John Knoll joined ILM in 1986 as motion control camera operator to work on INNERSPACE and WILLOW. He will then work on THE ABYSS. He’s promoted as VFX Supervisor in 1990 for the film THE HUNT OF THE RED OCTOBER (with Scott Squires). He will oversee the PIRATES OF THE CARIBBEAN and STAR WARS trilogies. He will also participate in MISSION TO MARS, SPEED RACER, AVATAR or recently RANGO.
He is with his brother Thomas, one of the creators of the software Photoshop. He received nine awards including the Oscar for Best Visual Effects for PIRATES OF THE CARIBBEAN: DEAD MAN’S CHEST and more than 20 nominations. In the following interview, he talks about his work on MISSION IMPOSSIBLE: GHOST PROTOCOL that marks his return on this franchise since he had supervised the effects of the first MISSION IMPOSSIBLE.
Can you explain your background before ILM?
I was born and raised in Ann Arbor, Michigan, the son of a Nuclear Engineering Professor at the University of Michigan. As a kid I was fascinated with film and especially films that depicted something that didn’t really exist and included imagery that had to have been manufactured in some way. I earned my Bachelor’s degree at USC in the Cinema program, and while I was a student there, I started working professionally around town as a model maker and motion control assistant cameraman. I worked at mostly small non-union companies like Greg Jein Inc, M3 Effects, Celestial Mechanix, Cinema Research, Introvision and many others. About a year after graduating I saw a job posting at ILM for a motion control camera assistant and I applied.
How was the collaboration with director Brad Bird?
I really like Brad. I’m a big fan of his animated films. I think they’re just brilliant, and I leapt at the opportunity to work with him. He’s got a really good eye for composition and shot design. As you would expect, he’s very finely tuned to see subtleties in animation from his vast experience in at Pixar. Anything we did that was animated, from the spy balloon over the Kremlin, to the IMF Secretary’s SUV splashing down into the Moscow river, to missile flight paths, etc., got Brad’s signature attention to detail.
What was his approach about visual effects for his first live action movie?
Brad was very easy to work with and very collaborative. Since this was his first foray into live action, Bad Robot Productions put together a great crew of very experienced department heads to help him along in areas he was less familiar with. With Visual Effects, Brad was appropriately focused on storytelling clarity, and was happy to leave most of the « make it look real in a live action world » kinds of details to us.
How did you design and create the gadgets used by Ethan Hunt to reach the Kremlin archives?
A 3d model of the hallway was built and textured using photogrammetry techniques, and used to create the synthetic views of the hallway shown on the front and back of the screen. Shots looking through the back of the screen seeing the guard at the far end featured perspective accurate views of what the guard should see from his position in the hallway. These were done by matchmoving the guard’s head in the plate, placing a CG camera at that location, rendering the synthetic hallway from that camera, and then projection texturing that image back onto the screen.
Can you tell us more about the impressive Kremlin explosions?
The Kremlin explosion was a rigid simulation of a CG model of the clock tower and Kremlin wall. The explosions and dust were computer generated using ILM’s GPU accelerated Plume simulation and rendering engine. The cars that are thrown into the air and the rest of red square are all synthetic as well.
What was the real size of the set in which Ethan is trying to flee the explosions?
The set was a 200 by 200 foot flat platform with rubber cobblestones in front of a blue screen. The rest of the environment was synthetic, created both with CG models, matte paintings and still photography of Red Square.
How did you create the car crash seen from the inside?
The Special Effects department built a huge mechanical rig that could roll a real SUV through two full revolutions in about three seconds. Tom Cruise and the other actors were thrown about the interior of the SUV as it was spun on that mechanism in front of a blue screen. Visual effects put in the background, debris, splashes, etc.
How did you prepare the impressive Burj Tower climbing sequence?
The scene of Ethan climbing from the 118th to the 130th floor of the world’s tallest building, the Burj Kalifa hotel in Dubai, was largely filmed on the real building at the full height depicted. Tom Cruise climbed the exterior of the building, and visual effects were used to remove safety cables and their reflections in the building. As dangerous as that was, some work was considered too dangerous or impractical to shoot at full height, and were shot on a partial set down at ground level, with CG building extensions. Ethan’s glove malfunction, kicking a break in the glass of the server room, and his desperate jump for the open window are examples.
Can you tell us more about the huge sand storm that surround Dubai?
The distant and approaching sandstorm shots were created using Plume. Once engulfed in the sandstorm, production attempted to shoot as much of the chase practically as possible, and largely succeeded. visual effects were used to even out dust density, patch areas where visibility was too high, or there were undesirable gaps in the dust. The car chase in particular presented serious safety concerns because of the high probability of an accident with such low visibility. Less practical dust was used in these shots with more extensive CG dust augmentation.
This sequence ends with a impressive car crash. How did you create it?
The car crash was done practically. It was shot separately from Tom Cruise for safety reasons, and we combined the two plates in a composite.
Did you used models for this show like the submarine or the cars?
We didn’t shoot any miniatures at all on this show.
Can you tell us more about the beautiful shot showing the missile going out the submarine and then through the sky?
All of the missile launch and flight shots are computer generated, done at our Singapore studio. The last few shots as the missile approaches San Francisco are built from helicopter plates shot over San Francisco.
Can you tell us more about your work on the final fight? What is real and what is CG?
The chase and fight over a nuclear briefcase is staged in a huge robotic parking garage modeled after the Autostadt in Wolfsburg Germany. A large practical set, two floors high on one side, and five floors high on the other was built in Vancouver for the sequence. The majority of the shots in the garage contain a smaller or larger set extension, with a half dozen or so shots that are entirely computer generated.
How did you create the shot in which Hendricks is falling to his death?
He starts off as a live action element when close to camera. As he starts to fall away, we switch him to a digital double. Most of the set is our CG model in that shot.
You have supervised the VFX for the first Mission Impossible movie. What was your feeling to come back on the serie?
It was fun to revisit the series. A lot has changed since that film about how we work and what tools we use.
How was the collaboration with the VFX supervisors of the different vendors?
Lindy DeQuattro was my Associate Supervisor at ILM on MI4. She supervised the third party work for me.
Was there a shot or a sequence that prevented you from sleep?
The Kremlin explosion shot was very difficult and went through many many changes. It was the first shot we started and the last one we finished.
How long have you worked on this film?
The MI4 schedule overlapped RANGO by four months. I had another ILM VFX Supervisor, Russell Earl, supervise plate photography for the first four months until I was wrapped on RANGO. That meant I didn’t really start full time on MI4 until January. We finished the show in late November.
How many shots have you done?
We did about 700 shots on the film.
What is your next project?
I’m working on PACIFIC RIM for Guillermo Del Toro for a summer 2013 release.
What are the four movies that gave you the passion for cinema?
It’s hard to choose only four, but I’d have to say they are:
FORBIDDEN PLANET, 2001: A SPACE ODYSSEY, SILENT RUNNING and STAR WARS.
Matt Jacobs arrived at Tippett Studio in 1997. As a compositor, he has worked on films such as MISSION TO MARS, HOLLOW MAN or BLADE II. And then as Lead compositor, he takes care of the films MEN IN BLACK II, MATRIX REVOLUTIONS or CONSTANTINE. In 2007, he will oversee the effects of THE GOLDEN COMPASS then TWILIGHT SAGA HESITATION and TEMPTATION and not forgetting HARRY POTTER AND THE DEATHLY HALLOWS: PART 2.
What is your background?
I’ve been with Tippett Studio since 1997 and supervising projects here since 2006. Before that I was a compositing supervisor at the studio. I was brought to visual effects and film by my initial interest in photography. I’ve been photographing and printing pictures since I was a kid.
How did Tippett Studio got involved on this show?
IMMORTALS was brought to Tippett Studio in large part by Craig Hayes a former VFX Supervisor at the studio. He had been hired on by production to help design the God/Titan battle sequence. Subsequently he brought us on to do a test for the Titans.
How was the collaboration with director Tarsem Singh?
