The wonderful scene of the tank in the official trailer is quite different from the one seen in the Superbowl teaser. How many scenes underwent this renovation treatment?
The version in the Superbowl teaser was rough, temporary work that was not done by ILM, but by another facility who had not done many shots with the red Mark 3 Iron Man suit and didn't have much time to create the first versions of the shots. To make sure the shots will be as good as possible for the film, we consulted with the other facility to help them improve those shots. In the end, ILM took on most of the remaining Iron Man shots from that sequence. All of the tank shots are even better in the final film version than the official trailer and several of the other shots in the trailer (including some of the suit up shots and the shots where Tony is stuck in the suit talking to Pepper) are improved or different in the film versions. This is fairly standard for trailers of these big movies to evolve in this way.
How did the fact that you had a very precise reference in the form of the comic narrative influence your work involving the character? Did this help you or did it bound you possibilities?
Having the rich legacy of the Iron Man comics was something we never felt bound by, but it was a tremendous help. Many members of our crew are enthusiastic fans of the comics and we were given the opportunity on this film to contribute suggestions and ideas along the way. Our collective knowledge of the comics and of the character of Tony Stark was helpful in coming up with ideas to make the action more dynamic as well as in the invention of the detailed mechanics of the suit and various machines. The invincible Iron Man: Extremis, by Adi Granov and Warren Ellis was particularly useful as inspiration for the visuals and action of the film.
I find the choice of the inside-helmet visualization very effective. Can you tell us how you arrived at this solution?
The head-up display graphics for the helmets were done at a facility called The Orphanage. Because Robert Downey Jr. is such an interesting actor to watch, the idea was to let the audience ‘see inside the helmet. The inside-helmet graphics were also used to help establish the identity of Jarvis - Stark's ‘virtual butler and computerized assistant’ through the graphics. Some of the initial inspirations were real pilot heads-up displays, the virtual screens in the film Minority Report and those in the Japanese Anime' Appleseed. The final
result incorporated a few new details as well.
What scene was the most problematic for your team?
The sequence where Tony first wears the Mark 2 (brushed metal/silver) suit and tests it by flying out of his house
and over Los Angeles were probably the biggest challenge. First off, we had to create an exact match to a real brushed metal suit which had been created by Stan Winston Studios, but do it seamlessly so we could replace any amount of it - from just an arm piece to the whole thing - with computer graphics and not have the audience be able to tell. This required new materials development by our artists and technical people to get the match exactly right. We then had the challenge of lighting this reflective suit for the flying scenes above L.A., including somewhere he is high above the clouds where there is nothing to really reflect. On top of that, ice starts forming on the suit at one point, so we had to add creating ice on the suit into the mix. It was a big challenge that kept on getting harder. The other challenge was to create digital costuming for the actors, where we would provide create some or all of the suit for a shot where the actor is clearly visible and track it in to match his movements. This required us to build on our iMocap on-set performance capture tools we developed for Pirates of the Caribbean, and increase the fidelity of the software such that it can now track the actor’s movements perfectly. We needed to improve the tools because in the previous use we'd been replacing the whole actor whereas for this project we had to match him exactly, sometimes having our CG suit pieces right alongside whatever practical suit pieces he could wear comfortably for the action.
How much creative independence did they give you to complete the scenes with special effects?
Iron Man was a tremendous opportunity for us, creatively. Jon Favreau invited myself, our Animation Director, Hal Hickel and our entire crew to contribute ideas throughout the process. For example, the sequence where Tony Stark gets suited up into the Mark 3 (red and gold) suit was something that was invented after we had completed principle photography. The animatic team from PLF created moving storyboards for the sequence, which were fairly loose and Jon challenged us to come up with the machines and have input into some of their actions. Some of the footage we shot on a blue screen with Robert Downey Jr. in a wetsuit but we created the rest of the scene including backgrounds, machines and the Iron Man armor completely in visual effects. We also got to suggest details for the suits and some of the action for the fight with the bad guy at the end of the film. Jon would accept some of our suggestions and reject others but he was always very encouraging and creative about it. Because this is a Marvel studio film and they have such a rich legacy and experience with Iron Man, the studio was involved as well. It was a very creative process but as the Director, Jon Favreau had the final word and kept us all on track. Everyone on the crew appreciated the feeling of being involved creatively. The only downside with coming up with creative ideas is that then you have the challenge of getting them on to the screen as actual shots - something that gets a bit harder with those parts you're making up as you go along!
