User:Wildroot/X-Men: The Last Stand

Production edit

VFX supervisor John Bruno (Alien vs. Predator, Avatar) started work on X-Men: The Last Stand in April 2005, at which point there was not a final director assigned or an art department set up. His most ambitious sequence was the Golden Gate Bridge scene, which was the first to be worked on and the last to be finished. The sequence alone took up on-sixth of all effects spending on the film. Initially, Bruno and Ratner considered having Magneto actually rip the bridge to pieces, then re-create it, but dropped the idea out of being too complex to approach.[1] No footage took place in San Francisco, California; the scene was shot entirely at Vancouver Film Studios in Canada. A second unit crew traveled to San Francisco to photograph elements of the bridge that were later composited into the final images.[2] To achieve the effect, a full-scale, 300-foot stretch of bridge[3] was built against a green screen, where the VFX crew filmed 30 cars sliding and moving as Magneto and the Brotherhood move across the bridge.[1] A 60-foot miniature of the bridge[3] was used for lighting references, enhanced with CGI to show snapping cables, crumbling concrete supports, and people fleeing the bridge.[1] Framestore and Weta Digital handled the majority of the digital effects on the film.[4]

Due to terrorism laws, the production could not film within 1/4 mile (or a 1000ft) of the bridge with a chopper, - it could not even take a camera car or truck on the bridge, "so we walked over it a lot" joked Bruno. What stills they could take would often have the wrong fog or lighting for the final shots. Huge models were filmed, and often in an exterior car-park so real skies could be captured. Another company who was responsible for a large number of shots was Weta. But while Weta handled a large amount of the final Alctraz, it did not start work until January. When we asked Weta's effects supervisor Eric Saindon why that was, he admitted that perhaps they could have started in December but after Kong "there was really no one at Weta in December, we were fried". Saindon was joined Weta in 1999 from Santa Barabra Studios to work on Gollium. (although his formal training is in Architecture) By comparison to Gollium's 15 mins a frame, and Kong sometimes 5 to 8 hours a frame rendertime, Wetas shots on Xmen "were closer to real time", he said. Weta work in Xmen 3 used both the procedural city texturing from King Kong, and LOTR's Massive software on X-Men. The Massive X-Men were not particularly complex - using the generic Agent that Weta already had as the agent base. Allit also wrote Weta's GRUNT renderer and thus is very comfortable with complex setups. Many of Weta's 240 shots were relatively static or locked off, "by comparisons to Peter's work" which Saindon explained are normally wildly hand held and wild. Match Moving was mainly done in 3D Equalizer. Shots were comped in Nuke and Shake. Weta had a large team on the project, given the short deadline, with "about 200 people in the final credits" VFX on Xmen 3 were also provided by CIS Hollywood , Cinesite , Framestore CFC , General Lift LLC , Kleiser-Walczak , Rhythm & Hues , Soho VFX , New Deal Studios , MPC and W.M Creations. Much of the success of planning the film was due to the films extensive previz. The film had 6 storyboard artists and 3 previz companies working on the project, which "saves money" according to Bruno.[5]

Opening

Their faces were digitally rejuvenated by a computer-effects company called Hydraulx, based in Santa Monica and owned by another firm called Lola Visual Effects. These firms, according to Bruno, do a lot of work using CG to make so-called vanity fixes. Early on in planning this scene, there was talk of getting younger actors to stand in for Stewart and McKellen. But Hydraulx got the gig instead, and the scene was shot normally with McKellen and Stewart. Then digital artists plumped McKellen's features out a bit. Stewart's mug basically just needed smoothing. Through various experimental attempts, the artists eventually went a little too far with the fountain-of-youth touches — it got creepy, so we pulled back, says Bruno.[1]

The actors had no special makeup, not even hair colouring. There were no special tracking markers, greenscreens, measurements or considerations given to the effects team. The sequence was initially scheduled to be a makeup effects, when tests were not satisfactory, alternative actors were briefly considered but thanks to a previous relationship with Visual Effects Supervisor John Bruno, Lola was given the chance to helm the sequence.[5]

Other

Three hours to apply the prosthetic makeup for Beast. The filmmakers did many tests for the character's dark shade of blue lighting.[6]

While Lola is a spin off from Hydraulx, separate from Lola - Hydraulx had a large number of their own shots, including multiple man, Some of Storm's effects sequences, and also the Convoy attack. For the famous claws, the real strap on claws could not be used during fight sequences as they are far too dangerous. For this sequence for example, Hugh Jackman would wear tracking markers - alternate colours for each 'claw' - " just so we can see what they are doing". In the end sequence when Wolverine is being "atomized', Bruno initially wanted to have Jackman act this naked, but Jackman was extremely concerned about images escaping onto the net, so a small pair of flesh coloured pants were worn. Ironically to maintain the film's rating, digital cloths had to be hand tracked and re-added back onto Jackman to cover him up. Not only was nudity an issue, but much work went into the death atomizing to make it "oatmeal" according to Bruno and not too vivid. Hydraulx used motion control footage for the multiple man sequence, with each "mutant clone" being a completely separate motion control pass. Most of the actors did their own stunts. When Angel dives out of the clinic window, it is the real actor on wires that pulled off the stunt.[5]

TM & (C) 2006 Twentieth Century Fox. X-Men Character Likenesses TM and copyright symbol 2006 Marvel Characters Inc.

