John Knoll Talks ILM, Disney, ‘Star Wars’ and Tough Times in the VFX Industry
Out from under George Lucas’ wings, ILM’s legendary visual effects guru takes the helm as Chief Creative Officer.
By Dan Sarto | Wednesday, January 8, 2014 at 6:04pm

Few people in the visual effects industry talk with as much clarity and perspective as ILM’s John Knoll. When he speaks, he truly speaks from experience. A career spent supervising visual effects on many of the most successful feature films in history provides only a glimpse of his impact on the world of computer graphics, filmmaking and visual development. Along with his brother Thomas, he created Photoshop, arguably the most important digital creative tool in history. Oh, and add in an Oscar for Best Visual Effects for his work on Pirates of the Caribbean: Dead Man’s Chest along with four other nominations. Clearly, he’s been busy.

But as they say in business, “That’s great, but what have you done for me lately?” Well, last May, John took on the role of Chief Creative Officer at ILM, ushering in a new era at the venerable visual effect studio as it settles into the fold of its new corporate owner, The Walt Disney Company. While the announcement of a new slate of Star Wars movies should ensure the hum of ILM renderfarms for many years to come, these days not even ILM is immune to the economic pressures that continue to cause such turmoil across the global visual effects business landscape.

I recently had a chance to sit with John, who shared his insights on the ownership change, new Star Wars production, continued problems within the visual effects “business” and how the loss of the visual “wow factor” in films is forcing directors to tell more compelling stories.

Dan Sarto: You’ve taken on a new role at ILM as Chief Creative Officer. What all does the job entail and how does it differ fundamentally from what you were doing previously? What does one of your days look like?

John Knoll: I’ve worked on a lot of really big films that required breaking the show up into different sequences. At that point, I’d put leads on in a sort of hierarchical structure. So in a way, what I’m doing now is just one level up. Now I oversee all the company’s creative projects, which are kind of divided into the seven or eight shows that are in production at once.

My day starts off with dailies. I make the rounds through the dailies of all the shows in production, just seeing what's going on. So, I stay in touch with the state of all the shows, what they are struggling with and where they might need some help, to make sure they get all the resources that they need. I’m there to help, to offer up suggestions to help the project, to encourage folks to share with their peers, something that's been a very successful part of ILM culture in the past. I am trying to keep alive this idea that good ideas are good ideas no matter where they come from. I used to run my dailies where I created an atmosphere in which people felt free to make suggestions. If they had an idea and if that was a cool idea, they should go ahead and say it. If it's a good idea we are going to use it.

One of the strengths we have as a company is our big bench of talent - all the supervisors - and that we are stronger working together than anyone is working individually. If we’re smart we’ll make use of all that talent. Something I used to do all the time was if I was struggling with a shot, and I had lost my first impression and had seen the shot too many times and didn't know what to do with it, I would get the guys together, get the group together and we would brainstorm it. We’d do those sessions periodically, something I am now trying to encourage everyone to do. It's interesting the most established supes [visual effects supervisors] are the ones that are most comfortable doing that. It's usually the younger guys just getting started that feel the most vulnerable and are least likely to want to do that. I really try to encourage them to do that, to take advantage of what we have here.

DS: What are some of the big challenges you’re facing?

JK: One of the things that I am trying to do is provide a little higher level and forward thinking guidance to the company both in terms of the projects and technological R&D we are pursuing, with the goal of pushing us towards a particular set of capabilities and skills. We have a long and storied history of technological experimentation on shows, where we will try this technique or that tool, but it's never really been coordinated at any kind of high level with an objective in mind. It's sort of hard to afford to do that anymore and so I am trying to provide a little bit of a large-scale structure. We want to keep trying these experiments but they have to be with some kind of end goal in mind.

What happened in the past was there would be some experiment, there would be some promise but then it wouldn't be followed up on the next show because that crew had a different idea about what they wanted to do. There was a bit of chaos in that regard. Things would sort of meander and not really progress as effectively as what I am trying to do with it now. Now, each experiment that's happening on a show is usually approached with the thought that, “All right, if this works then we’re going to pull that back in [to the studio’s core capabilities]. This show is going to tackle this problem specifically because we want to build up our competence in this area.” The next project along the line needs that expertise and is actually going to take that capability and push it a little further forward.

