State of the Art: A Cinefex 25th Anniversary Forum
Edited by Jody Duncan
Interviews by Don Shay & Joe Fordham
(they're all VFX Supervisors unless otherwise indicated)
RICK BAKER (Makeup FX Creator):
CRAIG BARRON (Casino, Titanic)
STEVE BEGG (Batman Begins)
BROOKE BRETON (VFX Producer)
ERIC BREVIG (Total Recall)
MATTHEW BUTLER (Fight Club)
STEPHANE CERETTI (Matrix Reloadedutions)
ROB COLEMAN (Animation Director: Prequels)
RANDALL WILLIAM COOK (Animation Designer/Supervisor: LOTR)
JAMIE DIXON (True Lies)
NICK DUDMAN (Creature FX Designer: Batman Begins)
SYD DUTTON (Dune)
JOHN DYKSTRA (Star Wars, Spider-man 2)
PADDY EASON (Chicken Run)
RICHARD EDLUND (Original Trilogy)
HARRISON ELLENSHAW (Tron)
VOLKER ENGEL (ID4)
ALEX FUNKE (VFX DP: King Kong)
JOHN GAETA (The Matrix Trilogy)
STEVE GAWLEY (Model Supervisor: Back to the Future Trilogy)
ALEC GILLIS (Creature FX Creator: Starship Troopers)
NED GORMAN (VFX Producer: Pearl Harbor)
KAREN GOULEKAS (The 5th Element)
MATTHEW GRATZNER (Miniature FX Supervisor: The Aviator)
COLIN GREEN (Pre-viz Supervisor: Panic Room)
KEVIN TOD HAUG (The Cell)
RICHARD HOLLANDER (X2)
IAN HUNTER (Miniature FX Supervisor: Waterworld)
JOEL HYNEK (The Matrix)
STEVE JOHNSON (Special FX Designer: Men In Black)
ANDY JONES (Animation Director: Final Flight of the Osiris)
JOHN KNOLL (Pirates of the Caribbean)
HENRY LaBOUNTA (Twister)
ROBERT LEGATO (Apollo 13)
VAN LING (T2)
KEVIN MACK (Big Fish)
GRAY MARSHALL (The Life Aquatic)
NATHAN McGUINNESS (Black Hawk Down)
TIM MCHUGH (Dune)
DENNIS MUREN (Jurassic Park)
ERIK NASH (O Brother Where Art Thou)
JEFFREY A. OKUN (The Last Samurai)
KEN RALSTON (Roger Rabbit)
ROBERT SKOTAK (Batman Returns)
BEN SNOW (Galaxy Quest)
MARK STETSON (Superman Returns)
PATRICK TATOPOULOS (Creature/Production Designer: Dark City)
BILL TAYLOR (Bruce Almighty)
RICHARD TAYLOR (Makeup/Creature Creator: Heavenly Creatures)
PHIL TIPPETT (Robocop)
JOHN VAN VLIET (Empire Strikes Back)
BRIAN VAN'T HUL (Master & Commander)
STAN WINSTON (Character Creator: Pumpkinhead)
TOM WOODRUFF JR. (Creature FX Creator: AVP)
HOYT YEATMAN (The Fly)
Reply
JOHN VAN VLIET (VFX Supervisor): I don't call it a digital revolution. I call it a digital tsunami. It came in, sucked everyone out to sea, and then the survivors washed back up.
CINEFEX: Are you surprised at how quickly that 'digital tsunami' swept through the film industry?
JOHN KNOLL: It was inevitable that it was going to happen. But how fast digital became the ONLY thing is pretty remarkable. I did one of the first digital comps for a feature film - at least here at ILM - on THE ABYSS, the shot of the door closing on the water pod. At the time, I thought; "This is the future. Five, ten years from now, we're going to be doing all this stuff digitally." And a year later, you couldn't GIVE optical compositing away. It happened overnight.
HARRISON ELLENSHAW (VFX Supervisor): Actually, I was surprised that it took as long as it did. TRON was in the early 1980s; and at that time, I said: 'Five years from now, digital will have taken over all of visual imagery effects, and even the gathering of the image itself. Film will have pretty much gone away.' I was wrong. And the reason I was wrong was that TRON was not a commercial success. It sunk - fast. So, instead of taking 5 years, it took 10 years.
KEN RALSTON (VFX... I'll just point out when they're NOT Supervisors...): In the public's mind, JURASSIC PARK is where it happened. But it really happened prior to that, at ILM, purely because George Lucas had a belief in the technology. Things happened there that I don't think people realized were huge - like the stained-glass man in YOUNG SHERLOCK HOLMES. Really, that scene is more important than JURASSIC. You look at it now, it doesn't look like much of anything. But at the time? My God, those guys beat themselves up doing that. And by the time JURASSIC finally came out, after years of CG in other movies, that was the showcase.
I remember there being quite a bit of buzz about The Last Starfighter (which I just caught on HBO yesterday, hence it springing to mind), since all of its space scenes were computer generated. Nothing close to the realism of today, but a step beyond TRON, and certainly a major move forward.
I also remember Starlog devoting a pretty lengthy article to the stained-glass knight from Young Sherlock Holmes. At the time, that was a pretty major deal to those who followed FX.
CINEFEX: Given the origins of digital imagery in movies - the fact that it was created using massive Cray supercomputers at the start, and then later on expensive, non-intuitive Silicon Graphics machines - did anyone expect that, 10 or 15 years later, people would be doing visual effects on what amount to personal computers?
DENNIS MUREN (Jurassic Park): THAT surprised me. I didn't know it was going to turn out to be something that you could basically do in your home. It isn't like the old days, when you had to get a camera, learn how to thread the camera, learn how to build a model, find the glues at the hobby shop, learn how to make a pyro thing - all yourself. In the CG world, there is all this great software easily available, not to mention hundreds of books on Maya and Photoshop and StudioMax to educate people on how to use it.
CINEFEX: What have been digital technology's most significant contributions to visual effects?
MARK STETSON (Superman Returns): The integration of VFX shots is much easier now. You don't have the limitations that you had 20 years ago trying to make shots fit within the style of a film. We've passed the place where the film stops, the VFX come up, then they go down, and the film continues. That interruptive feeling of the work isn't even an issue anymore.
ERIC BREVIG (Total Recall): I look at it as a much better tool with which to do the same things we've always done. Almost every digital tool we have has a counterpart in the analog world. Motion capture was rotoscoping, for example. Ray Harryhausen would shoot guys as skeletons fighting, and then he'd match it frame by frame - it was motion capture, without digital tools. Digital just allows us to do all of those old things at a much higher level. We can refine the craft because we can build upon our successes until the shot is done, instead of starting a whole new try at it everytime.
Reply
• Location: She works... at the library.
•
• Reputation: 1120
RICHARD EDLUND (Original Trilogy): And that means we can be much bolder. That's another one of the beauties of digital VFX compared to photochemical. I've always said thatMultiplicity would never have been made if we hadn't been in a digital world. No way would anybody have tried that.
JOHN KNOLL: Digital compositing, especially, has been a godsend. Once you fix something, you can be sure it will stay fixed - and then you can go on to the next problem. Optical compositing was always a performance. Load this element and this projector head on the printer, wind to this frame, set your position to 15,000 north and 12,000 south, set the focus ever so slightly out to soften that, put this color filtration on here, shoot with this exposure, then wind back - and you'd do all of that for dozens of passes! And invariably there would be some mistake; so you'd fix that, and then something else would go wrong. I don't miss the not-so-good-old-days of optical compositing.
BILL TAYLOR (Bruce Almighty): The promise of the digital era - and I think Dennis Muren was the first person to say this - is that if you do it right, every step is a step forward. You have the power to edit and refine, edit and refine...
ROBERT LEGATO (Apollo 13): And because digital allows you to do things over and over again - and opportunity you would never get on film - you get better at it. It's like editing on the Avid. I can make 10 mistakes and finally get it on the 11th try. If I did that on film, I'd be fired before I ever got to the 11th try! On the Avid, no one has to see my 10 mistakes - and so, all of a sudden, I'm a pretty decent editor. That's what digital has given us.
Reply
TIM MCHUGH (Dune): The speed at which we can work is truly staggering, as well. When we did Blade Runner, we'd shoot the spinner flying, shoot the background on another stage, take those elements to the optical dept. for hand-roto, get the mattes pulled, get the compositing done.