With a total of 900 post production shots, Twentieth Century Fox entrusted over a quarter of them to Hydraulx. Their sister company is Lola Visual Effects. Fittingly for X-Men: The Last Stand, supervisor Greg Strause relied upon the proven prowess of his Discreet Inferno systems pipeline and GenArts Sapphire Plug-ins to handle everything from ground-breaking vanity work to photorealistic special effects. GenArts Sapphire Plug-ins were integral to the overall success of this de-aging technique, specifically a patented threesome of effects which Strause and his team have previously used on CG shots to create organic, photo-real results."For the rejuvenation sequence, we added BlurChroma to the mix, which is effective at smoothing out the skin tone of an aging actor without removing detail. Sapphire Glow also mitigates inconsistencies while maintaining the sharpness of the pores and hairs. In general, adding a Glow makes people look more youthful--sort of like a digital facial." For this shot, Strause and his team relied on the Sapphire Zap effects to create organic, credible lightning bolts.[7]

11 visual effects companies were hired. Vendors such as Soho VFX, Framestore CFC, CIS Hollywood, Rhythm & Hues, New Deal Studios Inc., Kleiser-Walczak, Hydraulx and Lola Visual Effects all contributed work to the film, with The Moving Picture Co. (MPC), Cinesite and Weta Digital contributing the largest sequences. The latter three companies will be the focus. Responsible for the opening sequence of the film, Sue Rowe, visual effects supervisor for Cinesite, details that their participation on Last Stand was a project they aggressively pursued. “We were keen on working on the show as soon as we heard it had been green lit. Cinesite had worked on both previous films, with Cinesite L.A. having done the majority of work on X-Men 2, so we had a good understanding of the show’s needs. Also, the overall vfx supervisor, John Bruno, was a good friend of Cinesite’s, as we worked on Alien vs. Predator with him a year before.” Cinesite also created a number of X-jet shots after shooting the plates in the Canadian Rockies. Nicolas Aithadi, visual effects supervisor at MPC. MPC created and destroyed the all-CG den for the violent fight between Dr. X and Jean Grey in 90 shots. The vfx house put together a visual bible of the actual set and broke down the images using proprietary software. The whole sequence was maybe 10 minutes. Another aspect of the Dark Phoenix transformation that MPC had to tackle was the physical changes Jean undergoes during her violent episodes. "The idea was that when the Dark Phoenix is taking over, Jean’s skin darkens, some veining appears on her face and her eyes go black." We had to recreate the lake in CG entirely, which was quite difficult. “We also had to make a big explosion in the lake. It starts with a whirlpool created in the lake and then water bursts up and explodes. That was quite violent to take on,” he laughs. “It was the most complex effect because it was really, really difficult to achieve. The point wasn’t to do an impressive water sim, but to get a controllable template for the whirlpool and the big explosion. We used a combination of Maya Fluid for the simulation and for the bigger waves and RenderMan for all the small detail on the water and a lot of particles and 2D elements as well.” Battle at Alcatraz: Eric Saindon, vfx supervisor at Weta Digital, “We didn’t start in reality until January (2006), which is quite a bit later than most of the vendors. We were given the end sequence and did some random shots here and there throughout the movie. But they really wanted us to concentrate on the end. Impressed with their work on King Kong for destruction of cars, buildings and people." With their chunk of the finale taking about 20 minutes of screen time, Saindon says there were plenty of other elements they had to create also. “We atomized soldiers, which is similar to the effect in War of the Worlds, when the little pods blow up the people. We also did a lot of water simulation. As Jean gets stronger and stronger, she is supposed to lift the San Francisco Bay up and pull the water up around the island and shoot it up into a dome up above — which you never actually see in the final cut. They changed things around from what we did to the final cut. We are primarily a Maya, Shake and RenderMan house. Everything we do is through that process, but the R&D for the water was difficult mainly because of the time frame we had to finish. Weta did 190 shots in total.[8]

References edit

  1. ^ a b c d Steve Daly (2007-06-07). "X Marks the Shots". Entertainment Weekly. Retrieved 2010-12-11.
  2. ^ Cite error: The named reference San Francisco Chronicle was invoked but never defined (see the help page).
  3. ^ a b Cite error: The named reference premiere.com was invoked but never defined (see the help page).
  4. ^ Annie Stordahl (2006-06-09). "VFX Facilities Create Mayhem with Massive for X-Men: The Last Stand". fxguide. Retrieved 2010-12-11.
  5. ^ a b c Cite error: The named reference Seymour was invoked but never defined (see the help page).
  6. ^ Carl Cortez (2006-05-24). "Profile: Kelsey Grammer Unleashes The Best for X-Men: The Last Stand". iF Magazine. Retrieved 2010-12-11.
  7. ^ GenArts, Inc (2006-07-06). "Hydraulx Makes Its Mark With Sapphire Plug-Ins On X-Men 3". fxguide. Retrieved 2010-12-11.
  8. ^ Tara DiLullo (2006-06-01). "X-Men: The Last VFX Stand". Animation World Network VFX World. Retrieved 2010-12-11. {{cite news}}: line feed character in |work= at position 26 (help)