So I’m thinking a little bit more about where we want to go as a company. That's a little bit of a new thing for us, partly because of the historical roots of the company. ILM was originally formed just to do Star Wars. John Dykstra was running the show and he was top of the pyramid. As we gradually expanded to doing more than one show simultaneously, each using individual supervisors, the company was still small enough that there weren’t these more fundamental direction and resource contention issues. As the company got bigger and bigger, at any one time we were typically working on four or five big projects. Next year, when there are probably seven shows on the slate, all pretty good size, it gets more important than ever to start having a little higher level vision.

DS: In my estimation, ILM has always been the gold standard in visual effects production. But the business is a lot more crowded and competitive than it used to be. How does your studio differentiate itself in today’s visual effects service marketplace?

JK: We of course have a very long history, relatively speaking, in the business. We’ve been around since 1975. We have an unbroken track record of always delivering and always delivering on time. Our clients feel an assurance, a peace of mind that they don't have to bail us out. On most shows, client contact becomes more and more frequent as you approach the delivery deadline. During the early phases it may be once a week and then twice a week and then eventually, it’s every day. I have been on a couple of projects where right near the deadline, clients started canceling our review sessions. At first our feelings were a little hurt, like “Jim Cameron doesn’t want to talk to us?” Actually, it was the exact opposite. It was a vote of confidence, like, “You guys are the least of our problems, just keep doing what you are doing. It's fine.”

DS: That's a great position to be in.

JK: Yeah, it's good to be in a position where we provide some of that peace of mind, that we are the least of their problems.

DS: You cannot talk about visual effects these days without talking about the rough business climate. There seems to be a tug of war between the studios and the visual effects houses, with considerable finger pointing and frustration. Are there any answers? Is there much more shake out to come? Where do you see things headed?

JK: All in all, the picture is a little grim right now. Unfortunately, there are a lot of companies that engage in really foolish and short sighted business practices. You can't just blame the studios for wanting to pay less and less for the work. What's happening is that when a low target number goes out, somebody says, “Yes.” A lot of times companies are saying “Yes” to a number that they know is going to be below their cost. They know they are going to lose money, for a variety of reasons. Sometimes they don't know that's what they are doing. They completely misunderstood what it is they are bidding on and their bid is based on a wrong assumption. They are going to take it in the shorts because of their naivety. Sometimes they are doing it for strategic reasons, thinking, “We will buy this first one and then…”

DS: That never works as intended…

JK: …which you know, as soon as you establish the precedent that you value the work like that, that's where they’re always going to want to be. It’s bad for the industry to do that in the first place.

DS: They figure that maybe they will get to bill for certain overages…

JK: Maybe. I’m not sure what goes into that kind of thinking because I am not really willing to engage in that kind of stuff. I am very opposed to that kind of “bid below cost.” I think it's destructive to the whole industry so I pledge that we will never do that. But, you know, for every company that gets burned by doing that, that may vow never to do that again, there is somebody else that is willing to do it. It seems like almost every round of bidding that happens on a project, somebody somewhere is ready to do it for below cost. What that does is set the bar for the going rate for that work well below what most company’s cost is on it. That's just hugely destructive to the industry. That's poisoning the waters for everyone.

DS: But do you think part and parcel to that, the state of the art of visual effects is so tremendous, that it’s harder for companies to differentiate their capabilities, to show why it’s worth paying more for their services? Do the studios figure the quality of work is so high for the most part that there’s no big difference going with a lesser bid to save money? Has the perceived value of the quality and artistry of the work itself diminished in the eyes of the studios?

JK: Yeah. I’m not sure I see quality valued as much anymore. It really comes down to where they [the studios] have to make it [a show] for their number and if they have to live with a little lower quality, that’s fine. [They feel] That’s fine, it usually doesn’t affect box office, so it's not important to really care about that. It’s frustrating, but it's a business. The market that we’ve chosen to address is the high end, the highest quality. It's sad that so few people want to pay for that anymore.

DS: Speaking of “the business,” how does the Disney purchase affect the studio? Is George still involved at all?

JK: Well, George has sold the company so he is not involved in ILM anymore. We belong to Disney now. The thing is, ILM is a bit of an unusual company for Disney in that we are the only entity that provides services to all comers. Our business model is based on being able to work with the best filmmakers regardless of what studio they are affiliated with. It wouldn’t make sense to narrow the business down to just being a Disney house. There isn't enough work to support our facility that way. So we really need to be firewalled off from Disney because our business model is based on being able to work with anybody at any studio. That means Disney can't look into ILM’s internals, they can't look at the IP of the projects that are being worked on because we have to keep that very confidential. They can't look at any finances of the projects going through the facility because that's revealing competitive information. We really have to keep them at arm’s length and we have to represent kind of a black box to them.