It would be at least 2 weeks before you saw that shot as Take 1. Now, you can do all of that in hours. Another thing - back when we did Battlestar Galactica, there was oneGalactica spaceship, sitting on one bluescreen stage, with one crew. So there was a finite number of shots you could get. Today, if you built the Galactica as a digital model, that ship could show up on an infinite number of workstations, simultaneously. How much work we can get done in a given length of time represents a huge revolution.
VAN LING (T2): I would say that the pedestrian, unheralded things - like wire removal - have really been the most beneficial. Those kinds of utility shots, as opposed to the hero stuff, are probably a good 80 to 90 percent of the business.
CINEFEX: We have all this amazing technology now, and it's available to virtually everyone doing this work; and yet, there's still a significant disparity in the quality of digital FX. How do you account for that?
KEVIN MACK (Big Fish): Like any other field, there's a population of people who do this work. And within that population there are people who do it extremely well, people who do it adequately, and people who aren't very good at it at all. So there is an inconsistency to the work, depending on who does it. It is still the rare, freakish talent who can really do it well. You cannot just hire anybody and train them to do this. You're looking for rock stars.
RANDALL WILLIAM COOK (Animation Designer/Supervisor: LOTR): The technology actually allows mediocre artists to produce passable work. It's just like an only competent musician having the ability to orchestrate without having to study orchestration, due to the instruments that are sampled for use in the computer now.
KEVIN TOD HAUG (The Cell): It's all about the person who's pushing the pen - it's not about the pen itself. There was a period of time when the tool was more important than the artist - which was a sad state of affairs. I can remember when I worked at The Post Group, they sold Harry time. They didn't even tell the client who was on box. The client just bought X amount of Harry time, and they assumed that whoever was running it was as good as anybody else. That thinking still hangs on a bit; but, for the most part, the technology is not what is driving things anymore.
DENNIS MUREN: I don't know... I've always thought that one of the reasons people came to ILM was because we were seen as leading technical innovation - which we were, but only by default. We weren't innovating just to be innovating. Back in the Star Wars days, it was the quest for imagery that drove the innovation, not the other way around. When John Dykstra looked at the animatics George had done using WWII dogfight footage, he innovated motion control programming because he saw that it was needed to get those kinds of images on the screen. Now there seems to be a push for innovation that isnt' at all connected to a particular project.
CINEFEX: An example - ?
DENNIS MUREN: Hair and fur. This industry is totally hung up on hair and fur, totally hung up on fire. The audience doesn't give a hoot, but that's what is going on all over the industry. People are thinking that the innovation comes first; but I thnk that if you want to do that, you should be at a university, doing research. If you're going to work in visual fx, you have to assemble the tools that are needed for the project, and only write new ones when you need to - as opposed to trying to fit a really cool toolset into a project. Star Wars wasn't done that way; neither were T2 or Jurassic Park. A lot of people don't know that. They think we're innovating, and that those projects grew out of that innovation. We were innovating, but it was all directed toward a specific application.
Reply
CINEFEX: Digital technology has been a liberating force, in that it allows you to do just about anything. But, creatively, is there a downside to the limitless possibilities it affords?
JOHN VAN VLIET (Empire Strikes Back): Sometimes we're asked to do some really stupid things, just because we can. It's like watching a monkey with a machine gun. A lot of the forces that use it just have no idea when and where to use it. So it's a rather undisciplined application at this point.
KEN RALSTON (Roger Rabbit): The lack of limitations is the worst enemy of an artist. Because when you get to the point where anything can be done, it also means any kind ofcrap can be done. Any filmmaker with no talent can put all kinds of shit in a movie that you wouldn't watch with a gun to your head. On the other hand, in the right hands, that technology can be used for all the right reasons, to make films you can be excited about.
ROBERT LEGATO (Apollo 13): Any creative endeavor needs limitation as a point of departure. The limitation of a budget or something else. Jaws is the best example. If they could have done a CG shark, Jaws would not have been a classic - because they would've relied on the visual of the shark, and not on the very terrifying thing of what you imagined the shark looked like lurking under that water. Those yellow barrels went plop, plop, plop on the water, and you were scared shitless. But if there'd been a CG shark, that shot wouldn't have been necessary. You'd show the shark - "Wow, isn't that scary?" And you know, it's not. Seeing the monster true-to-life is not scary. When you see monsters too much, they don't work. Limitations forced Spielberg to create something that was truly terrifying.
PHIL TIPPETT (Robocop): It's a matter of restraint. How many shots of that shark did he really need to tell that story? And how much more satisfying is it, really, when you're left wanting to see more, rather than dotting all the i's and crossing all the t's?
NED GORMAN (VFX Producer: Pearl Harbor): When we were prepping The Lost World, Steven said, "Ok, I had 60 3D shots of dinosaurs in Jurassic Park; I'm giving myself 75 this time." We wound up with 78, because we decided to go crazy and add 3 shots. But that is the kind of artful restraint shown by Spielberg. He can look at a movie and know - it's one of his brilliances - 'I can tell my story and get a greater impact with less.' It's not even about the budget. He knows that visually, story-wise, it's better if he holds himself to fewer shots. That results in great stuff.
CINEFEX: The contrast is startling. A big-budget dinosaur movie made not that many years ago, and it had 78 digital fx. Now, it's not uncommon for a vfx movie to have in excess of 1000 shots.
NED GORMAN: And if you're going to do a movie with 1000 vfx shots, there'd better be a really good reason for all of them...
PATRICK TATOPOULOS (Creature/Production Designer: Dark City): The limitation of smaller budgets, too - maybe they make us more clever? Think about it. You have a great young director who does a fabulous low-budget movie. Then they give him 3 times that amount of money, and somehow he gets lost, he loses the heart of things. Sometimes I think we are going down that same path in special fx. When we have to work with little money, we really have to brainstorm and be creative. It can be frustrating, because you want to take things to the next level - but it makes you fresh again.
BRIAN VAN'T HUL (Master & Commander): A lack of money does lead to ingenuity. The older small fx shops, just to survive, couldn't always throw money at something to solve a problem. Now, it seems people will spend money to find a quick solution, rather than try to think of a new, innovative way to get around the problem. They just say, "Let's buy that thing off the shelf; let's hire that person." Rather than, "OK, with the people and tools we have, how do we get this done?"
MATTHEW BUTLER (Fight Club): Look at the phenomenal fx in Blade Runner - none of them CG, all of them very clever. I look at that scene in Brazil where Robert De Niro is literally swallowed up in paperwork, and my jaw drops. What do we think we're doing? Who do we think we are? It was all done before - and done brilliantly. Let's try to do it as good as that, and concentrate on making a good movie!
Reply
VAN LING: Before the computer handed them a solution to everything, visual fx artists really had to use their imaginations. They were always having to find new techniques to meet the director's vision. It wasn't - "Well, we have this technique that allows us to do anything, so let's just come up with whatever we want." In the old days, they had to figure out very specific ways of delivering shots. It was customized, hand-crafted fx. It was inherently creative. And that's very different from the assembly-line fx you get with the computer.
JOHN VAN VLIET: It's basically a digital factory. Everything is menu-driven, a multiple choice thing. I'm not discounting the talent of the people doing it; but it's not like the giants we had before. Before digital took over, we had some incredibly talented people who could basically carve out a movie by hand. Some of it was crude; but it was remarkable what they could do. People would draw and animate smoke, electricity, fire - and you could count on 2 hands the people who could do it well. And not a single one of us is doing it today, because now it is all plug-ins on software programs, or digitally capturing something and manipulating it. It's a lost art - the end of the age of wizards. What was the effort of a few people is now done by an army of thousands.
Are you transcribing it by hand? Wouldn't scanning it be easier or is there some copyright legalities trying to be avoided?
Reply
I have no idea of the copyright stuff in this case. If I'm told to stop, I will. Mostly... it's just that I don't have a scanner. Hahah. It would actually be a lot of scanning too.
Reply
CINEFEX: You're saying that, due to digital technology, VFX practitioners don't have to be as ingenious as they once were?
VAN LING: And not just the VFX practitioners - the filmmakers, as well. On the one hand, filmmakers are now free to do anything, because there are no limits within the digital realm. But what I've found is that this completely open capability has left some filmmakers - the ones who aren't visionaries - rudderless to some degree. You go to them as an FX person, and say, "What do you want?" And they'll say: "Well... what have you got? You show me your cool techniques, and I'll come up with something." Many directors, without this technology, were forced to be creative. They had to start with the story first, and then figure out the specific way they were going to get those shots into that story.