Fantastic. Tarsem was extremely collaborative throughout the production. He knows what he is looking for and at the same time is very open to others ideas.
What was his approach about VFX?
Not to put words in his mouth but I don’t think he cares much for visual effects as a whole. I think he would get everything in camera if possible. So we knew the Titans would be a challenge. They had to look photo real and they had to be seamless across the cut so that the effect wasn’t given away before we started dismembering the Titans.
How was the collaboration with Production VFX Supervisor Raymond Gieringer?
Working with Raymond was as well a very collaborative experience. Raymond had a great insight
into what Tarsem was looking for from the picture and helped guide us to that end.
What sequences have you made on this show?
First and foremost the Gods battling the Titans in the tomb. We also created a hyena type creature dubbed « the Mongrel » as well as working on the final shot of the picture that was referred to as the Sistine Chapel shot.
How did you create and animate the Hyena?
The Mongrel was essentially a bastardized canine we had as a pre-existing asset. We re-groomed the fur and markings on the beast as well as tweaked the ears, head and body proportions to take our canine in a more hyena-like direction.
How was filmed the Titans fight?
The Titan fight was a complicated choreography created by Jean Frenette in Montreal. He had worked out this elaborate fight where the five gods were simultaneously dispatching of the Titans. We had to work hard to break down the choreography ourselves so that once principle photography was done we could recreate the pieces we needed on the motion capture stage
Can you tell us more about the use of the motion capture for the fight?
Motion capture became the basis for our animation. We were able to use the same stunt men that played the Titans and their God counterparts for the mocap session. That made a huge difference. When we had the processed mocap peformances lined up to the production plates it was amazing how well the stunt men had done in recreating the exact performances they did in front of the camera.
Have you created digi-doubles for the Titans and the Gods?
There was basically one Titan we used throughout the sequence. We made alterations to the paint when needed to better match a particular actor when needed. For the Gods we did have scans of all of them but for those characters we were really only creating wounds.
How did you created the Titans cuts in many parts?
Basically we would split the model where we needed, cap it if say it was a leg or head being chopped off then use Z-Brush and vector displacement to add the gory details inside.
Can you tell us how you created the blood?
All of the blood was made in Houdini. We chose to go with Houdini because of it’s ability to create highly stylized fluids.
How did you manage the slow-motion aspect of the fight?
We created the slow motion effects by stretching out the animation. From the basic mocap data our animators would take over to drive and finesse the performance, ramping into the slow motion death was just a matter of taste. Sometimes we did it faster, other times slower. The blood was the more difficult thing to work out in slow motion. To get an interesting look from the dynamics the blood had to be emitted at a normal rate and the drag fields applied to get it to linger in space.
How did you created the beautiful shot of Sistine Chapel showing the Gods and Titans fighting in the sky?
Our work on the Sistine Chapel shot was to replace some of the practical Titans with our CGI ones so that the shot could have some more of the character of the rest of the film. Initially there wasn’t any blood but in keeping with the intensity of the other battle scenes we were asked to gore things up a bit. That meant killing a few Titans an eviscerating a horse.
Have you developed specific tools for this show?
We made some great improvements to our toolset on this show. Obviously the introduction of Houdini and motion capture into our pipe was a big deal. This was also the first show we used Nuke. For me that was really neat since I came from a comping background. There were things we did in Nuke that made life much easier. Our compositors added a lot of dripping blood and splatter to the shots inside Nukes 3D space.
What was the biggest challenge on this project and how did you achieve it?
There were challenges on multiple fronts. I’m not sure any one trumped the other. I guess if I have to choose one it would be slow motion blood.
What is your next project?
I’m working with Tarsem again on MIRROR, MIRROR.
What are the four movies that gave you the passion for cinema?
JAWS, LAWRENCE OF ARABIA, APOCALYPSE NOW and AIRPLANE!
A big thanks for your time.
// WANT TO KNOW MORE?
– Tippett Studio: Dedicated page about IMMORTALS on Tippett Studio website.
Vincent Poitras began in the visual effects on the movie 300 at Buzz Image Group in 2006. He then worked on films such as FLOOD or DOOMSDAY. In 2008 he joined the team of Rodeo FX. He will participate in all major projects of the studio such as MR NOBODY, REPO MEN, DEATH RACE or recently THE THREE MUSKETEERS.
What is your background?
I started doing compositing a couple of years ago. I did not really learned anything at school in that particular field. I mainly learned by my self and with the many mentors I was lucky to have. I was fortunate enough to have an internship at a visual effect company and they hired me right after starting on a Flame system. So I comped on Flame a couple of years and switch to Nuke.
How did Rodeo FX got involved on this show? Rodeo did get involve since the start doing concept to help develop the look of the movie and the environment. We send a small team to LA. It was before the show was green lit. They work closely with the director and production designer to help create the environment. We even give a little hand for some 911 emergency for the DI.
How was the collaboration with director Tarsem Singh?
It went pretty well. Watching his old flick, we knew Tarsem had a certain style and we tried our best to reflect that in what we did. Also, one thing that rarely happens here in Montreal, he did came by at our studio to design on the fly some camera move. It really helps to have a direct access to the director sometimes.
What was his approach about VFX?
He does not really have that super picky technical approach. And, for our shot, there was no crazy camera move but instead a lot of shot had no camera move at all. He mainly wanted good composition without caring to much about continuity. It was all about creating beautiful « tableau », more about being graphic than accurate. It was perfect for a matte painting/comp approach.
How was the collaboration with Production VFX Supervisor Raymond Gieringer?
It went pretty well. I think Raymond did a great job at understanding what the clients wanted. When Raymond said a shot was ok, it was almost certain the client would like it as well. I think that is in big part why internally the process went well. We did not feel like we had a multi level of decision making happening.
What have you done on this show?
We started doing concept, I think it was almost a year before they actually shoot the movie. Then when the real shot came our way, we did the dam environment with the army camp and the red shrine. We also did shots around the quarry environment, some shot of the village (where Theseus was born). The last package was a bunch of shots we had to finish from other vendor so there was a couple of different locations as well. It was mainly a matte painting and comp approach.
What was your approach to design so many different environments and aspects of the movie?
They shot the movie with a MoSys system, which allowed them to preview in realtime where that talent and set would be in CG environment. We always had those as a reference with the Lidar scan from the set. For the rare shot that had movement we started match moving the camera. We did that also on some static shot that required more complex set extension.
For the dam, we had a CG asset to have a good start for matte painters, and Lidar scan from the production that really made our job much easier, but most of the environment were matte painting asset projected that we brought to live in comp. Especially for mount Tartarus and mount Grillus, it was easier to give a sense of scale in painting. Even the army shot from above the dam was a matte painting that we animate in comp and added some particles. We also did sometime a rough passes of lighting in cg. We had a cg army for a close up of all the army when Hyperion walks in front of his soldiers. All water shots we did was practical water.
What references and indications did you received for the different environments?
Well the concept part was really important for this and since we were involve in the concept we already had a good idea where we were going. So we knew pretty much where we were heading for the the environment. As for the look there was the paint Odd Nerdrum that was given as a reference. And as I was saying earlier we had the MoSys as reference for the 3d environment and the Lidar scan for the sets.
Can you tell us more about the Dam and the huge environment around it?
The Dam protect a city in a valley surrounded by mountain. On top of that Dam is a red shrine, that lead the way into Mount Tartarus where the Titans are imprison. So the only way to get into Mont Tartarus is to open through the bug Dam. This had to be enormous in scale, band we had to feel the height of those mountains but without really have some other elements to help feel the scale as those mountains were surrounded by a desert.
Can you explain to us the creation of the beautiful shot in which Hyperion is throwing the arrow against the Dam?
Hyperion was shot live. Because of the fast pan, the army extension was a matte painting animated in comp with small warps, glint flickers. We also had some small cg flag to bring life to the army. The shot was all matte painting and projection.
How did you create the huge army of Hyperion?
