong. We then ratcheted it down to 300 kmph to accentuate the scale more.” The wheels contacting the ground, displacing mud and striking large gouges into the earth, all had to be simulated within Maya.
SHRIKE
One of the most compelling characters in the Mortal Engines movie is the part-human, part-automata, Shrike. Hundreds of years old, Shrike is the last known Stalker roaming the Earth, he is described (and looks like) a dead man resurrected by technology.
“They were built to kill,” Producer Peter Jackson says. “Human bodies were taken off battlefields and turned into the most ruthless, powerful soldiers possible.” His presence in the story is jolting and for good reason. Their movement is unhuman-like. There were early tests in motion capture but the result proved to be too human, so much like the actor’s human form performance that it didn’t fit into the concept of Shrike as the part-automata. So the final version was key-frame animated, yet keeping some elements of the actor’s presence within him. Shrike had to be a blend of the two performances.
AIRHAVEN
This bouyant city suspended from enormous balloons filled with a helium-type gas was one of the largest of the production sets. “Airhaven is a sort of Switzerland in the air,” Producer Phillipa Boyens says. “It’s primarily a place for the great aviators of the static settlements and the Anti-Tractionists, but it’s also visited by scavenger-traders and the occasional airship from a predator city.”
Airhaven was modelled in 3D, using HoloLens Holographic, a mixed-reality technology. “Augmented Reality is fantastic when it comes to visualizing totally impossible structures,” Concept Art Director Ra Vincent says. “The headsets gave artists an opportunity to experience their drawings and sets at full scale. Not everyone can navigate a 3D model, but everyone can wear glasses and navigate a hologram.”
The massive set was a complex maze of gangways, bridges, ropes, wires and billowy parachute-like fabric. Lanterns and lights hung throughout.
If creating the huge Airhaven was a big deal, destroying it as it collapses toward the ground in flames was an even greater challenge. Airhaven proved especially challenging for the practical special-effects department. Wind too was a big factor. There was a total of 10 wind machines on the studio floor, as well as air movers. For the later scenes, on set, they filled the studio with a haze effect, and introduced blasts of CO2 to help sell the explosions, designed with air mortars, lighting and flash bulbs. Parts of the gantry were rigged to break away as stunts took place, with technicians dropping debris and ash from the studio grid.
“We had to simulate a lot of individual sheets of burning cloth,” explains Ken McGaugh. “Basically the globes of the balloons which made up the walls and structure of this vast airborne vehicle, were on fire, falling slowly then impacting the ground. That’s an awful lot of flame simulation.”
The end third of the movie is full of immense, explosive action. The destruction of the Sheild Wall by several direct hits from the London gun is one such event which is a visual feast. The ultra-bright laser appears out of the top of StPaul’s Cathedral on London, (I’m not making this stuff up) and focuses an attack on the wall in front where the Resistance lives. Below are some detailed stages of the build.
Energy effects were the most crucial and effective way to show the power of the attack and the Weta crew had a fair idea of how to create it. “We blocked in the entire look in Maya and Houdini,” says Ken McGaugh, “and once we had something that was selected and the director liked, they legitimise the scene and put it all together. This is sometimes called the ‘Salad Bar’ approach because all the elements are created for the director to peruse. Much like a self-serve really. They come along and select the various parts of the meal, and then these are brought in quickly afterwards and put together.” Artists then assemble the whole thing, and the Effects TD then does all the qualifying for rendering of the scene to completion.
Related links:
[All images: © Universal Pictures and MRC]
[All text © Hellard Media]
Written by Paul Hellard
VFX SCIENCE 