RICHARD EDLUND: You had to be inventive before the fact, in the old days. You'd get all your material ready, go on the stage, set up your machine gun nest because you knew the enemy was coming. It was like fighting a battle. I'm not one of those who's singing the blues for the passing of that. I spent many sleepless nights in a wrestling match with the photochemical process. I developed a lot of ways to master it, and I was pretty successful; but it was always a real tough grind. The digital world is a lot more fun. In the digital world, you can show up on the stage like a gunslinger. You're more free to kick ideas around with the director, and just figure out on the spot how you're going to shoot what you need to make it work. It's much freer.
DENNIS MUREN: It still requires ingenuity. Ingenuity is going on all the time. You think about the ingenuity of the old days, and it's Richard Edlund blowing apart the house inPoltergeist with a shotgun. But we didnt' invent that shotgun to do that - we had it. It was just putting pieces together in an interesting way to do something new; and that still goes on.
Reply
CINEFEX: Not that long ago, most filmmakers were relatively ignorant about digital technology. Has that changed? Do directors now have a pretty good grasp of the capabilities of CG?
NED GORMAN: Oh, yeah. When I came to ILM in 1982 for Return of the Jedi, we were still sort of the 'mysterious wizards to the north.' There was very little coverage of VFX at the time, and so studio people, directors and producers were still in the dark. But the directors coming up now, most of whom came of age in that period, are very savvy.
ERIC BREVIG: I think directors are definitely more FX aware, but that doesn't mean that they really understand what's involved. As an industry, we may have hurt ourselves by making things look too easy - bragging about how nowadays you don't have to shoot or light things properly because we can fix it with our digital tools. But, that is never really true. It's only true if mediocre work is OK. If you want it to be great, you still have to work really hard to shoot and light things right.
KEN RALSTON: Some directors, if they're not intimately involved with how the work is being produced, think that anything they ask for can be done, just like that. In their minds, the world of digital is easy. You just press a button, and there's your ten shots. They don't realize the difficulty involved. And, to some extent, that has changed the attitude of directors toward VFX. In the photochemical days, FX work held a kind of mystery to it; and so, the directors were a lot more embracing of your ideas.
TIM McHUGH: It's true. When I was 20 years old and really didn't know what I was doing, I'd walk out on a bluescreen stage and the crew was always very deferential to me. The DP would ask, "Is this the way you want it?" The director would walk me through the staging. There was a sacredness to FX, because it was only understood by a few people. Now that I'm older and actually know what I'm doing, I feel as if we get almost no respect. People couldn't care less about what we do. We're just sort of in the way. And they all knowit can be fixed in post anyway, with a button push. There's the 'dinosaur' button and the 'tornado' button, and the 'army' button...
JEFFREY A. OKUN (The Last Samurai): I'm a huge fan of George Lucas; but in a sense, his publicizing of the magic in the first Star Wars movie started a mis-education project - and now, everybody knows how to do FX! The mis-educated think that there's a button on the computer that says 'fix it,' and you just hit it, and we all go out for drinks.
CINEFEX: You'd think FX people would have more clout than ever. Movies are marketed as FX extravaganzas, and most of the 20 or 30 highest-grossing films of all time are FX films. It seems VFX would get more consideration now.
JEFF OKUN: Not at all. My experience is that we're considered a tremendous nuisance all the way through. You can be the 2nd or 3rd person hired on a film, and involved in an inordinate amount of preproduction time, but nobody listens to what you have to say. It's like they're doing 'due diligence' by having you there - but, really, they just wish you'd go away.
CINEFEX: VFX - the Ugly Stepchild of production. Are there any signs of that actually improving?
KEVIN MACK: I think so. I think VFX are finally being recognized by the studios and by producers as being a significant creative aspect of the production. I've been fortunate in that most of the productions I've been involved with, I've been embraced as a key creative collaborator.
MATTHEW BUTLER: My experience with Rob Cohen on XXX was great. It was the first time I saw VFX working with production at that level. So, yes, it is better - but it needs to get better still. CG is still somewhat new in the filmmaking industry, and VFX are still not fully recognized as part of the team. When it happens, like on XXX, it's a dream. But it's more often like pulling teeth.
Reply
CF: This greater awareness of digital technology, this 'mis-educated' perception that it affords a quick and easy solution - does that encourage a more lackadaisical attitude on set? Is the 'fix it in post' mentality increasing?
STEVE BEGG (Batman Begins): The tendency to ignore problems on set and worry about them later - in other words, to make them someone else's problem - is increasing. The post crews are having to fix more stuff than they had to in the past, because the ability to manipulate every pixel in al film is promoting sloppy filmmaking up front. Bad cinematography has to be resurrected, a bad stunt has to be retimed, bad takes have to be rescued by split-screening two characters into a single take. That's just cheating it too far, for my liking.
ROB LEGATO: By relying on the computer, you shortcut what you would have shot on the set, and you eliminate the discipline it takes to decide if this angle is better than this angle. "I'll just grab a plate anywhere and fill it in later" is a pretty cavalier, generic attitude - and it's a recipe for a generic shot. When I shoot something, I am very rigid about it being just so. I don't trust in my ability to manipulate it later. I act as if I've got one crack at it. When I do that, the shot turns out well. When I don't do that, when I become lazy, and say, "It's going to take another hour to set this up right - I'll play with it later," it always looks processed. It's like I've lost the soul of the shot.
TIM McHUGH: There is this reliance on digital as a safety net. "Whatever it takes to get it off the stage, we don't care. We're not shooting a 2nd pass. We're not going to measure anything. You go figure it out later." It's frustrating.
MATT BUTLER: I've been on shows where the special FX people told me, "Normally, we would have painted those wires black to make them disappear in the shot; but we didn't, because you guys are going to fix it anyway." That drives me nuts, because it's so wasteful. No! Put your energy into making that special FX shot as good as it can be, and spend the CG money where you really need to.
ALEX FUNKE (VFX DP: King Kong): Here's the problem with fixing it in post - if you spend all your time fixing a shot, you don't have time to make it look cool!
JOHN V. VLIET: Right - the best it's ever going to look is 'ok.' But it's very seductive to avoid problems on the set. Everyone's tired, and they just want to make their pages for the day. So they ask, "Can you fix that?" You say, "Sure." - and what would have taken 3 minutes to fix on set windes up costing $10,000 and God knows how much effort in post. It's used badly to cover poor planning and mistakes. They'll squander all of their resources fixing stupid things they they could have taken care of while they were there.
JEFF OKUN: I got to a point with one director where, every time he said, "Fix this in post," I'd say, "OK, but you'll have to sign this piece of paper - it's an estimate of what it's going to cost to do that." On one show, I ran up a tally of $8 million to do all these fixes! And the director said, "Where are we getting that $8 million?" I said: "I don't know. You're the one who said 'you fix it' on all these shots - you figure out where the $8 million is coming from!"
Reply
CF: With the availability of previz, the ability to plan everything out before the shoot, you' think that there would be less need to fix stuff in post.
BROOKE BRETON (VFX Producer): The more you go in prepared, the more work you do up front, the better. But studios try to save time and money in preproduction, and then everything starts to skyrocket at the end - and often, the VFX team becomes the fall guy. When, really, it was just that they needed that additional four weeks or so of getting their act together in the beginning.
JOHN VLIET: The joke is, 'Prep was on a Tuesday - after lunch.' We often jump into shows with no clear idea of what's going on. A show that will remain nameless asked if I was interested in supervising, just as they were starting to film - and they had over 1000 shots! It was like, "What, are you crazy?"
PADDY EASON (Chicken Run): I've always throught it would be fun to get some T-shirts made up for the VFX team that said 'Fix it in Pre' - because, really, the way to do it right is to figure it all out in preproduction.
VOLKER ENGEL (ID4): If you don't want to fix it in post, you have to get with people as early as possible and plan everything out. On Coronado, we went to Hunter-Gratzner 4 months before they even had to start building anything; and they said to us: "This never happened to us before in our entire career. Usually, we're brought in 2 weeks before it has to be done!" I had the feeling that everybody there was so much happier building that stuff, because they had time to deliver quality work.
IAN HUNTER (Miniature FX Supervisor: Waterworld): What usually happens is, preproduction times are cut, and so the director, producer, and cinematographer don't have time to adequately plan whta they are going to do. So, during production, they'll go out there in desperation, throw up a greenscreen, and say, "Oh, in post we'll add something there." Sometimes they don't even have time to throw up the greenscreen! They just have to go with what they've got, knowing that some poor Cal Arts grad is going to have to sit there and rotoscope someone's head to pull them out of the shot.