For the close up, we hag a cg extension of the army and matte painting for the soldier really far. As for all shots from above the dam, it was matte painting mix with live element and some particles done in nuke. The army was animated using warps and distortion receipt.
Did you share assets with the different vendors?
Yes, with Scanline. They did the destruction of Mount Tartarus.
What was the biggest challenge on this project and how did you achieve it?
There was one shot that Tarsem did not like the camera move they shot. There was a lot of live action so in order the redo the camera, we had to re-project all the scan with different layers for the parallax. Which meant cleaning and adding all the info that was not there, painting soldier walking in front of other soldier. For that shot, all the painting and projection was done directly in nuke (except for the matte painting). Also the length of the plate was not long enough, so we had to morph soldier into 3d soldier, extending the fires. It was a complex puzzle with lots of roto and paint work. The camera revealed the army camp in the quarry at night. I think that shot was around for a long time and ended up at Rodeo. We basically restarted it form scratch and we did not had that much time to do it. Raymond told us that when Tarsem approved it, he just stood up and clap!
Was there a shot or a sequence that prevented you from sleep?
Nothing prevent me from sleep! But you can copy/paste the answer above!
What do you keep from this experience?
You can do a lot with a kick ass matte painting!
How long have you worked on this film?
Almost a year and a half. With some time of here and there jumping on another project to help for deliveries, but yeah one year and a half.
How many shots have you done?
We did 123 shots.
What was the size of your team?
Around 45 persons.
What is your next project?
Rodeo has 2 movies that we worked on coming out in January – RED TAILS and UNDERWORLD AWAKENING. We are working at the moment on MIRROR, MIRROR (the next Tarsem Singh movie), JACK THE GIANT KILLER, an undisclosed project with Sony Picture Imageworks and another undisclosed project with Industrial Light & Magic. We also just finish a print advertising for an agency in New York for the NHL.
What are the four movies that gave you the passion for cinema?
That is a hard question. When I was a kid I watched things like SPACEBALLS, THE BLUES BROTHERS, SMOKEY AND THE BANDIT. But to be honest I do not have a ultimate favorite movies. I just like when it’s stand out a little like when I first saw MEMENTO. I always loved old classic movie though. I just wanted to hear or tell a good story. I was even making small movies with friend when I was 15. But really what pushed me into VFX is when I started playing with a 3d software showed to me by my older brother in law in the time and realizing I could create anything from scratch.
Matt Aitken joined Weta Digital just after its creation in 1994. He participated in all major projects of the studio as THE FRIGHTENERS, the LORD OF THE RINGS trilogy, I ROBOT, KING KONG or AVATAR. He won two VES Awards for KING KONG and THE LORD OF THE RINGS: THE TWO TOWERS. In the following interview, he talks about his work on THE ADVENTURES OF TINTIN.
What is your background?
My background is in a combination of filmmaking and computers. I’ve loved movies from an early age and used to make short films in the Super 8 format. Then when personal computers first became available I was immediately drawn to them, I particularly loved programming computers to create images. I studied film, mathematics and computer graphics systems programming at university then joined Weta Digital in 1994 when the company was just getting started. Since then I have had the privilege to work on some fantastic projects here including LORD OF THE RINGS, KING KONG, AVATAR, DISTRICT 9 and now THE ADVENTURES OF TINTIN.
How was the collaboration with director Steven Spielberg?
Steven was fantastic to work with, always enthusiastic about the work and incredibly astute on what changes needed to be made. We reviewed the work with him via video conference calls, by the time we got up to speed with the show we were doing these calls with him daily. He might’ve just been being polite but on more than one occasion Steven said that these calls were the highlight of his day. I found it fascinating to watch him in action, within a few seconds of seeing a clip he would come back with very specific notes that were always bang on the money and ultimately about ensuring that the story he wanted to tell was clear. And if there were no notes, he would say « put it in the movie! » which we always loved to hear.
Can you explain to us the preparation and design process?
Weta Digital formed its own internal art department for the first time on THE ADVENTURES OF TINTIN. We realized that as our role on this film was wider than usual, Weta Digital was effectively the production company for this film, an art department was crucial if we were to realize our goals of making the film look as rich and beautiful as possible. Building on early concept work from the Weta Workshop design team under Chris Guise, the Weta Digital art department under art director Kim Sinclair embarked on the process of designing interior sets, buildings, vehicles, furniture, props, weapons and costumes for all the characters. There were 2 primary points of reference for the movie’s design: Hergé’s original panel art and reference of actual locations and items. In practice these two weren’t hard to bring together into a unified design aesthetic because Hergé himself drew his environments very accurately, often basing his panel art on photographs of actual locations and vehicles. Translating this reference into textured, shaded digital models we used the flexibility offered by working in an entirely cg space to build on reality. Whereas Hergé translated the scenes in his reference photos into his signature ‘ligne claire’ drawing style, distilling each panel into an image of high graphic impact, we were able to use the tools we had developed over the years to present not so much a stylized version of reality but more an idealized reality. So Tintin’s home town is a picture-postcard version of a northern-European city. The idyllic ocean dawn that breaks over the heroes in the longboat is particularly beautiful, one that a live-action shoot might have to wait days or weeks to capture. And the desert can manifest itself as an infinite stretch of pristine golden sand.
Have you created previs for the whole movie?
A « template » version of the whole film was created out of a combination of cg from the performance capture shoot and keyframed digital previs. Typically the action sequences were prevised while the more performance-oriented scenes came off the motion capture stage. Our previs artists relished the opportunity to contribute original ideas to the film. These action beats, gags and chase sequences would be presented and discussed in review sessions with the director, Steven Spielberg said that he found these sessions, which took place early in the production process, to be amongst the most creative times he had had on any project, I think he found them to be huge fun! Previs also informed the performance capture shoot, particularly in more complex sequences. Steven Spielberg edited the resulting footage into a blueprint of the whole movie and turned that over to us as individual shots, the starting point for our detailed animation, lighting and fx work.
Can you tell us the process about the look and style of the characters?
Getting the look of the characters right was one of the big challenges of this production. Right from the start we wanted to honor the design that Hergé created for his characters, that signature look. But our characters also had to believably exist in the world we were creating for them: a highly detailed world of idealized reality. We used exaggerated facial features and the proportions of the comic book characters with their oversized heads and hands to carry the look of Hergé’s characters on into the world of the film. But we made the details of the characters natural: their skin textures, the way their faces moved when they spoke and especially their eyes are much closer to real humans than their overall form.
How did you decide with the director what should be the final look?
Our early review sessions with Steven Spielberg involved a lot of discussion about the look of the characters, with us responding to requests from him and also showing him proposals for how we thought the characters should look, move and perform. Our goal throughout was to create characters that could convey the range of emotions that the director needed to tell his story, with a high degree of subtlety and nuance in their facial performance. This process overlapped with our work delivering final shots for the film, there are key close-up shots for each character that are the ultimate proving ground for all the character design decisions we had made.
What were your references for the lighting?
Steven Spielberg gave us detailed lighting notes at the start of each sequence, in fact he has a credit on the film as Lighting Consultant. He has a cinematographer’s eye for how to light a shot to emphasize the essence of the storytelling that has to happen in the shot and it was great being able to draw on that experience. As with other aspects of the visual design of the film the lighting aesthetic could be described as ‘idealized reality’. Our basic lighting setups were derived from the practical lights in the scene and resulted in a natural lighting look. But we never limited ourselves to that and if the initial result made it hard to read an action beat or a nuance of performance then we would have no compunction about bringing in an extra light off camera to clarify the point. The trick was to do this in such a way that didn’t interfere with the natural feel of the lighting we had established. We referred a lot to scenes from Steven Spielberg’s live-action films as a guide to this, MUNICH, WAR OF THE WORLDS and especially the original RAIDERS OF THE LOST ARK.
Tintin takes us to a lot of different sets and landscapes. Can you explain to us your references for them and how you create them?