We also have to be really careful and make sure we have equitable distribution of talent across all our projects. We can't be seen as providing favoritism to Disney projects. All our shows must succeed. They all need to have great talent. If there got to be a perception that ILM puts all the best artists on Star Wars or on Disney projects then all the rest of the work would evaporate. We really can't have that happen. Every creative team we put together has great pride in the project they are working on. If I’m working on Pacific Rim, which is a Warner Bros. project, I will put my whole heart and soul into that project. I’m always pushing for that to be the best work that we can do. We are never going to do a lesser effort.

DS: At Disney’s recent D23 event, Bob Iger spoke and said this is the greatest period in Disney’s history, that we have Star Wars, we have Marvel, we have Pixar. The collection of IP and creative talent is quite formidable. There are many people who are happy the Star Wars franchise is going to get more attention. There are also many people worrying that now it’s under Disney’s creative control. What can we expect from the Star Wars franchise? Where is it headed?

JK: Well, let me say that Star Wars is extremely important to Disney, to Lucasfilm and to ILM. We appreciate how important they are and how much is riding on these being really great films that do really well. So there is great excitement about the opportunity here. The creative teams that are in place to do these [next Star Wars films], I’m feeling pretty good about them. I have every expectation these are going to be awesome.

DS: So you are bullish on the future of franchise?

JK: Yeah. There are lots of good choices being made.

DS: Do you think the ability to create phenomenal visuals has reached a point where they don’t generate the same “wow factor” with audiences that they used to? If that’s the case, how do you help director’s find those new “wows” in their films?

JK: Visual effects is a storytelling tool. It exists to help filmmakers tell their stories. The most important thing is that they tell good stories that are engrossing and that have characters you can really relate to. There are some films that have managed to be successful and skate along with very little other than their visual appeal. Perhaps audiences are tiring of lazy storytelling, shows that rely mostly on visuals to support them. That's probably a good thing. Audiences have higher standards. They want good stories. I am all in favor of that. It's not so much that we have to keep figuring out ways to top the visuals from the last set of big pictures. You can look at pictures that have done extremely well, with visual effects that were well executed, but weren’t some kind of new visual effects revolution. They’ve been really successful and connected with audiences because of good characters and a well told story.

DS: Basically, the onus is on filmmakers to make better films.

JK: Yes. And that's a good thing.

DS: You spoke earlier about taking a more strategic and global view of ILM’s R&D. The last few years, much of the narrative within the technological side of visual effects has been virtual production, global production pipelines and pipeline efficiency. None sound very sexy, but they’re critically important for production efficiency and scalability, empowering artists to iterate more and wrangle with technology less. How does this impact your studio?

JK: Generally a lot of the development I’ve been pushing has been geared towards reaching the very simple goal of trying to have artists spend more of their time creating art. The tools we use are still immensely complicated. There is a lot of tech that has to be put in place, a lot of scene management, lots of file listers and all sorts of things that have to be worked through to make these kinds of [high-end visual effects] images. If you prepared a histogram of what the artists are doing during the day, a small portion of that would show them actually doing their art. A lot more of their time is spent on a lot of noncreative technical things. I would rather have artists spending more of their time doing art. So I’m in favor of anything that we can be doing to streamline the process, so there is less filtering through the error logs looking for what broke.

A decade or so ago, I always thought it was a bad sign when you could walk around the company, looking at what was up on people’s monitors and half or two-thirds of the TDs had no images on their screen. It was stuff like file listers and logs and script editors on their screens. It always seemed wrong to me for a company that produces imagery to have so many people without imagery on their screens.

DS: Last question. You have been involved in so many great films. I’ve heard you speak a number of times at conferences and I always got the sense you really enjoyed working on the projects you were presenting. Is there any one particular film, or particular aspect of the work you’ve done, that you can point to and say, “Yah, this was pretty cool. This brought me a great deal of personal satisfaction.”

JK: Yeah. Actually, on most projects I work on I ruin them for myself. There are always mixed emotions when you get onto a project. First of all, I am attracted to the projects. These are the kinds of films that I want to see, that I want to be involved with. But you know, as soon as you work on a project like that, you pick it apart down to the minutest detail. You see every shot a thousand times. You know all the arguments and compromises that went into the visual effects. So, you completely ruin the experience of watching that film yourself, because you can't look at it with any objectivity by the time you get to the end of the project. Some amount of time has to elapse before you can look back at a project you worked on with any kind of fresh eyes. When it’s still fresh, I look at it and all I can say is, “Yeah, you know, I should have gone darker with the background on that shot and boy, I always wish we had better atmospherics here…” You experience this giant flood of memories of all the things that you wish you’d have done a little different.