JEFF OKUN: It's not money, it's not time, and it's not people. It is planning and setting it up properly. It means, "Let's talk about it, agree on what we're going to do, and then do it." It doesn't mean, "Let's talk about it, agree on what we're going to do, and then revise everything because when I said I wanted a boat I really wanted an airplane." It goes back to my point about mis-education. A lot of behind-the-scenes documentaries and publications don't go into what really happens - rightfully so, I guess. They don't say: "We designed the sequence, we argued and fought abou it; but then we finally agreed on it. Then we shot it and put it all together, and it was a huge disappointment. So we panicked and had to figure out how to do something else. We shot new stuff, and put that together, and he still didn't like it, so he threatened to fire me" - nobody ever talks about all of that.
Reply
CF: The 'party line' is always "There was planning, there was execution, and it worked - The End."
JEFF OKUN: That's right. And that's a problem, especially when you're working with a first-time director who has read this stuff and believes it! I had a first-time director on a high budget movie, and we'd already missed the turnover day when he came to me, and said:
"I don't like the suits, I don't like the sets, I don't like the creatures - and I know you can fix all of that in the next 3 days."
"Uh, no, we can't, man - that's crazy talk." And he's looking at me like I'm lying to him, like I'm just not cooperating.
SYD DUTTON (Dune): A last sunrise shot in Van Helsing was shot only a month before the movie was released. So good old Bill Taylor went out at four o'clock in the morning and shot runrise after sunrise. But the DP said, "No, let's do this" - and it was already 10 o'clock by then. So we had to make that into a sunrise shot. By the time we got done with it, it looked pretty good. But afterwards, I went to the filmmakers, and said, "Please don't do that again." And they said: "Why? You just proved that you could do it!"
RICHARD HOLLANDER (X2): Unfortunately, or fortunately, we have been very successful at 'fixing it in post.' People in the industry will tell me: "They shot it wrong; then they handed it to us all wrong; and then they asked us to repair it - and we did, and we did a good job. And the bad news is that now they're going to think that's the norm." Well, you know what? That is the norm!"
CF: In past interviews, many of you have mentioned ever-decreasing preproduction and postproduction periods. Why is that happening?
RICHARD EDLUND: Because in development land, you have 4 or 5 writers who contribute to the script: and, meanwhile, the movie's been greenlit - but the script still isn't ready. So this incredibly lugubrious machine lumbers to the production starting gate, and you have a script that nobody feels is ready to shot, but the studio has bought off on it. So the movie takes on a life of its own - and sometimes, it doesn't have a very high IQ!
IAN HUNTER: Here's the problem. The studios used to be run by people who knew how to make movies. But when movies took off in the 1970s and became these amazing profit centers, big corporations decided to get into the movie business. So now we have middle managers reporting to bigger bosses, and they are under pressure to make the product for less. They cut preproduction and post-production, because in the corporate world, they don't understand what those mean to a movie. They just want production, because production equals product. So they go right on shooting, without a finished script. They get something shot, then they go to post - which has also been cut short - and try to assemble it. Of course, the corporations still want all the bells and whistles in the FX - but they don't give you enough time to do it.
MATTHEW GRATZNER (Miniature FX Supervisor: The Aviator): We look at how many weeks we've been given, historically, and every year it's less and less - to the point where we're turning stuff around in 3 months, when it should have been a 6-month schedule. It's crazy.
RICK BAKER (Makeup FX Creator): When I did American Werewolf, people said: "Wow, we haven't seen stuff like this before! What's the new material that made this stuff look so amazing?" But it wasn't the material; it was that we had some money, and we had some time. By the time I did Planet of the Apes, I had 4 months for preproduction, when I should've had a year. The unfortunate thing is, we did it in that 4 months! And the next time, they think, "Well, they did that in four months - couldn't they do it in two?" And it keeps going like that; and it will continue to go like that, until somebody fails.
JEFF OKUN: I visited a movie set recently that's in grave trouble. The FX people are working around the clock, with emergency stuff farmed out all over the place. And they were all saying, "If we just fail, maybe they'll learn that you can't make a movie this way." But nobody wants to fail.
KEVIN MACK: It's an escalation thing. The technology gets faster and better, we get more experienced and have better tools; and because of the pressure to get the work, companies bite off more than they can chew. And they wind up getting it done, one way or the other. But, unfortunately, it comes out of the hides of the people doing the work. As soon as you heroically do the impossible, then that becomes what's expected.
MARK STETSON: We deliver miracles on a shoestring again and again - and then everyone just assumes we'll continue to deliver miracles.
KEVIN MACK: And then there's a backlash. You burn out a bunch of people, and they go work somewhere else. Sony has learned that if they don't keep an eye on that, they lose their best people.
MARK STETSON: I think the next step in the maturing of the industry will be further integrating ourselves into the production process, to the point that we can defend ourselves against that kind of thing, just as any other department might.
Reply
• Reputation: 1120
CF: Increasingly, it seems that FX are being integrated into the production process. Do you find there is now a blurring of the boundaries between production design, cinematography, and visual FX?
BROOKE BRETON: I'm seeing that. It used to be, "Here comes teh VFX team, and they're going to shoot their plates, and then they're going to go away and work on this for a period of time." Now, there is the opportunity to work closely together, because all of these crafts are overlapping.
CF: The blurring of boundaries requires greater cooperation between departments that haven't always seen eye to eye. For example, how is it affecting the relationship between cinematographers and the VFX team?
GRAY MARSHALL: We can be a real asset to the cinematographer; but we are more often seen as an impediment. I got a chance to work with Conrad Hall on a commercial once - he was, of course, one of the great cinematographers. We were doing a lot of greenscreen, and he wasn't having a good time. "I could be back on my island in Tahiti," is what I believe he said. And I was crushed. But it wasn't personal. What I realized is that cinematographers are constructionists - they work to bring everything together in front of the lens. And VFX artists are deconstructionists - "Give us the pieces, we'll build it for you." And the two are often at odds. It takes time for them to realize you're on their side and trying to create a movie just as much as they are.
JEFF OKUN: Part of the problem is that a lot of DPs are very worried about what people are going to see in the dailies. Their careers rise and fall on that, because no matter how educated the studio exec or the producers are, if they see a shot in dailies that doesn't look good because it's a VFX shot, that reflects on the DP - or the DP thinks it reflects on him. So you get a lot of resistance from the DPs if you want to shoot something that might make them look bad. You are asking them to put their careers in your hands, and I don't know many of them who willingly do that.
Reply
CF: Is there closer interaction between the FX people and production designers today?
GRAY MARSHALL: We try to involve the production designers as much as possible. Unfortunately, in many cases they are off the show after production wraps, so they aren't around during post. It would be nice to see a shift in paradigm, where production design is more a part of post. We have a couple of upcoming projects where, in many ways, our duty is more to serve the production design than any other aspect of the story.
PATRICK TATOPOULOS: It's always in the best interests of the movie to keep the integrity of a design all the way - but, very often, you lose track of it. You design something, and then it goes into animation, for example, and it begins to take a different turn. More and more, as a production designer, I'm fighting to make sure that I'm involved all the way through, to maintain the integrity of the design.
KEVIN HAUG: The FX facilities try to hang on to that design integrity with digital art directors, effects DPs - all these people who are really good at matching someone else's work. But, still, they haven't been around to see why the original designers made certain choices, and they can't get ahold of those people because they've already moved on to other movies.
Reply
CF: The ability to change the intent of the production designer or even the DP after production has wrapped is greater than ever now, due to all the tweaking that can be done in post, in the digital intermediate suite. Is that being abused?
KEVIN HAUG: There's always been a kind of polite antagonism between the production designer and the DP on set - but the digital intermediate is where the rubber really meets the road. I've seen directors and DPs conspire to change the color of a set that was very carefully chosen by the production designer. They'll go, "Hmmm, why don't we make it a little greener?" And all of a sudden, everything the production designer spent weeks agonizing over is gone.
CF: With the malleability of digital - both in the D.I. suite and at the VFX workstation - it's now possible for filmmakers to massage a shot or a sequence to death. Has that become a problem?
TIM McHUGH: Absolutely. When it was so much harder to do things, and you weren't going to see the results for weeks, you had to commit to it. Now, the blessing and the curse is that you can see something right away and make a change. You can change it. The director can change it. The producer can change it. Lots of people make changes, just because they can.