One of the things I love about THE ADVENTURES OF TINTIN is the huge range of locations we go to in the film, and their different moods and lighting. From the soft light of a northern European city, to the bright colours and harsh sunlight of a north-African port town, from the ocean at dawn to a night-time docklands we are always racing from one location to the next. The design of these environments was a collaboration between previs and the art department. For action sequences such as Snowy chasing the red van and the chase through the streets of Bagghar previs determined the overall layout of the environment to support the action. For other environments like the Karaboudjan cargo ship and the docklands the art department lead on the design. The art department amassed a large library of photographic reference to ensure that architecture was period-correct and material textures were accurate to a highly detailed level. The digital build for the environments involved constructing more of the locations in geometry and less use of digital matte paintings than previous vfx projects we had worked on. This is partly because the film is presented in 3D stereo, and so more of the environment needs to be created in true 3D space. But also the film is an old-school road movie, we are constantly moving through the environments and see them from an ever-changing perspective.
The movie features a lot of reflections and refractions. Can you tell us more about their creation?
The film includes many shots where characters are seen reflected or refracted, many of these originated in work from the previs artists. Visually these shots bring elements together in a single frame without having to cut from one to the other. The first time we see the model Unicorn it is reflected in a dozen mirrors in a market stall. And our first look at Captain Haddock is as a comical distortion refracted through an empty whiskey bottle, giving us the opportunity to introduce him as a facsimile of his comic-book appearance. We decided early on that the only way to get these shots working effectively, especially in 3D-Stereo, was to render them in a physically accurate space using ray tracing. We are fortunate in that a lot of our rendering pipeline is built on ray tracing architecture so our processes lend themselves easily to working this way. As an example there is a shot towards the end of the film where the Captain holds a whiskey bottle in which we can see Sakharine reflected, holding a lighter flame up to the scrolls. A pre-rendered image of Sakharine could have been texture-mapped onto the bottle but it wouldn’t have moved across the surface of the bottle, as the Captain moves the bottle through space. By raytracing the reflection of Sakharine we get a correct spatial relationship between the bottle and Sakharine, and consistent 3D-Stereo across the shot as a bonus.
Can you tell us more about the creation of the beautiful Ocean?
The ocean is a full fluid simulation throughout the film. Building on techniques we developed for AVATAR our water simulation software has developed and can now handle much larger events. This meant that we could create calm ocean scenes like the seaplane sequences, as well as the storm-ravaged ocean of the pirate battle. Another key development was enhancements to our toolset for simulating the interaction between objects and the ocean. This includes bow waves and wakes around ships, down to the splashes when characters fall in the sea. These smaller events are in many ways the most difficult to achieve: the shot where the Captain falls off the longboat and creates a big splash is the most complex water simulation event ever calculated at Weta Digital. In conjunction with the simulation work the shaders that we use to render water have also been enhanced and now incorporate a more physically accurate light scattering model to great effect.
Can you tell us more about the impressive continuous shot at the end of the chase?
This shot lasts over 2 and a half minutes and is one of the most complex shots we have ever done at Weta Digital. Previs was the key to designing the shot. The shot was prevised over a 5 year period, starting late in 2006 and previs for the shot was eventually approved in April of this year. In several intense periods over that 5 year span our animation supervisor Jamie Beard worked with his team to shape the sequence in close collaboration with the director. Different gags were explored, the timing and camera work were refined, motion capture was used to help shape the action where appropriate. Once approved and turned over into animation and lighting a « divide and conquer » approach was taken to making working on the shot manageable. The shot was split into 17 sub-shots of a more typical shot length. Care had to be taken to ensure that the transition from one sub-shot to the next was invisible, so the transitions were placed at times where the camera was moving fast to help disguise the switch. Added complexity came from the large and up-close water simulations featured in the sequence and the need to subtly change the key light direction so that the action was well lit no matter what direction the camera happened to be pointing.
How did you manage so many elements?
Our work on AVATAR taught us that it is crucial to get the tracking of assets correct from performance capture on. So on Tintin we had asset management tools in place so that if the director called for a specific prop in the middle of a motion capture session, the use of that asset in that scene will be tracked from then on. All assets are entered into our proprietary asset management system as they are created. If we didn’t do this, the world of Tintin, where every item in his apartment, every leaf blowing around on the street and every basket of fruit in the marketplace of Bagghar is a unique digital element, it would become impossibly complex to manage.
What was the biggest challenge on this project and how did you achieve it?
Notwithstanding everything I’ve mentioned above, achieving great performances out of our digital characters is always the one single thing that we take most seriously here at Weta Digital. We knew that the success of this film hinged on the characters having great performances, anything less would detract from the emotional resonance of the film and distract the audience from the story the director wanted to tell. To achieve this, Weta Digital built on the performance capture processes it has been running since creating the character of Gollum in THE LORD OF THE RINGS, employed to great effect recently in AVATAR and RISE OF THE PLANET OF THE APES. As always our goal is to ensure the actor’s on-set performance is clearly conveyed in their character’s digital persona. Key to achieving this is a highly developed facial motion capture system. But that in itself is not enough, we have always found that the ability to enhance the performance that we get « out of the box » with keyframe animation is crucial. To this end our facial rigs have been set up to allow input from both captured motion and keyframe animation of the facial controls together. This fine layer of finessing that we are able to add on top of the captured motion can make all the difference, it brings the artist’s craft to what would otherwise be a procedural system and somewhat paradoxically enables us to create a performance that is more true to the actor’s original performance.
Was there a shot or a sequence that prevented you from sleep?
There’s always one! For me it was the climactic docklands battle. This sequence was the last one to be turned over to Weta Digital and we didn’t have much time to finish it. Work on the sequence was further complicated because there was a lot of complicated destruction work for our fx department . And a lot of the destruction was sequential: dock buildings and in particular Haddock’s crane are subjected to blows that change their appearance significantly over a series of shots, so the destruction work for those shots had to be done in sequence which slowed things down. Detailed planning got us through and we ended up delivering the sequence of 100 shots a few days earlier than scheduled.
What do you keep from this experience?
I loved working in an entirely computer generated space on this project. Our vfx work usually involves us ensuring that our cg matches the lighting in the plates we are delivered from production, and there can be an element of compromise involved there. With projects like THE ADVENTURES OF TINTIN we have complete control over all aspects of the production: the look of the characters and environments, the animation of the camera and particularly the lighting. So we could make things look as beautiful as possible. I relished that freedom.
How long have you worked on this film?
Off and on since April 2004, that was when we got involved with testing a digital Snowy for a live-action THE ADVENTURES OF TINTIN which was Steven Spielberg’s initial plan. In 2007 I spent some time in Los Angeles preparing for the performance capture. I supervised work on the show through 2009 when most of the rest of Weta Digital was busy finishing AVATAR. And I spent the last year of production working full time on the show.
How many shots have you done?
Weta Digital completed nearly 1300 shots for THE ADVENTURES OF TINTIN.
What was the size of your team?
At the peak of production crew size was over 800. In total around 1100 Weta Digital artists worked on the show throughout the production.
What is your next project?
Weta Digital is returning to Middle Earth with Peter Jackson, I think I hear Gollum calling me now!
What are the four movies that gave you the passion for cinema?
2001 A SPACE ODYSSEY holds a special place in my heart for the poetry of its visual effects, and some very early computer graphics. STAR TREK II: THE WRATH OF KAHN for the Genesis Effect sequence, there was a shot that could only be achieved with computer graphics. The light-cycle chase in the original TRON for the dynamic use of a digitally animated camera. And JURASSIC PARK for the creature work, that film changed cinema for me.
After working on the animated series POLLUX in 2007, Sirius Buisson joined the teams of Buf Compagnie where he worked on projects such as SPEED RACER, AVATAR or THOR. He was also on set VFX supervisor for ADÈLE BLANC-SEC. He has joined MPC to work on WRATH OF THE TITANS. In the following interview, he talks about his work as supervisor of the short film WELCOME TO HOXFORD.
What is your background?
I started to work on the animated series POLLUX at Action Synthèse at the end of 2007.