DS: All you see are the “mistakes.”

JK: Yeah, so it takes a while. I found for me it takes three to five years before I can go back and look at something I worked on with a little bit of perspective. Last year I was flipping around on cable and Pirates of the Caribbean II was on. I came in just as Davey Jones was making his big appearance and you know what, I feel that work really holds up. I look at that and I can't think of any way to make that any better than it was. So I am proud of how Davey in particular turned out. It's a very memorable character that was really very successful. I’m super pleased with how well that worked. That one has a special place in my heart.

--

Dan Sarto is Publisher and Editor-in-Chief of Animation World Network.

Phil Tippett wurde auf Episode VII angesprochen
Samstag, den 01. Februar 2014 um 13:55 Uhr von Aaron

Eine der bemerkenswertesten Entwicklungen seit der Lucasfilm-Übernahme ist wohl, dass zahlreiche bekannte Gesichter in den Krieg der Sterne zurückströmen: Egal ob nun Lawrence Kasdan und John Williams für die klassische Trilogie oder Doug Chiang und Iain McCaig für die Prequels, bei Lucasfilm legt man offenkundig Wert auf personelle Kontinuität.

Wie sich nun herausstellt, ist Lucasfilm auch an einen Altmeister herangetreten, der durch seinen Einsatz wesentlich zum visuellen Realismus der klassischen Trilogie beigetragen hat: In einem Interview mit dem Hollywood Reporter äußerte sich der frühere ILM-Mitarbeiter und zweifache Oscargewinner Phil Tippett über Episode VII.

"Ich habe mit Kathleen Kennedy darüber gesprochen, [an Episode VII mitzuarbeiten], aber zunächst steht für mich Jurassic World auf dem Programm. Danach werden wir sehen, wo wir stehen."

Colin Trevorrows Jurassic World soll, wie auch Episode VII, 2015 in die Kinos kommen. Produziert wird der Film u.a. von Frank Marshall, Steven Spielberg ist als Produktionsleiter mit dabei. Über seine erneute Zusammenarbeit mit ILM sagte Tippett:

"Ich bin schon lange raus aus dem Dinosauriergeschäft, aber dann hat mein Unternehmen zusammen mit einer Firma namens Efexio eine App erstellt, die es den Nutzern ermöglicht, selbst neue Geschöpfe zu erstellen. Das führte dann wiederum dazu, dass ich gebeten wurde, an Jurassic World mitzuarbeiten, wobei mein Studio eng mit ILM arbeiten wird. John Knoll hat mir diesen Vorschlag einer Zusammenarbeit gemacht und dabei gemeint, er wolle 'die Band wieder zusammenbringen'."

Tippett stieß erstmals 1975 zu ILM, um die Dejarik-Effekte für den ersten Kinofilm zu erstellen. Bei Das Imperium schlägt zurück leitete Tippett dann zusammen mit Jon Berg die Trickabteilung und entwickelte für die imperialen Kampfläufer und die Tauntauns die Go-Motion-Technik. 1981 wurde Tippett für seine Arbeit an Der Drachentöter für den Oscar nominiert, 1984 gewann er den Preis für die Effekte von Die Rückkehr der Jedi-Ritter. Im gleichen Jahr verließ Tippett ILM, um sein eigenes Trickstudio zu eröffnen, das sich verstärkt auf Modelleffekte konzentrierte und bereits mehrfach mit ILM zusammengearbeitet hat, darunter u.a. an Jurassic Park, für den es einen weiteren Oscar gab.

Inzwischen setzt zwar auch Tippett Studio auf digitale Effekte, die Stärke der Firma bleiben aber ihre Modelle, animatronischen Tricks und vor allem ihre tierischen Kreationen.

tatsächlich wieder echte Raumschiffmodelle hier und da vorbeifliegen zu sehen,

Den Rancor finde ich viel bedrohlicher und "greifbarer" als sämtliche Prequel-Viecher, genauso wie die Dinos aus JP durch das perfekte Zusammespiel zwischen Animatronik und CGI.

Modelle sind genauso wenig "echte" Raumschiffe wie Computeranimationen.