HENRY LaBOUNTA: It's been a problem in VFX for quite a while - the directors know they don't have to final a shot until the very last minute. They can go in there and noodle the hell out of it - and that makes it difficult for companies to get work done.
CF: All of this tweaking - is there a danger of making a shot too perfect, of losing the 'soul' of the shot, as Rob mentioned earlier?
ROBERT SKOTAK: I think so. There's an impulse to scrub the dirt out of shots, to buff and polish them, to perfect nature through the computer. But to make our shots feel more real and gutsy, we have to learn to live with the sloppiness of nature. I'm a fan of the ugly matte shot. Not an unacceptable matte shot, but a shot that is just ordinary, that doesn't call attention to itself. Or - something Jim Cameron has done in the past is to look for the not-perfectly framed shot, catching an action that, because it happens so fast, you don't quite capture ideally. There was a nice thing in Signs - the camera comes through a window at a birthday party, and this creature kind of strides by, and you don't quite see it. It's a haunting shot.
CF: We used to hear the expression 'happy accident' all the time; but in CG, there are no happy accidents, because everything can be so perfectly controlled.
ROBERT SKOTAK: And that's what makes it look wrong. A prime example is the shot of the elephant stepping on the car in Jumanji. It was technically brilliant, very well done; but, to me, there was something not quite right about it - and it applies to this idea of perfectly nature in CG. In the real world, with real rampaging, out-of-control animals, that elephant wouldn't step on the car so perfectly. It might just step on the trunk, and the car might pop up and block the view of the elephant. A part of the car would fly off, toward the cameraman, and so he'd duck and the camera would suddenly move to the side.
CF: Peter Jackson did that with the cave troll battle in the first LOTR. It really looked like the cameraman was running around with a handheld camera, trying to get this action while also ducking out of the way of the troll.
ROBERT SKOTAK: That's the kind of thing we should be looking for.
JOHN VLIET: I remember some of the early CG aircraft and spaceships. A lot of the people who had been shooting spacecraft models on motion control stages really hated the look of CG because it was so mathematically perfect. I've been on shows where we had both real aircraft and CG aircraft, and it was amazing to see the difference. In the real aircraft, this guy would be flying along in a so-called straight line, but he would do a hundred different different calculations to keep the plane moving correctly. There would be all these little flaws. But if you told teh CG airplane to fly from point A to point B, it would be absolutely precise, with no deviation. And it would look wrong.
RANDY COOK: It applies to character animation, as well. With the computer, you can move to a finer degree of incrementation than real skin, bone, and muscles will allow. Consequently, we often wind up with animated motion that is smooter and more precise than what you see in nature - and that gives it a kind of fakeness. If you're attempting to simulte life, you need to put in those idiosyncrasies and irregularities that you would find in an organic creature.
BRIAN HUL: I think people are realizing that 'flawless' can mean a lack of style. So FX artists are throwing back in faults as a last step in their comps. "OK, now that it's all perfect, let's mess it up a little bit. Let's put some dirt on the lens, let's shake the camera up."
JOHN VLIET: One of the software programs now has something where you can 'add noise' to a move. You can basically just add that imperfection all the way through. Somebody finally got it. Just the fact that people are now addressing that is a huge leap.
CF: Another component of over-massaging by directors - it seems they want their VFX shots to be absolutely perfect; but some don't apply that same standard to other elements of their film. They beat up their FX companies to get everything right, in support of a movie that clearly had a lot of shortcuts taken with the screenplay and other things!
JAMIE DIXON: I think what happens is - with live-action, you shoot your movie, you blow your millions of dollars, and you get what you get. And at some point, the filmmakers say, "The live-action is what it is, that's all I'm going to get out of it - but this we can still mess with." On one of our projects we spent so much time noodling ridiculous details - truly ridiculous details. One of the most difficult things about filmmaking is the decision-making process. And it almost doesn't matter how early you get started trying to solve a problem, because nobody's going to make up their mind until about the last 6 weeks of the show anyway. The pressure of the deadline is an amazingly clarifying force.
PHIL TIPPETT: The fear factor really hits in postproduction, when the director and the editor are trying to make sense of what was shot - "Oh my God, what have I done?" And it's not just the VFX companies - everybody in the postproduction pipeline gets it. The further on down the line, the worse the intensity. Some people just don't have the temperament or the organizational skills to do these kinds of big visual FX pictures, and everything falls apart at the post stage. And, you see it coming! You see it coming a million miles away!
Reply
CF: We're now 15 years into digital effects technology. Is the initial excitement wearing off a bit? Is the pendulum swinging back, away from CG and toward more practical approaches?
ROB LEGATO: I think we're swinging back to a point where people realize they don't have to replace everything with a CG image. People are saying, "I'm going to shoot a real plate, then let the computer help me move it about the way I need for the shot" - but not totally creating it, not reinventing the wheel. I don't want to create a CG car if I don't have to. I've done it, and it worked, ultimately. But man, it was a big deal to do. It took 4 or 5 weeks of, "I don't know what the hell's wrong with it..." You'd never have to spend that kind of time on a set, shooting a real car. You'd have an instant visceral reaction as to whether it was working or not.
JOHN VLIET: When I was on Mortal Kombat, the producers desperately wanted to do it all CG - like for lava flowing out of this one character. But, CG was teh wrong approach for a number of reasons. It was wrong from a financial standpoint. Fluid dynamics technology really wasn't there yet. They would have been working on this thing for another year! But the guys who held the purse strings were just convinced that they had to have CG; and it was really tough to talk them out of it. We finally got them to go the practical route. We went in and plumbed up a couple of robots, and they looked great!
PATRICK TATOPOULOUS: If you go back a few years, everyone was saying, "We only want to do it CGI." But now, there are new directors coming up - the young generation that you would think would be partial to CGI because they were born into this world of videogames - and, strangely enough, these younger directors are pulling the other way. That's been my experience. I find it very interesting.
ALEC GILLIS: We're starting to get directors who like model spaceships, real explosions, practical FX. Of course, we're listening for what we want to hear, to a degree. But I do think directors are coming back to live effects a bit.
ROBERT SKOTAK: I've run into a lot of directors recently who have said, "Is there any way we could do this non-digital? Because I don't like that look." I see the pendulum swinging back to what it really should be - a mix of all these things.
NED GORMAN: If we learned one thing from Peter Jackson and the LOTR trilogy, it's that you can achieve great success by mixing different techniques. He used every trick in thet book, and saved the digital punch for when he really needed it.
STAN WINSTON: He used techniques that we'd all seen before - just never put together in quite that way. He mixed up digital matte paintings, men in makeups, digital characters, miniatures. He had teh vision to use those tools to tell an amazing story. You have to open up the toolbox and be comfortable with everything that is out there; but, as Phil Tippett always says, if you can do it with a hammer, use a hammer.
CF: Jim Cameron is another director who mixes it up. He's always been a proponent of using everything from high-tech tools to old, tried-and-true VFX techniques.
VAN LING: That was one of the greatest lessons I learned from him. One of the phrases he always used was, 'horses for courses.' Use the right horse for the race you're going to ride. I've seen people dismiss the older techniques as quaint and old-fashioned. But it would be very sad to see the FX world become a world of CG artists who consider themselves CG artists, rather than artists who are willing to use any technique.
CF: Won't there be all-digital artists who rise to the top, who are artists in their own right, even if they work only in CG?
VAN LING: I wonder, though, if the technology doesn't make it more challenging for people to be artists. It can encourage laziness, because the art can be an automated process. You can buy software packages today for flying logos, for example, and all you have to do is swap out where your logo goes and choose one of a hundred preset motion paths. There's a lot of canned stuff, too - and you see it all the time. You can be watching something on TV - 'Oops, they're using Pyromania. They're using Pete Kuran's fireball explosions.'
TIM McHUGH: I find myself decoding the FX packages or plug-ins in FX movies all the time.
"Oh, they have the new hair plug-in that does that."
Or, "Oh, the subsurface scattering on the skin."
Or, "Oh, they've got that new particle generator."
I sit there, knowing who worked on the movie, and I can tell you what software they own by seeing their effects.
VAN LING: There are a lot of people who just use the defaults. And that's not art. That's Chinese menu work - pick one from column A, one from column B, put it together, and you've got a shot. It's a struggle for the VFX artist - especially the up-and-coming ones - to ignore all of that preset white noise and learn the basics of how to tell a story, how to fit a shot into the overall.
CF: As far as digital technology has come, what limitations does it still have? What tools do you wish you had in that toolbox?