Then I joined Buf Compagnie where I worked at for three years and a half, and where I had the opportunity to work on different feature films as well as commercials. I worked on animated feature films such as ARTHUR & THE REVENGE OF MALTAZARD, but mostly on live films that integrate vfx like SPEED RACER, AVATAR, L’ASSAUT or even THOR.
During this time, I also had the chance to go supervise on the ADÈLE BLANC-SEC film shoot, which have been very enriching for me.
And last September, I threw myself into the project WELCOME TO HOXFORD.
How was the collaboration with director Julien Mokrani?
The collaboration with Julien was very interesting. We didn’t necessarily have the same background, yet it was very easy to get on with him and we could quickly decide together on which direction to take.
Julien is ready to listen to what we can bring to him, he always lets a bit of creativity while knowing very well what image he wants to obtain. And also, we share a common passion for comic strips, which enabled us to collaborate easily on the whole work of graphic adaptation.
How did you get involved on this project?
I got involved on this project thanks to Hélène Saint-Riquier, the vfx producer of the film, with who I had already worked on a Nina Ricci advert at Buf Compagnie. She told me about it very early, when the film was still being written. They were looking for an on set supervisor. The project, as well as the idea of working again on a film shoot, seduced me straight away.
Can you explain to us in details your role in this project?
Firstly, my role was to prepare and supervise the film shoot for the vfx part. Therefore, I worked before with the whole team to determine what would be live and what would be CG, and to anticipate the film shoot to the maximum in advance to assure a good post-production material(green screens, trackers, clean shots…).
During the film shoot, there were a lot of photo shoots in order to have as many references as possible (hdri maps, props references, positioning and measures of the different sets…), and, of course, a rigorous gathering of much information on the camera, the sets and the lighting to be able to recreate each element in 3D if necessary.
It really was a permanent collaboration with each department in order to obtain as much helpful information as possible for the post-production.
Afterwards, I took care of the supervision of the post-production. The work was essentially based on research on the effects designs and also, of course, on a team management.
It was about coordinating several CG artists teams, giving them all the necessary elements for their tasks and conveying the director’s view to them, while finding solutions to succeed in bringing all the plans out within the short given time.
It was also very important to insure the cohesion between every department and to maintain the coherence of the look throughout the whole film (the vfx were divided over 6 different studios).
As everyone came working when it was only possible, I had to take care of the continuity between each step; therefore, I worked on a bit of everything (retopo, script, animation, modelisation, lighting, compositing…).
What was one of your typical days like?
During the film shoot, you run everywhere, especially when two camera teams are shooting at the same time. You try to anticipate everything that’s happening and, mostly, you take down a maximum of notes because if you miss something, you won’t be able to do another take the next day.
Generally, I was getting there early and I was leaving last to make sure I had access to the set and to receive all the information.
In the post-production, it was a lot and a lot of working hours. I was spending my day supervising the CG artists, sending emails and giving phone calls to coordinate everyone.
Later in the evening, when it was a bit more peaceful, I could start doing some shots 🙂
The days were pretty long but very motivating because everybody was passionate, they were there because they wanted to and we all had the same motivation: do something good in a good atmosphere. They really invested themselves, it was a real pleasure to work with everyone and to overcome the difficulties together (and there were some!).
How did the film shoot go?
We shot in the Parisian suburb for 5 days.
It was amazing, we had the feeling we were on a Hollywood feature film because the human and technical means were impressive.
Everybody came to work voluntarily, driven by the same motivation: do something huge.
I had the chance to work again with Thierry Arbogast and Jean-Christophe Spadaccini who I had met on Adèle Blanc-Sec. Their help and implication on the project were amazing.
Concerning the vfx, this film shoot really was very pleasant and interesting. I was really able to do all what I needed. The whole team was attentive to the post-production’s needs.
What was the real size of the sets?
There were five different sets, three cells, the bus, the Warden office and the cellar for the scene with the mother.
Almost everything has been built in the studio, which represented an area of about 120 m2. Caroline Thibouville, the production designer, and her team’s work was extraordinary. Everything was built in record time (about 10 days) and with an impressive quality. Everything was planned for the needs of the direction and the vfx: most of the walls and ceilings were removable as well as the bars in the cells.
The sets were a true material for the post-production, we had an amazing base to work on the extensions of the sets, and very rich textures to find the prison’s look.
However, the sets in the transformation sequence, which takes place on the prison roof, were entirely made in 3D for mise en scène reasons. The Parisian studio « Circus » dealt with this part and did an amazing job and, once again, in a short given time.
Can you tell us more about Raymond’s glasses?
It’s one of the effects that we anticipated the most because it really is one of the important characteristics of the comic strip. It defined the main character. The idea was not to do something too much 2D like, while finding an aspect that’ll keep the strength of comic strip.
During the preparation, we already asked ourselves about Jason’s glasses lenses and we decided, with the director of photography, to use reflective lenses as well as reflectors of different shapes to bring back graphic sides to the reflections.
Therefore, in post-production, we could create our effect directly using the actual reflections.
It took a bit of time to find the right look; the technique we choose brings a more realistic aspect than if we had to replace everything.
Of course we had to rotoscope the glasses carefully and restore the camera team on every shot.
How did you create this huge environment outside as well as the prison?
This has been very complicated to set because, at first, we had to go shoot some real sets and then, rework the images to insert the prison and the bus in! Later on, we were forced to make everything in 3D and, inevitably, it took a long time for the set to take shape.
We worked with several speed painters to define the aspect and the volumes. Julien wanted something very precise, hard to get in the given time.
A whole team worked on this part, under the supervision of Cédric Chapeleau, who is used to the construction of large-sized sets. A group of students in third year at Méliès School helped in the setting and the construction of some elements of the sets.
We had to create a plants database and other props to bring a sufficient level of detail to the desert and to the area around the prison.
Then, we used some mapping camera techniques to be able to cover the important surface of the desert and quickly set a working basis for the lighting.
Various Sfx, dust for example, were then necessary to bring life in this full 3D atmosphere.
The compositing was finished on Flame, at Bepong’s, by the brilliant Micha Sher, in order to blend the whole thing and give some shades back. A work on the optics has also been done at this point.
For the inside of the prison, speed painters Paul Chadeisson and Olivier Le Gall were proactive. We saw a lot of very different things and we managed to determine the volumes and the lighting from the pre-production. This early work was very useful both for the film shoot and the vfx, the atmosphere was clearly given, and we could progress in the right direction immediately.
Circus, Chez Eddy, Cutback and Cube studios worked together on the creation of the inside of the prison. Each studio worked on specific axis. There were, of course, some common elements such as matte paintings and models.
It really was a common work, some shots came under the extension, some were full 3D. The challenge was to ensure continuity in every axis. In the end, the colour grading on smoke by Cyril Chanterault enabled us to polish this part.
Can you tell us in details about Raymond’s dantesque visions?
This effect isn’t much illustrated in the comic book, we had to create something credible and that can be integrated in the atmosphere of the film. We did a lot of 2D research by helping ourselves with other comic strips from the same author, but also by inspiring ourselves with other evident references such as Mignola’s drawings, Lovecraft’s monsters and other creatures from the abyss!
The design concept comes from Mathieu Vavril. He managed to find an insane look that perfectly suits Ray’s fantasy world.
How were they created and at which studio?
It’s the company Chez Eddy who gave life to the visions, under the supervision of the talented Jean-Charles Kerninon. We made many tests to be able to integrate the whole into the live images without compromising the work from the DOP. For me, the result works perfectly and we’re completely immersed in the hallucinated world of the character. Their team worked very hard to deliver the final result, an amazing job, produced in only 4 weeks !
In the same idea, the opening sequence was composed in Flame by Micha Sher from 3D elements from the shots at Chez Eddy.
Can you tell us more about the magnificent final shot?
The final shot represents the apogee from the character in his own world. We needed to finish with something big!
The shot was tracked in order to extend the camera movement and reveal the immensity of the set.
For the set extension, Pierre Fabre produced a huge matte painting (6K), which the team at Chez Eddy then remapped on the 3D model.