Can you talk about how you work as you go along, describing the whole process?
You could write a whole book on the subject. What we do is add the visual effects in the film - that is, the effects that can't be done on set, but that are added later through animation, computer graphics, miniatures and so on. I'll give you an example of adding a CG character, which is the bulk of our work on Iron Man. Briefly put, the process starts with taking the script (and hopefully storyboards or rough computer animations of the action we call "previs") and working with the director, art directors, cinematographer, stunt team, the team from Stan Winston studios that built the practical Iron Man suits and practical (on set) special effects teams to decide what will be done on set and what will be added later by us as visual effects. Because ILM was not the only facility working on the project, we did this in conjunction with Victoria Alonzo, the studio's effects producer, and John Nelson, the studio effects supervisor. Next we work out what models we'll need to make on our computer and as miniatures to be able to create the shots we decided are visual effects, and begin creating those and working out any new technologies we'll need. Making the computer model is just like making a regular model or miniature - you have a modeller, someone who paints the model but in this case also someone who makes all the surfaces look realistic when we start lighting it in the computer. While this is in progress, another artist will be working out how we can create Iron Man's RT effect (the trail when he is flying or shooting).
At the same time they begin shooting the movie. I'll be on set during photography together with someone to collect all the information about the cameras and measure the sets etc. In addition, we consult with the film-makers and to help explain what we'll be adding later - for example we might put a tennis ball on a stick to indicate where we'll be adding a CG character so the actor knows where to look - or describe how he'll be fighting with Iron Man to help the stunt people. Sometimes we'll shot a bunch of different takes to give us lighting references or versions with an without the actors to provide more flexibility when we start adding the visual effects. The film is then roughly edited and any backgrounds that were photographed for our shots are scanned onto the computer. We match the real camera with a camera in the computer. On Iron Man we sometimes recreated the backgrounds when Jon Favreau or the animators would come up with a new action or idea that we didn't have background photography for. We'd shot a lot of still photographs on set that helped with this process. An animator then animates Iron Man in rough form and this rough version gets translated to a fully detailed realistic version by the rigging and skinning work done by the Creature Technical Directors. The end result being a detailed model of Iron Man or any other creature moving over the background.
The asset then goes to a lighting Technical Director who lights the character to match into the original background photography and finally to a compositor who integrates the character with the background by adding shadows, dust from him walking, any other special effects elements. Throughout this process we review the work with the director and studio team two or so times a week. When everyone is happy, we “test film it out” and send the images to the editorial department for inclusion in the final film.
How much does it take to come to an end of a sequence with special effects?
We started the project in January 2007 doing the preparation and starting the models. We started a full crew later in the year – around May/June and delivered the last shot in mid-March 2008. Shots can take between a day and a few months of artist time depending on their complexity.
Can you talk about the ILM team members who worked on this movie?
ILM Visual Effects Producer, Wayne Billheimer; Animation Supervisor, Hal Hickel and I headed up a crew of over 100 people working on this project at ILM. The team included Aaron McBride our Art Director and our CG team was headed up by Doug Smythe, Philippe Rebour, Jeff Sutherland and James Tooley. Beyond that we had a terrific team of leads and artists across the board. Bruce Holcomb was our model lead and Ron Woodall our paint lead. They are two of ILM's leading hard surface model specialists and when I went to talk to them about the project early on they were tremendously excited, since they are both Iron Man fans. They both pointed out their screens which had Iron Man screen savers and that type of enthusiasm was shared by many in the crew. This helped make the show an absolute pleasure to work on.
What kind of technologies do you use to carry out audience's dreams, in this case comic lovers, in a true reality?
We used every trick in the book on this film. From traditional miniatures to the groundbreaking computer technologies for motion capture and animation rigging we'd created for ILM's recent work on Transformers
and Pirates of the Caribbean. On Ironman we used Maya for animation, Renderman for rendering and Shake and Inferno for 2D compositing. The lighting, camera and motion tracking and effects animation tools were part of a big set of tools written here at ILM based on our Zeno development platform that continue to evolve with every project. This allows us to create the highest quality photo-realistic imagery and get new ideas to the screen for each director we have the opportunity to collaborate with.
[by Redazione Comicus] [16-05-08]