Das meinst du nicht ernst, oder? Der Rancor besser als die Viecher aus der PT? Dieses hakelig animierte Latexpüppchen? Da malt die Erinnerung mit güldenen Farben, oder?

Sorry, aber wenn ich dieses Argument noch einmal irgendwo lesen muss, stranguliere ich einen Hutten. Darum geht es in dem Vergleich doch gar nicht. Natürlich ist weder das eine noch das andere echt, deswegen habe ich "echt" auch in Gänsefüßchen gesetzt. Es geht um die Illusion, und die fand ich bei den Modellen der alten Trilogie nun mal besser, objektiv betrachtet und ganz ohne Nostalgie-Brille.

Ist nur meine subjektive Meinung. Ich finde gerade bei der PT sind die Effekte nicht sehr gut gealtert, es sieht alles sehr nach Videospiel aus, und sowas trübt bei mir persönlich immer das Sehvergnügen. Und man hat halt immer im Hinterkopf, diese Wesen/Raumschiffe existieren nur im Computer und wurden nie materialisiert. Ich finde ja die Szene mit Dexter Jettster in Pepsidose II z.B. ziemlich grenzwertig. Man kann Ewan McGregor richtig ansehen, wie unwohl er sich fühlt und nicht so recht weiß, wo er hinschauen soll.
Ich sage ja gar nicht, dass bei Episode VII wieder genau die gleiche Technik wie bei Episode VI zum Einsatz kommen sollte. Jurassic Park hat aber meiner Meinung nach schon vor 20 Jahren gezeigt, wie es richtig gemacht wird.
Episode I ist übrigens mein Lieblingsprequel und man erkennt dort auch sehr gut, dass dort noch viel mit Modellen und echten Drehorten gearbeitet wurde.

Das war nicht als ernsthaftes Argument gemeint, das sollte nur klarmachen, wie absurd die ermüdende Forderung nach ach so tollen Modellen ist. Da muss man auch keinen Weltraum-Wurm foltern. Sorry, aber die "Illusion" finde ich bei CGI einfach gelungener. Ich kann mich noch gut daran erinnern, wie ich 1997 in der Special Edition saß und auf der großen Leinwand den Wechsel zwischen Modell- und CGI-Aufnahmen beobachten konnte. Schon damals fand ich die Aufnahmen mit den per Travelling Mattes einkopierten Modelle mit ihren komischen Triebwerkslichtern nicht so gelungen wie die Computeranimationen der X-Wing, bei denen man sogar den Kopf des Piloten im Cockpit und die Reflexion von Lichtern in der Pilotenkanzel erkennen konnte. Mal davon abgesehen, dass das Geschwader nicht mehr wie an einer Schnur gezogen künstlich hintereinander herflog, sondern "realistisch" in loser Formation angerauscht kam.

Gerade Episode I legte bei den Weltraumszenen mit den reflektierenden Naboo-Fightern und den detaillierten Explosionen mit Trümmerteilen und indirektem Licht noch mal ordentlich einen drauf. Von dem Podrennen, das in einer "echten" Umgebung stattfand und nicht nur vor schwarzem Hintergrund mal ganz abgesehen.

Einige Effekte sind in der Tat nicht gut gealtert. Die waren aber genauso der Zeit geschuldet wie durchschaubare Modelleffekte irgendwann in den 70ern. Der englische Rasen auf den Naboo-Ebenen zum Beispiel sah nicht hundertprozentig überzeugend aus. Wenn man aber mal den Maßstab des Projekts bedenkt und dass das alles gerade mal sechs Jahre nach der "digitalen Revolution" gemeistert wurde, ist das alles schon erstaunlich.

Dexter Jettster fand ich übrigens großartig gemacht. Genauso wie den digitalen Yoda (den ich übrigens deutlich besser fand als Gollum, der zur gleichen Zeit entstand). Dass Ewan McGregor sich unwohl fand, ist sowieso wieder so ein unhaltbares und reininterpretiertes Argument von Leuten, die unbedingt einen Nachteil von CGI suchen. Gerade Ewan McGregor hatte in der PT eine mehr als überzeugende Darbietung abgeliefert. Und zumindest ich habe ihm abgenommen, dass er mit Dexter gut befreundet ist.

In Episode II wurde übrigens auch mit echten Drehorten gearbeitet. Episode III entstand zwar komplett im Studio, aber das trifft auf heute hochgelobte 80er Jahre Fantasy-Streifen wie Legende auch zu.