The other challenge was the animation of the hundreds of tentacles in that shot. Djelloul Bekri and Sebastien Corne wrote a script to generate and animate each tentacle with an ondular and random system in order to produce the swarming aspect.
Can you tell us about the monsters and their creation?
The Wolf:
At the beginning, we wanted to keep the creatures’s head as a live effect and create an extension for the rest of the body. But we very quickly realized it was easier to recreate the whole beast in 3D to obtain more conclusive results. Nevertheless, we kept the live head whenever it was possible, when the rest of the body was not in frame.
We had to reproduce CG creature identical to Jean Christophe Spadaccini’s in order for the shots to line up.
Then we had to invent the body and the process to find a design that pleased Julien was very long. He wanted a character that was fat and thin, weak but potentially powerful, slow but lurking at the same time. A real headache!
Julien really wanted to go as far as possible from the classical werewolf representation. An innovative design had to be found while staying true to Ben Templesmith’s work. Again, we made some different versions, we started from scratch a few times to get to the final result, very original, and which stays true to the creatures in the comic strip.
The animation and the setup also needed some research in order to define how the beast was going to walk with its unusual proportions.
Fabrice Lacroix, who was supervising the animation, and Simon Reynaud, who made the rig, reached a good compromise by basing the acting on slow movements punctuated by lurking actions from the wolf. We kept in mind that a creature that almost doesn’t move seems much more menacing, confident and ready to jump on its prey.
In regards to lighting, Mathieu Malard set up an IBL system (image based lighting) with an hdri map photographed during the shoot. Then he added the 3D sets and the lights of the prison that did not exist.
The last touch was produced in Flame, again by Micha Sher, who added life to the prosthesis to line up with the full 3D shots and to blend the whole work.
The Lion:
It was also a very important character, even though he appears in only two shots. The lion represented a true challenge. We were lucky to have worked with Stephane Levallois who created a great design. The stakes were high, we needed to succeed in reproducing that powerful design in 3D.
Farid Dridi from the GFactory collective managed to recreate all the intentions of the design in the modeling and sculpturing.
We then created a believable facial setup for the lips as well as an automated setup for the tentacles representing his mane. Remi Munier, the rigger, did an amazing job and the result is so successful that the animators wished for more shots. Everyone wanted the character to have more onscreen time.
The project gathers a great amount of Parisian VFX companies. How was it collaborating with them and how did you choose them?
Indeed, a lot of studios responded to our call and I would like to thank them all again for their support and their great work on the project. We worked with Cube Creative Computer Company, Chez Eddy, Circus, Bepong, One More Prod, Cutback, Nightshift, Chez Louis.
Their contribution was made in two ways: some, like Circus, Chez Eddy, Bepong and Cutback took over the shots; others like Cube, One More Prod, Nightshift and Chez Louis opened their doors and provided us with everything we needed for our teams to work.
There’s also the George Melies school that welcomed us for a month. We gave lessons to their students in the morning and in the afternoon, they worked on the film. The idea was to give them an insight into the real production working conditions.
We turned up to each studio in the same manner, with the storyboard and the designs under the arm.
We pretty much knew what to offer to each one of them..
For many, it was the desire to work on a fiction film that attracted them. And the raw images of the shoot were already very nice, a good argument.
We also found out that many companies wished for collaboration with other studios.
Can you also explain how the work was split between the different studios?
First, we split the shots by type of effect / atmosphere, as it would be for a feature film. We went to see the studios and offered them different packages of shots. Each chose what interested them, depending also on the current projects they were working on and their human capacity.
The idea was that every company who worked on the project got something out of it.
The biggest task for me was to coordinate everything. To finish on time, the work had to be split to move faster. Each and everyone agreed on the method and it’s thanks to it that we completed the project.
What pipeline did you put in place in order to keep a visual coherence?
To keep coherence throughout the whole project, I tried to maximize the research by having speedpainting / concept design done in order to lock the universes in their volumes, the lightings and the creatures as early as possible. For the full 3D shots, we even made a lightboard immediately after having created the animatics to gain research time on the 3D. It is above all a coordination job between all the departments so that everyone moves forward in the right direction and avoids setbacks.
What was the main challenge on this project and how did you deal with it?
The main challenge of the project was the amount of effects and the time that was given to produce them without any budget!
To get there, we had to prepare the shoot and the preproduction well so that we could convince people to work with us. We were very lucky to have a seducing project, out of the ordinary. It helped us gather a lot of people.
The big challenge was to keep an artistic as well as a technical coherence over the whole project.
Were there shots that kept you awake at night?
The deadlines for the film were so tight that I didn’t sleep much until the end of the production.
Each shot was a challenge but I think that the visions and the confrontation with the wolf were the most difficult ones.
The attack of the wolf was a crucial scene and we absolutely had to pay tribute to Ben Templesmith and to Jean Christophe Spadaccini’s work.
For the visions, we had to find the look that fit the idea of the director while keeping a believable ambition to deliver something on time.
How many VFX shots does the movie have?
The film counts 119 shots, 13 of them in full 3D.
How many graphic artists worked on this project?
129 graphic artists were involved in the project as regards the image postproduction!
The core was made of 6-8 people. Then everyone came depending on their availability, their speciality and on what they wanted to do on the project.
For how long did you on this project?
I worked on this film for about nine months, shoot and preproduction included.
Initially, the postproduction was supposed to last 6 months but due to various complications, the work really started in February and finished at the end of April.
What will you keep from this experience?
It was a really rewarding experience from a technical point of view. It was the opportunity to use everything I had learned over the years to use it on a project which really motivated me and for which we had to start from scratch.
Above all, from a human point of view, the project really was an adventure where everyone gave everything.
I don’t think I’ll be able to thank these people enough. They helped us and supported us to the end – a dream team!
What’s your next project?
I hope WELCOME TO HOXFORD will allow us to develop new projects with Julien. We cross our fingers.
Right now, I’m in London working for MPC on WRATH OF THE TITANS.
Which four films made you passionate about cinema?
Ermmmm…
BACK TO THE FUTURE, JURASSIC PARK, ONCE UPON A TIME IN THE WEST, E.T, SHINNING, HEAT, SEVEN, INDIANA JONES, BLADE RUNNER, GREMLINS, TERMINATOR, LEON, SNATCH, JACKIE BROWN … did you say four? (laughs)
What is your background?
Damien has been a graphic artists and supervisor for the past 10 years (SupInfocom degree), Antoine post-producer for 5 years (French and German Management degree). They both met at BUF. Christophe Beaucarne is DoP, he joined us to found L’Atelier VFX at the end of 2009.
How did Atelier VFX got involved on this project?
We met Marjane and Vincent through Christophe. They were expecting from the post-production team to be very attentive and detail-oriented. Also, they wanted not more than 1 or 2 people to talk to, as they wanted to avoid the big studio way of production. As soon as we met we got along together very well.
Can you tell us about your collaboration with directors Marjane Satrapi and Vincent Parronaud?
M&V are actually one of our best experience in terms of collaboration. They knew from the beginning what they wanted, but were also sensible to any advise that we could give to them. We think that their comic artists (cartoonist) background was very helpful because they could judge better than anyone else how good was a work in progress: they were never freaking out because a review was not presenting something done/finished, as it sometimes happen with directors watching some WIP.
What was their approach about the visual effects?
First way: for example for the set extension, they wanted something very graphic, but not photorealistic as we composited plates from the shooting with paintings done by a russian painter. The same was for the snow flake. They wanted to have something that looks like a piece of cotton, because the snow on set was very thick already. It was more charming to have this unrealistic cotton flake than to have a microscopic realistic snowflake. Otherwise, like for the smoke, the texture had to be photorealistic even if the volume was not (Parvine’s soul).
How did you create this environment and the other in this film?
Environment was made according to the 50’s film look they wanted to have. The story had to be graphic as a tale. It was not important to be realistic but more relevant to transport the audience into another world.
What were your instructions and references for the different environments?
For the night shots: THE NIGHT OF THE HUNTER.
