Two Lucasfilm tech wizards reveal how they make Hulk just the right shade of green and why directors love tidal waves with personality
The artists, engineers, and technicians at San Francisco-based Industrial Light & Magic (ILM) are a crucial drivewheel of innovation in the movie business. They number as many as 900 during peak production time in advance of the summer blockbuster season, making ILM the largest unit of Star Wars creator George Lucas' Lucasfilm.
Over its 30 years, ILM has laid claim to such innovations as the first completely computer-generated movie sequence (in Star Trek II: The Wrath of Khan, 1982) and the first fully computer-generated character (the "stained-glass man" in Young Sherlock Holmes, 1985).
WHITE HOUSE HONORS.
It's also responsible for a list of ever-increasingly realistic computer-created depictions of humans on screen (The Mummy, The Hulk, and all of the recent Star Wars releases).
For its body of work, ILM won the "nation's highest award for innovation," the National Medal of Technology, from the Commerce Dept., announced Nov. 15. The award presentation will take place at the White House on Feb. 13.
In the wake of the news, BusinessWeek Corporate Strategies Editor Brian Hindo chatted by phone with Cliff Plumer, Lucasfilm's chief technology officer, and Tim Alexander, a visual-effects supervisor at ILM, about the company's innovation process.
Plumer oversees all research and development and info-tech functions across Lucasfilm's three divisions (ILM, video-games unit LucasArts, and Lucasfilm Animation, among others). At ILM, visual-effects supervisors, along with a producer, direct film projects. Following are edited excerpts of Hindo's conversation with Plumer and Alexander:
Most of the work done at ILM is structured around movie projects, or "shows." How are the teams selected?
Alexander: A lot of negotiation [laughs]. You try to start with a core group of people you think are going to be experienced in that area. For The Perfect Storm, we picked people who had previous water-particle experience. Typically, ILM will have more than one show in-house, so it's a matter of looking at all the resources and trying to put them on the right show -- that either they'll be good at technically or artistically, or even sometimes because of their personalities.
A visual-effects supervisor and a producer usually oversee each project. How big do the teams get, and how are they organized?
Alexander: It's actually quite a range. I've been on small shows where we've had only about six people working on a show -- something like The Majestic. And then shows like Star Wars: Episode III can get up to 150 to 200 people. The average show is maybe about 50 to 60 artists. On most projects, we'll start small and then grow big. So the initial setup might only be about 10 to 15 people, but then as you get into a full-shot production, you'll get larger and get up to 50 or 60.
Sometimes the production teams will be the same across projects. But more recently we've been trying to mix that up. It used to be that one team would do three or four movies together, and that's become less prevalent recently.
Alexander: We try to do different things on every show. We're given the flexibility to try different organizational structures. And I think we've found that, with teams that stick together, if there are problems, those problems will persist. Sometimes we get into a rut on how we think about projects -- both socially and technologically. I think it's good to get different perspectives.
Plumer: Also, sometimes it's really driven by our clients and the marketplace. In the past year, we had Star Wars: Episode III, War of the Worlds, The Island, The Lion, the Witch, and the Wardrobe, Harry Potter and the Goblet of Fire -- five very large projects, all in production at the same time.
We're finding a lot more that these films' production schedules are getting much shorter. That's really driving us to look at some of these operational changes, and build some more flexibility in being able to move artists and technical people between projects a lot more fluidly. In the past [an artist or engineer] might be assigned to a project for anywhere from a few months even to a couple years.
What advantages does that flexibility give you?
Plumer: One of the positives is it enables digital artists to do more. We used to have a very specialized assembly line, and we're now enabling artists to do more than one specialized task.
Do you mean that, for example, your water specialists would focus only on water effects?
Plumer: It's more simplistic: someone who might build a model -- historically that's all they would do. You'd have a separate person who might work on the texture applied to that model. We're now trying to create an environment where one person could perhaps do both tasks.
What are ILM's hiring processes like? You must need to recruit people with both creative ability and technical ability.
Plumer: There lies the challenge. We have historically been able to find great engineers and very technical people with strong computer-science backgrounds coming out of universities or research labs. And you can find great artists coming out of more artistic schools. I'd like to think that, over the last few years, we've started to see more computer-graphics training with a balance of technical and creative training. It's probably not to the degree we'd love to see, but it's growing.
What has enabled all of the technological innovation at ILM? Is it more a matter of simply getting the best people together or of establishing the right work processes in order to enable people to innovate?
Plumer: I think it's really putting together the right team of people. You can have a lot of brilliant people, but if they can't work together, you're not going to achieve the results. Innovation can come in many different ways. It could be a digital technology we developed. It could be building a practical model and integrating that with digital effects.
And I think that's something that's still unique to ILM: We have this long history, and talent that can leverage expertise and techniques in so many different fashions to achieve a look. Every project is slightly different, but what we can pull on is 30 years of history and expertise attacking a problem.
How important, then, are the processes you have in place that allow you to execute?
Plumer: What defines ILM, at least in my opinion, is that its whole history has been based on trying to figure out a solution to achieve a vision of a creative.
In most cases, it's a director's vision. Whether that's working on a project for George Lucas, or if Steven Spielberg or any other director comes to us with an idea, we have to figure out the solution. It's not usually that we have something just sitting on the shelf, and we just pull out the dinosaur button.
Alexander: The other interesting thing is that we have to find a solution that looks good to them. We can go off and simulate water, but it may not look like the water they're thinking of. You have a little uncertainty because of the artistic vision that's involved. You have to be able to both build technology that does what you want but also solves the creative vision of the person driving it.
Plumer: Going back to Perfect Storm, we were testing the technology needed to create a 100-foot wave. We brought in experts in fluid dynamics in order to create this wave. When we showed the initial test to the director, he immediately started to direct the wave. Here we thought we visually accomplished something terrific, but we just completely lost sight of the fact that the wave needs to perform like a character in a movie. Even though things may be mathematically correct, at the end of the day we're forming an effect.
Culturally, what do ILM managers do to promote creativity?
Alexander: One of the things that they do really try to promote is that ILM is successful because of its artists -- those people creating the shots and solving these technical problems that ultimately have to have artistic results. It's artists that are the power of ILM.