JOHN KNOLL: Nothing is ever fast enough. Our render times have remained fairly constant since The Abyss! They're always at the pain threshold. If you look at how long it took to render one frame on The Abyss and what it takes to render a typical frame on the new Star Wars, they are fairly comparable. Of course, that's because we're doing far more complicated renders - denser, wiith more sophisticated shading models. If we had tried to render that back in 1989, it would have taken a month to do what we're doing in an hour now. But people are always trying to figure out a way to make the process faster. Faster is always better.
JOHN DYKSTRA: It's still really tough to do random stuff in the computer. The computer doesn't like to do random stuff; it likes to do organized stuff. So that's a limitation - even when it's done really well. I was astonished at the work in Pearl Harbor. I thought the explosions, fire and smoke were phenomenal in that movie, some of the best-looking procedural stuff I've seen. But it is still really hard to make that kind of random stuff - water, smoke, clouds, fire. Those things are a huge challenge for digital imaging.
ALEX FUNKE: There's also no good digital lighting program. Software designers have never sat down with real cinematographers and said, "OK, let's see how you do this and let's design a program that actually works the way the cinematographer lights." We're way behind on that.
ERIK NASH: Given the complex nature of CG lighting and how non-intuitive and non-interactive it is, I don't know how the artists do what they do. Artists who really understand how digital lighting works do remarkable things with tools that don't mimic real-world lighting tools in any way, shape or form! It's liberating, in that you can do things you could only dream of doing in the physical world; but the non-interactivity of it, to me, would be very frustrating. That's one aspect of it that still seems incredibly arcane - how slow it is to see what it is you're really doing when you're lighting a CG object.
RICHARD HOLLANDER: That's what we need - a way to realistically light environments and get results that don't take 40,000 years to render. It seems to me that if I was to put a synthetic creature sitting next to you in this room, since the lighting is not changing for all of our shots, no matter where my camera is, there shouldn't have to be too much human intervention on the lighting of that creature. We are striving for that. We want to understand this room from a light-sourcing POV - and we want that to happen fast. That's the next layer - to light 90 percent semi automatically, then tweak the rest.
CF: It seems ilke lighting, since it is physics-based, ought to be something that could be proceduralized fairly easily.
ERIK NASH: Yes, it can all be quantified. The problem is just the complexity of the calculations, the number of calculations necessary to mimic what happens in the real world. All of that is so staggering, to be able to run that simulation in anything approaching real-time is just currently beyond the realm of possibility. But, as computer power grows, interactivity will improve in leaps and bounds.
HENRY LaBOUNTA: The past couple of years at Siggraph, it seems like more than half the presentations are about real-time issues. The stuff Nvidia is doing with Gelato is brilliant. It's a whole new package they've come out with that is basically intended to do RenderMan-style rendering in real time.
PHIL TIPPETT: That would be ideal - to be able to go in and manipulate a thing with all of the lights on it, in real time. When I was doing stop-motion animation, I found that lighting had everything to do with how you moved the character. And when you're dealing with a fireframe thing, you don't have that. You're just dealing with movement in the abstract until you see the thing rendered out.
RANDY COOK: That's the biggest disadvantage in computer animation, versus stop-motion or some of the older techniques - you don't see what the light is doing as you are animating. On LOTR we tried to get our lighting setups as soon as we could and incorporate them into our work. We also suggested lighting positions, so that our character's face would 'find the light' at the right moment. But most animators today don't get to do that. They're not allowed access to that part of the the toolbox, which is unfortunate.
ROB COLEMAN: I think we're going to have that real-time rendering in the next 3 to 5 years. It may not have every level of specular highlights, or subsurface scattering or all those beautiful things that they do to make realistic skin these days; but it would certainly be good enough for me to judge the animation at a level beyond what we have now.
CF: What is the role of previsualization in VFX now?
IAN HUNTER: In some respsects, previz is a way to get some of that lost preproduction back into the schedule. If you can sneak that in by calling it 'previz' and everyone thinks that's a sexy term, that's great. It's just preplanning, which you should have been doing in the first place.
ROB LEGATO: If you do it correctly, previz allows you to go through the mechanics of a scene beforehand and solve problems that would be harder to solve later. It gives you a sense of rhythm. You can cut something together with music and sound FX, and it starts to feel like a movie. It makes you a better director, because it gives you a couple cracks at directing a scene before you actually have to shoot it. I think, in the future, we're going to have a previz artist on all productions, as a matter of course. People are becoming more dependent on it.
ERIC BREVIG: I think previz is the best application of digital technology we have in the movies. FX are so expensive, if you can save the slightest misstep by solving a problem in a cheap video animatic, it's a great thing. Especially when so much of your shot is synthetic. Unless you have this road map of what the shot is supposed to look like when it's done, you can't go out and shoot any of the elements. If you're doing a big orbiting helicopter move around something you're gonig to put in later, you'd better have some indication of what the shot is supposed to look like before you shoot live-action. Animatics are very important - if the director sticks to them, of course.
CF: Has previz extended beyond its initial purpose?
COLIN GREEN: It has changed a lot. Initially, previz focused on addressing technical challenges. How big of a greenscreen do we need? Is the crane long enough to get the shot? And we would do some creative development in the midst of that to see what it all looked like. Now, we're doing almost exclusively creative previz. The technical questions aren't even asked until the very end. It's there purely to define the visual approach.
CF: Is previz generally accepted now? Or do you still have to sell the concept of doing it?
COLIN GREEN: Early on, we'd have to do a lot of explaining about what the benefits were. Now, people just come to us, and say, "We need previz," and we have to ask them: "Why? What are you expecting to get out of it?" The term is thrown around for a lot of different things. But we always make sure that the process doesn't evolve into some sort of fluffy eye-candy phase at the beginning. We want to make sure it is a decision-making vehicle for the director and crew.
CF: From the standpoint of FX work, is previz always a good idea?
MATT BUTLER: I see both sides of the coin. On the upside - if I was a director, I would want previz, because I could cut it right into my movie, and it would help me plan. Given a choice to previz or not to previz, I'm sure every director would say, "Sure, let's do it." The downside is the exhaustive resources that can be spent on it. Previz can be endless. Your resources are limited, and it can be a negative thing to spend some of them - a lot of them, in some cases - on previz. leaving you with fewer resources to do the final work.
JAMIE DIXON: We suffered badly from over-previz on Riddick. We were noodling shots forever in the previz stage. It killed us, because we couldn't get things signed off, and we wound up wandering all over the place.
CF: So, instead of acting as a quick-and-dirty intermediate step between storyboards and final shots, previz can be abused a bit.
MATT BUTLER: That's a good way of putting it. Previz needs to be something that's turned around quickly and is done in a disposable manner. Often, directors will say, "I can't judge a shot when it is done at that low a level - I need to see it more fleshed out." And I understand that, in some cases. In other cases... For example, how do you previz a volumetric rendering of a cloud? It is so abstract, the previz is useless. So you've spent all this time previzing a cloud, when you should have been doing the final shot. You can wind up in a situation in which you've used up nearly all your time and money - and all you've done is previz.
CF: If you previz in-house, you risk wasting time and money. But if it is done somewhere else, at a place that specializes in previz, doesn't that take the VFX people out of the creative loop?
COLIN GREEN: I'm sure that's perceived as a problem occasionally by the VFX companies; but if you approach it in the right way, vendors that get involved later are generally quite happy to get the data we've created. If that process is done with good, open communication, and you hand over solid usable assets, everyone is happy.
MATT BUTLER: It's thte director's dream anyway; and if that outside previz company is getting what is inside the director's head into the previz, it just gets you further along. But if you are completely locked to the previz, then you might as well not do the show - because, what are you offering? We are artists and technical experts, and if you take too much away from us, you are being inefficient. You're hiring an expensive place like Digital Domain, but not letting us do what we do well.
JAMIE DIXON: Another thing that can happen is, because the previz artists are working at such a fast pace and experimenting with things, you can end up with a tangled mess of details that is very hard to manage when you actually get into a shot. Stupid little technical things, like maybe you cheat a camera move because it looks all right in the previz - and that's all that matters when the director is standing over your shoulder, saying, "Yeah, that's great." But then, the animators doing the final shot get saddled with that mess.
Holy shit. Thanks for transcribing this very interesting interview...but typing all this is borderline insane.
But hey, if you like it - more power to you! Do you want to type my dissertation by any chance? I would give you a cookie. Or two.