For the extension: Hitchcock movies, for example.
Can you explain the creation of animated sequences?
There was one decorator (environment designer) and one animator on After Effect.
It was done at the end of the post-production. Marjane and Stéphane, the editor, dubbed the voice of the characters.
How were shot the shots with Jamel Debbouze?
For Houshang (Jamel) when he appears in the café behind Nasser-Ali and his brother, Jamel was shot on green screen. In his shop and as the beggar in the cemetery, he was shot in real set on the normal background.
What was the real size of the sets?
Like two huge hangars/warehouse. Very impressive. The streets were built inside, as well as the house’s garden, cemetery, etc. Each shot of the film has been shot inside, not outdoor. It was meant to be this way as we wanted a « 50’s » look.
How did you create the different scenes including the small plane or the shot in which Mathieu Amalric jumps into the void?
Actors were shot on green screen. Then, we had to adopt the look of a polaroid with short depth of field, small models and real sized elements were mix together. Mathieu jumping for suicide was actually made by a stunt, and shot on green screen as the background was a painting. Foreground elements such as the cliff were models in real set decoration, like old time optical effects.
Are you involved on the winged creature?
It was not made with fur or anything complicated but it was several painted layers composited digitally.
How did you create the different smoke and especially the one of Mathieu Amalric?
Several invisible transitions were made out of smoke plates (smoke shot on black background).
Also, when Nasser smokes in bed and is supposed to make rings out of the smoke, Mathieu did know how to do this, so it was also shot multi-passe and composited.
And the other one over the tomb?
This one is entire full CG particles. We tried several shapes (flat, round, cloudy) before picking this one.
Can you explain in detail the creation of the beautiful smoke that moves through several sets?
This smoke was also full CG (no smoke shot for real). It took more than 2 months of particles simulation to find the right shape and animation. At the beginning, we did not start this way, it was more the shape of a stream, but it did not work out and Marjane preferred to have more like a ball symbolizing the soul/spirit of Parvine. Then this soul was supposed to travel around on its own axis and fly to its grave. Simulation was very long to calculate on computers because it could not be divided into several workstations: you just have one and only one fluid container file for the whole scene. The container file was the size of the whole garden… It took ages to render/calculate between each test, some shot were longer than a 1000 frames.
The aspect was easier to find as we proposed different alternative: diffuse smoke, heavy smoke, etc. M&V chose the heavy smoke that almost looks like a mix of smoke and milk.
What was the biggest challenge on this project and how did you achieved it?
The smoke in the garden. We made it by being patient (laugh).
Was there was a shot or a sequence that has prevented you from sleeping?
We made sure that everyone got 8 hours of sleep each night (laugh).
What are your softwares at Atelier VFX?
Mayan, Nuke, Smedge to dispatch rendering. Shotgun to manage the whole bunch of shots.
What do you keep from this experience?
Very positive experience. As a VFX studio, we are often asked to create something that has already been done. Not this time. It was very particular both at an artistic level as at a human relationship level.
How long have you worked on this show?
4 months pre production.
2 months shooting.
4 months of VFX post production.
What was the size of your crew?
Up to 10 people: VFX and animation.
How many shots have you done?
Altogether were 100 VFX shots done.
What is your next project?
We are working on a commercial right now were the decoration is totally transforming from an urban environment into a nature environment.
Also, we are preparing the post-production of next Anne Fassio’s film L’AMERIQUE.
What are the four films that have given you the passion for movies?
Not 4 films to name, but more directors: Kubrick, Hitchcock, Polanski, Tarantino, Cohen’s Brothers, Woody Allen. In Visual effects category, the first digital effects on JURASSIC PARK’s T-rex was very impressive.
A big thanks for your time.
// WANT TO KNOW MORE?
– Atelier VFX: Dedicated Page about POULET AUX PRUNES on Atelier VFX website.
// CHICKEN WITH PLUMS – VFX BREAKDOWN – ATELIER VFX
What is your background?
I got into VFX because of my love for film as a kid. I used to make stop motion lego, cartoons and I would direct sci-fi with my friends in them with my family´s super8 camera. We would make these toilet roll laser guns, and I would sit and hand draw laser beams onto the print once I got it back from the lab. From there, and to an actual career in VFX, I had some detours, but eventually I started doing web, motion graphics and photo retouching.
Then I picked up an early copy of Fusion and got into compositing, went to Shake, and I moved over to Nuke in 2008. I have been working at Storm Studios since 2005.
Because the Norwegian VFX industry is still young we have had to learn as we have gone along. This has given me the opportunity to get deeply involved in exciting projects, work with a lot of talented people and learn from my experiences.
How did Storm Studios got involved on this show?
We had worked on a couple of other shows with 4 1/2, the production company, and so they approached us with an early version of the script. We then did a couple of commercials with the director Marius Holst and DP John Andreas Andersen, just to see how we would all get along.
The director had not done a lot of visual effects before, so this was also training ground for him, getting to know the possibilities and boundaries. The cooperation was successful, and so we went into preproduction of KING OF DEVIL´S ISLAND (KODI).
How was the collaboration with director Marius Holst?
The collaboration with Marius Holst was a really good one. He is a great filmmaker with an amazing ability to connect with his actors. Most of the cast were young boys with no prior acting experience, they had all been cast as authentic. Marius Holst managed to convey that authenticity onto the screen, they felt real.
What was his approach about the VFX?
He was very humble. Coming into this with the intent to learn more and to use visual effects as a storytelling tool. This was never supposed to be a big effects film, but visual effects was crucial in telling the story. The climactic point of the movie relied heavily on VFX that would blend in seamlessly in order for the audience not to be taken out of the movie. If we lost them in the last 10 minutes, all would be lost. This was the directors main concern, and the point he emphasized.
How did you create the wide shots of the island?
The wide shots of the island were just modifications of the existing plate. Basically just shaping it to resemble the actual shape of Bastøy, which is the Island the movie is based on. They shot the arrival to the island during winter, and we had to make it look more like fall. The matte-painting was all done in Nuke.
Can you tell us more about the shot in which Braathen is hit in the face?
The stunt coordinator, Kimmo Rajala came up with the idea of putting up a 2×4 in front of Braathens face, paint it blue, and remove it in post. That way the guy hitting him could put some muscle into it and make it look more authentic.
What have you done on the fire farm?
To keep costs low, and spend production money wisely, they only built 1/3 of the barn, and we would extend the rest. For the interiors we shot with flame bars, and added the burning debris in post.
We shot the elements in our parking lot during lunch. Budgets are generally too small for VFX, so we strive to come up with inexpensive ways to get what we need. You get a long way with a 5D and some black molton.
The day of the fire, SFX rigged the barn with pyro, and the whole thing burned to the ground in 4 minutes. The footage that we got out of it looked spectacular, and we built the missing pieces of the fire with a mix of shot elements and just offsetting the plate. Morten Moen did the comps on all the fire shots, both interior and exterior.
Can you tell us more about the big war ship coming out of the fog?
This movie had quite a big budget in scandinavian terms, but that didn’t rub off on the effects budget. From a war movie we had done a couple of years earlier, I learned not to overdo levels of detail unless absolutely necessary.
The brief was that the shot was going to be a large military vessel coming ghostlike and menacing out of the fog, to end the rebellion and restore order to the Island.
What the camera can´t see is money wasted, so we constructed this scene with the philosophy of not spending more money on CG than what was actually visible. So CG supervisor Henrik Karlsson, built a fairly low detail model of the ship.
We began with animation and timing, and got that approved, then started figuring out how much we would see of the actual ship, and how thick the fog would be. The idea was to model fine detail where it was needed and not spend too much time UV mapping, shading and texturing an entire ship. We rendered out one AO hi-res angle of the boat and handed it to Nikolai Lockertsen, Head of the Art department, and he would paint the rest and we would project it back on to the ship.
The plate was just the two actors shot on green-screen back in Norway, while the shots leading up to it and in the middle were shot on location in Kalvi, Estonia. Matte-painter Stig Saxegaard, made the backdrop with stills from Estonia and Norway and painted in fog and water. With this matte-paint, all we would do was utilize depth passes for the ship and start to layer fog on top and see how much detail we needed to add in the model and what could be painting. This kind of buildup made the shot possible, because we managed to keep costs low and still come up with an impressive shot. Samuel Karlsson, Lead Compositor on KODI did an amazing job of pulling all the elements together.
How did you created the digi-doubles?
The soldiers on the boat were needed to add a sense of scale and believability to the ship. We already had some footage of soldiers shot on green that was used in earlier project. We discussed it and found that we needed some more specific action from the soldiers in this scene. A colleague and I went to a nearby parking lot that had a tall shipping container where I put on a beret and climbed on top. The day was overcast and so shooting from the ground pointing upwards gave us the right angle for the ship and an easy lumakey. This gave us the elements needed in no time, and within half an hour we could feed it into the comp to see if it worked. If not, we could always just run outside and shoot some more.
How was filmed the sequence with Erling and Olav escaping on the ice?
The great escape was filmed over a long period of time due to bad weather in Estonia. The temperature dropped just when they were to shoot these scenes. The ice started melting, and they ended up with having to move parts of the shoot to Norway.
The same issues arose back in Norway, and finding solid ice was not easy. We ended up going to a water in the mountains in the middle of Norway, to have a one day shoot with a small crew. On the way up snow set in and covered the ice completely, so a team of locals spent all night plowing the ice so we could shoot the next day.
That left us with a small area of cleared ice that had wheel- and scrape-marks. For shots where the camera was low or parallel to the ground, we could use most of the footage with minor adjustments, but for crane-shots and some shots angled from above, we had to replace everything. Since the whole sequence consisted of footage from Estonia, Norway and shots on green from the studio back-lot, the biggest challenge was getting it all to blend together seamlessly. There were also a big variation in some shots that had direct sunlight while others were freezing cold. Getting skin-tones to match from shot to shot was one thing, but the ice needed extensive work in order to have continuity. The director wanted it to feel as if they were out in the middle of nowhere, that we had gone so far out that we could´t see land. This helped us in our backgrounds, but also meant that we had to haze out or remove mountains/land from a lot of shots. The number of ice-shots in need of work was 50.
How did you create the ice breaks?
At the different locations, I shoot a great number of reference pictures of ice in all shapes and sizes.
We ended up using matte-painted ice and cracks, and animated reveals of the cracks in comp.
The ice cracking animation was done by Morten Jacobsen, and he set up 3 different animations that we reused at different places. We needed some cracking ice particles for the crane shot looking down on the two boys, but we were in desperate need to come up with a way to save time and money from the budget. So I grabbed a flower pot, put some soil on a white tabletop, set the 5D at a 90 degree angle and started shooting passes of dirt that i threw into frame. We then graded it and scaled it down to fit in with the cracking ice. This was a quick and dirty trick, but it worked and saved us time and money.
Can you tell us more about the shots in which Olav is trying to rescue Erling in the water?
They were shot on the studio backlot in Oslo, in a container filled with water. SFX had dressed with paraffin wax, that would serve as an icy surface, but we ended up replacing most of it in post, because it just didn’t blend well with the rest of the ice. After shooting for a while, the wax started to break off into really small pieces, that were floating around in the water. This took a lot of time to clean up, and we were only working on it until it became less noticeable.
Have you created new tools for the ice or the snow?
Early on in production our Effects TD Magnus Petterson set up an ice cracking system in Houdini. It would fracture where we would put « pressure points » and we could define how many branches it would make, how far it would go and how deep.
The initial script had lots more cracking ice than what actually made it in the film, and so we ended up not using it because of time constraints. We chose to go with the animated Nuke projected system instead, mostly because of the instant feedback it gave us.
What was the biggest challenge on this project and how did you achieve it?
The biggest challenge was getting it all done on time, to a high level of quality and within budget. I think the key was to know your team, play to their strengths and give creative freedom.
Besides the VFX, I worked closely with sound designer Tormod Ringens on this movie. We would meet with the director in the editing suite and figure out the best way to tell the story.
Which shots needed visuals and what could be told with sound. I also worked closely with Colorist Julien Alary, because some of the shots got finalized in the last days of grading.
Was there a shot or a sequence that prevented you from sleep?
The pressure was on with the ice sequence, and so that kept me up at night, racking my brain trying to find ways to get as many shots inside the budget without compromising on quality.
What do you keep from this experience?
Working with a great team, both on location while shooting, and back at Storm studios on the shots, was a great experience. I also learned that a bit of resourcefulness and some good old camerawork can save on time and money. We don´t have to go CG all the time.
All in all the whole show was a great collaborative effort between different departments that gave it their all, because they all believed in the project and have great respect for Marius Holst.
How long have you worked on this film?
From preproduction to release took a year and a half. But I only worked full time on the project 6 months. Marius Holst had worked on this film on and off for 10 years, so my contribution is fairly short in comparison.
How many shots have you done?
Actual comp work on my part was limited, as supervision took up most of the time. But I did shot design on a lot of the shots. Setting up slap comps and post-viz that could be handed over as reference for the comp team. All in all we did around 70-80 effects shots, and the total for the movie with online tweaks and fixes ran to around 180.
Not a massive number compared to other movies, but a big challenge on a tight budget. The biggest compliment we get is when people find out there were that many shots in the movie, because they thought it was mostly done in-camera.
VFX producer Camilla Fossen and myself also got an Amanda nomination for the VFX. This is Norway´s oscar night if you will, but we lost to TROLLHUNTER, which is another great movie that we worked on, so it felt like a victory just the same.
What was the size of your team?
The team consisted of 10 people on and off throughout the course of the project.
What is your next project?
Currently we are working on KONTIKI, a movie about Thor Heyerdahl´s expedition to prove migrational theories. A great challenge because most of the movie takes place on a raft.
We are hard at work on digimattes of New York and water sims for the raft. KONTIKI is to be released the fall of 2012.
What are the four movies that gave you the passion for cinema?
RETURN OF THE JEDI, E.T., BLADE RUNNER and AMADEUS.
A big thanks for your time.
// WANT TO KNOW MORE?
– Storm Studios: Dedicated Page about KING OF DEVIL´S ISLAND on Storm Studios website.
// KING OF DEVIL´S ISLAND – VFX BREAKDOWN – STORM STUDIOS
Pour offrir les meilleures expériences, nous utilisons des technologies telles que les cookies pour stocker et/ou accéder aux informations des appareils. Le fait de consentir à ces technologies nous permettra de traiter des données telles que le comportement de navigation ou les ID uniques sur ce site. Le fait de ne pas consentir ou de retirer son consentement peut avoir un effet négatif sur certaines caractéristiques et fonctions.
Fonctionnel
Toujours activé
L’accès ou le stockage technique est strictement nécessaire dans la finalité d’intérêt légitime de permettre l’utilisation d’un service spécifique explicitement demandé par l’abonné ou l’utilisateur, ou dans le seul but d’effectuer la transmission d’une communication sur un réseau de communications électroniques.
Préférences
L’accès ou le stockage technique est nécessaire dans la finalité d’intérêt légitime de stocker des préférences qui ne sont pas demandées par l’abonné ou l’internaute.
Statistiques
Le stockage ou l’accès technique qui est utilisé exclusivement à des fins statistiques.Le stockage ou l’accès technique qui est utilisé exclusivement dans des finalités statistiques anonymes. En l’absence d’une assignation à comparaître, d’une conformité volontaire de la part de votre fournisseur d’accès à internet ou d’enregistrements supplémentaires provenant d’une tierce partie, les informations stockées ou extraites à cette seule fin ne peuvent généralement pas être utilisées pour vous identifier.
Marketing
L’accès ou le stockage technique est nécessaire pour créer des profils d’internautes afin d’envoyer des publicités, ou pour suivre l’utilisateur sur un site web ou sur plusieurs sites web ayant des finalités marketing similaires.