CF: Presumably, if the FX supervisor is involved in the previz process, he or she can keep that from happening.
ERIC BREVIG: I get involved, as needed. I lay down some rules, because I'm going to have to go out and shoot live-action to match whatever it is they're creating in the previz. They can't use imaginary cameras that I could never use, for example, or mix focal lengths in the same shot.
ANDY JONES: It's essential to have people doing the previz who are really good with cameras; because if you don't, they'll animate the animatics in strange ways, and it will start to look very CG. A lot of camera work can be established in previz - but it is important to develop camera work that is actually going to be possible on set.
JEFF OKUN: Sometimes they give you an FX shot that they've created, and you look at it and go: "Whoa! You want this? The perspectives are wrong. It looks good here because of the low-resolution, but when we do it big, you're going to see a problem." But they make you put it together that way, and then they yell at you because it looks like crap. Those 'cheap' shots can end up costing a lot of money if you are frame-by-frame matching. Fortunately, some directors do realize that previz is only a template.
JOEL HYNEK: The directors who use it best are those who realize it is only a departure point - they have the option of using it or not, and going from there. That's how Rob Cohen uses it. He loads the previz on the day, and says, "OK, that's what I want to do," or "Nah, that's good, but we can do it better," and departs from that.
It passes the time at the store and it forces me to really read the stuff again without skimming. Not a bad price to pay.
Three cookies.
Alright, 2 and a Half.
CF: Any other good applications for previz?
PADDY EASON: It can be a sales tool, if the FX work hasn't been awarded yet. On Big Fish, we mocked up some previz of an airplane flying over a nighttime Korean landscape, with guys jumping out and parachuting to the ground. We offered that up, and the production liked it; so we just worked the previz out, and it became final rendered stuff that's in the movie.
CF: Is that happening a lot - taking previz files directly and using them as the basis for the final shots?
COLIN GREEN: We've done sequences where you can almost superimpose the final on top of the previz and not see a significant diffrence.
NATHAN McGUINNESS: I think it's great when you can take the previz and move it into the actual shots. It's the same artist working on the previz who's working on the final shots, which keeps you from wasting time.
MATT BUTLER: It seems like a good idea to use the same package for the previz that you are going to use in the final animation. It seems like a good idea to use the same artist to do the previz and the final shot. But, again, is that an inefficient system? Is the artist who is doing the final shot necessarily the best previz artist? I don't know. It's a conundrum. It's like Communism - in theory, it sounds great, but in practice, it isn't very successful!
JAMIE DIXON: I think it's safer just to treat previz as this completely disposable thing. Do it quickly, get the idea you need, get the basic timing that you need from it, then sit down and do the shot for real.
CF: Previs is also sometimes used on set as a through-the-lens visualization tool - to help directors, actors and camera operators on a bluescreen stage see the physical geometry of a digital set. How valuable an application is that?
ALEX FUNKE: Certainly, if a live action sequence is being done where large amounts of digital set are being incorporated, it is tremendously valuable to be able to see what the final composite is going to look like. But I would be reluctant to say to the camera operator, "Here's the previz, and this is how we have to shoot it" - because, that operator is an artist. The previz may be a good guide, but why do we have this accomplished camera operator if we're going to tie him down to a digital image?
COLIN GREEN: The main backlash we've had from the beginning is that seasoned grips and camera people are quick to say we don't know how to do what they do. If we have a digital crane in our previz, for example, their attitude is: "What do you know? How many cranes have you pushed around on set?" So we have to present ourselves as humble problem solvers, rather than dictators, dictating the way a crane gets used. But once we've created that relationship and they realize that we can help them solve problems, we become a good resource for them.
CF: Do studios ever use previz as a feasibility study, to determine whether or not to greenlight a film?
COLIN GREEN: That's definitely increasing. Studio people and investors want to see what's special about a project, and previz is a tool that can help showcase that earlier in the development process. Ideally, we'll get to a point where we'll make a totally CG movie for packaging purposes; and then the studios will be able to decide how they want to develop it - as a live-action movie, CG movie, TV series, or game.
CF: What about 'post-viz' - a term we're hearing more often lately?
COLIN GREEN: We're getting involved in that more and more. There are sequences and elements that have been shot - but how do you design and develop the cut when 80% of the screen is green? We fill in the green parts, so they can tell the story and cut the sequence.
CF: So previz is no longer just a preproduction tool. It is being used in production, all the way through postproduction.
COLIN GREEN: Yes. For Matrix, we were involved up until the last phase of postproduction, continuing to tweak things and generate new ideas.
KEVIN HAUG: On Panic Room, David Fincher wanted to previz the entire movie - to literally previz every single shot as an experiment in another way to make movies. Had everything gone according to plan, he would have been able to shoot really fast, because he would have already prevized every shot and cut the movie. He could have walked down to the set, and there would have been markers on the floor where every setup was going to be, and he would have been able to just go from one to the next with the actors - especially since this was a house and, presumably, everything would be lit the same.
CF: Interesting idea.
KEVIN HAUG: It didn't work out, for a number of reasons. But I think that, on some level, Panic Room is a watershed VFX movie that will never be seen as a watershed VFX movie. It did things that have reverberated out into other movies. And I think the standard for what to do with previz changed irrevocably with Panic Room. It was no longer good enough to do previz just to show the director what he might get in his FX shots some day. It has to be something that's useful to the grips and the AD and everyone on set, or there's just no point.
Reply
CF: Let's talk about motion control (MC). When Cinefex first started publishing, motion control was the hot new technology. 25 years later, it seems to be waning as an FX technique. Given the advances and elegance of new tracking software, is there a future for motion control?
ERIK NASH: I think the frequency with which motion control will be used will continue to decrease, as it has over the past 10 years - but I think we will still need it. One place where MC will continue to be used is in twinning situations, where you've got one actor portraying multiple parts.
JOHN DYKSTRA: Or if you want to do a splitscreen, where a raging fire is on one side, and a highly paid actor is on the other, and you want to make the two appear as if they are a foot from each other - MC works great for that. If you're going to shoot miniatures in multiple passes and then combine them in the composite, motion control is a terrific way to originate those elements.
ERIK NASH: There are also offshoots of MC that are going to play bigger roles. On I, Robot, we used Joe Lewis' Encodacam system, which uses a Kuper computer and encoders on the camera crane and camera head to feed positional information into a 3D package that renders a virutal set real-time. As virtual sets become more and more a standard part of the filmmaker's toolbox, being able to frame shots on set and replace that blank greenscreen with at least some representation of what will ultimately be there will become absolutely mandatory.
KEVIN HAUG: A classic example of what MC can do these days... I was walking out to the set on the 2nd reshoot on Panic Room, and the 3 MC guys we'd been using off and on throughout the show were talking towards me. "What are you guys doing here? I didn't bring you in today." But they weren't there for VFX, they were just doing the move. David Fincher had decided that he could previz, export the move, have somebody come in to slap the MC track down, and then already have the move plugged in so that once he'd set up point A and point B, the move would work just like it did in the previz, and he could start tweaking it.
JOEL HYNEK: MC definitely has deviated from its traditional role. I just got off a fairly elaborate setup for Stealth, where we had the front end of our futuristic airplane on a gimbal inside a huge greenscreen stage. We shot the gimbaled airplane with the actor in it, sometimes using shots that were based on our previz. And then we flew our Spydercam in real time as the plane went through its gimbal motions to get a sense of actual flying shots. We knew that most of the plane was going to be replaced with CG, but this gave us a way to get the actor actually moving and reacting to forces. And the whole setup was driven from a Kuper system.
Reply
CF: Going back to its more traditional uses - what are the disadvantages of using MC as it has been used for the past 20 years?
JOHN DYKSTRA: Well, it slows things down. If you are shooting a page and a half a day, bringing in motion control equipment can take that down to 3/4 of a page - and that's a big deal. An expensive deal. So, I tend to track after-the-fact. If I can give filmmakers the flexibility to shoot, to point the camera wherever and however they want, and then track it afterwards, why wouldn't I do that?
GRAY MARSHALL: But if you can get it to work in the production flow, MC can be a much more cost-effective way to do things. There are now systems that don't require a separate team - a MC equivalent of a HotHead. These systems allow the regular camera crew to integrate it as part of their package, just as they would the rest of their technical camera gear.
JEFF OKUN: I'd much rather have MC on the set. But, sadly, it is perceived as slow and cumbersome by the people who pay the bills. Maybe if we came up with a better term for MC - 'real-time digital replication' or something...
MATT BUTLER: One thing I love about MC is that you get the approval of the director on the set, and there is no argument. That's better than the incessant tinkering that can go on in post. So, MC has its advantages for us in that regard; but, in terms of what can be done synthetically, after the fact, sure, virtually anything can be tracked.
ALEX FUNKE: A caveat to that, however, is that you must have very good on-set documentation. Postproduction tracking is only as good as the surveys that are done on stage. We can get rid of the MC and the tracks and all of that stuff, and do it in the computer - but we've still got to have that accurate documentation of the structure of the set.
MATT BUTLER: Lens mapping makes a huge difference, too. Typically, a crew will shoot with the same set of lenses, and there can be significant defects in them. If you can get your hands on those lenses, you can actually record barrel deviations and make a remarkable improvement in your matchmoving - because the real camera and the digital camera are the same.
CF: Has post tracking become relatively easy? Or is it still a somewhat painful process?
PADDY EASON: It's become much easier than it was, particularly with the advent of software like boujou. Things that previously would have taken weeks to address can now be done almost as a matter of routine, in a few hours.
BRIAN HUL: And, knowing that tracking isn't going to take up a huge amount of time in post, people feel more free when shooting plates. I'll often offer up handheld to a director on set; whereas, in the past, that's the last thing any VFX person would offer.
ERIK NASH: On, I, Robot, the production team insisted that there would be no motion control on set. We didn't quite make that; but for all of the robot shots we did at Digital Domain, we derived camera moves 100% through image-based tracking. We have a great piece of software and a great system in place in terms of surveying the sets, and then tracking from the images to sort of reverse-engineer what the camera did. That became the foundation for all of the set extensions and CG character work that followed. It's become a key part of our entire process, being able to derive camera moves from the images themselves, without any sort of encoders or MC stuff on stage. It means we can do Steadicam shots, handheld shots - you name it.
CF: Will matchmoving and tracking continue to get better? Or is it as good as it's going to get?
ALEX FUNKE: I think it is just in the Model-T stage - competent, but not a Ferrari. There's a lot to be done, and a lot that is being done. The only way to make it better is to get it out in the field, use it, then listen carefully to what the users say.
RICHARD HOLLANDER: Tracking software is still on its ramparts - it hasn't reached that plateau yet. There's an awful lot of information in that picture in a single frame, and we're only using some small percentage of that information for our 3D tracking. So the tracking gets you part of the way, but then you still have to fudge it. I measure the success of it by how many times it has to go back. It is too often, still - and we have a pretty good record in terms of tracking. How about getting it right in the first place, so you don't have to fudge it?
Reply
CF: One of the things that the virtual camera has introduced to movies is the 'impossible' camera move - moves that could never have been done with a physical camera. What are your thoughts on this new style of camera movement?
PHIL TIPPETT: I think it's a trend that's going to date everything. We'll look back on this stuff in probably less than 20 years, and we'll have a similar experience to looking back at the 1925 Lost World. There will be various levels of skill that we can appreciate, but we'll see the technology for what it is.
ALEX FUNKE: It's analagous to doing a camera move on a miniature that is impossible on a full-size set - at some level, the audience is pulled out of the story because their brain is going, "Wow - how did they do that?" And if they're thinking, "How did they do that," then they're not thinking about what the director is trying to say.
VAN LING: There's this whole 'because we can' technological imperative. 'We can fly the camera from here to anywhere, so let's do it.' People love doing powers of ten shots, doing snap-zooms into microscopic views, and so on. Those are all great techniques, but they can quickly become cliche. The first time you saw bullet-time - amazing. But before long, you started seeing it in every commercial, and it didn't have the same impact.
ROB COLEMAN: Putting the camera inside the computer has completely changed the look of filmmaking. Years ago, you'd have to cut around an action sequence because you had a stuntman doing some things, and then you would go to your main actor - you were forced to cut. Now, you can do a morph from your digital double to your main actor right on screen, while the camera is moving 360 degrees around him. But is that always a good thing? Have we forgotten the roots of filmmaking? When I see student reels, it's the camera work that drives me insane. There is no understanding of the camera, and some of that has crept into films.
TIM McHUGH: I get demo reels from animators - and they're videogames! They're not VFX shots. They're robots skating down the hallway blowing things up. They're all first-person shooter kind of demo reels.
Reply
a generational thing. It bothers the older generation, but not the younger one. In the 1930s, VFX shots had to be locked-off. If there'd been camera movement in those shots, the 1930s audience probably would have been freaked out.
ERIC BREVIG: That's a good point. They wouldn't have been able to process it, because it wasn't part of their visual vocabulary.
CF: So if this unrestrained camera is the new look, is there any reason VFX artists need to have an understanding of traditional art, composition, and photography techniques?
HARRISON ELLENSHAW: Well, yeah. Not to be an old curmudgeon about it, but a big part of my training was taking photographs, understanding field of view, different lenses, different sizes of negative, perspective, horizons, vanishing point. I was lucky. I had a mentor, Alan Maley, who was really good at all of that stuff. You must have that basic knowledge - at least to make a matte painting look right. There's no slider that says, GOOD COMPOSITION - BAD COMPOSITION. And composition is so much a part of it.
BILL TAYLOR: One of the virtues of the old, locked-off matte shot was that the composition was static. So the composition could be designed to lead the eye toward the important elements of the scene - toward the live-action if the live-action was very small in the frame, for example. Whereas, in a moving shot, especially a shot that's moving in 3D, the composition is changing constantly.
HARRISON ELLENSHAW: Composition is becoming a lost art. There are a lot of people at the entry level, and above, who just don't know what the hell they're doing.
CF: Are the rest of you noticing a general ignorance of these principles among some CG artists?
KEVIN HAUG: Oh, yeah. There's not a show that goes by where I don't have to tell a 3D animator why what he's done doesn't work, because he doesn't understand the rules of perspective. Once the software cheats the perspective for them, they never learn it. I rarely get aggravated at animators, but the first time this happened, I got aggravated. It was a guy who was a hotshot in the video world and he was pulling an arrogance trip on me, and I finally said, "You know, they worked out all the basic rules of perspective in the15th century - how did you fucking miss it?" When you find an animator who is getting that that kind of money and doesn't know the first 3 rules of perspective, you want to smack them.
JOHN KNOLL: It used to be that just about everybody in the business spent some time out on the stage, shooting stuff. But I see a lot of people doing work in VFX now who have never had any real experience with cameras. I think that people who have never had that experience, have never had to make the tough call about where to set the exposure on something, tend to put together unrealistic composites. You'll see an interior with a wnidow to a daylight exterior, and they'll make the outside the same brightness as the inside. If you've ever tried to do that for real on a location, you know it's really hard to make that split. Another thing - there isn't infinite depth of field. So you see mistakes like a sharp background that should be soft.
CF: Some people defend those things as 'artistic choices.' They'll do something that is patently incorrect from a photographic point of view; but their view is that those rules don't apply anymore.
JOHN KNOLL: Well, that's a choice. But my feeling is that if you do something you couldn't do with a real camera and a real object - a camera passing through the propeller of an airplane, or a shot that booms from a thousand feet down into a closeup in a second - that is going to stand out as false.
BILL TAYLOR: It depends on how it fits in dramatically. You can do an impossible shot in the course of a very tightly cut, fast-paced car race - such as shots we did in The Fast and the Furious - and it doesn't stand out. There's a shot in the race that is literally impossible. We got in between 2 cars, when there was no way we could have put a camera operator - or even a robot camera - in that space. If we had done that same shot out of that context, people would have said, "Wait a minute - where was that camera operator, and how is it possible for a car to go within 2 inches of the lens on the left side, and then an instant later go within 2 inches of the lens on the right side?" But it doesn't stand out in the context of that race sequence.
JOEL HYNEK: That's the trick - you have to be so involved in what is going on in the movie, you're not even aware of what the camera is doing. Even if it's doing something that could never be done for real, you don't really take that in.
JOHN KNOLL: I'm just not a fan of things that couldn't be done for real. When I'm doing a digital shot, I think about how we could do it if we were really shooting it. Would it be a helicopter shot? A crane? A car mount? OK, if it's a car mount, there would be vehicular vibration on it, so we should put some low-amplitude, high-frequency vibration on the shot. OK, this is a helicopter shot, which means you can't get too close to the subject, so there's got to be a long lens, and there's got to be a little float as the operator tries to follow the subject. Those little things add a lot of realism to a digital shot.
