Interview With Jon LaBrie Of Weta Digital

Author: by Daniel Sánchez-Crespo Dalmau

Let's start with who you are - tell me about your background.

I'm the former chief technical officer at Weta Digital. I started in that role in late 1995 and held it until 2002. In that role I was the principle architect of the hardware and software infrastructure used to accomplish the effects used in The Lord of the Rings film trilogy. My official title now in respect to Weta Digital is Digital Systems Architect, although that is now largely an honorary title and I have moved on to found a wireless gaming company called Blister in New Zealand.

Tell me about your team and the tools you used to make The Lord of the Rings.

When I arrived in 1996 at Weta Digital, there were seven people. At our peak, when we delivered The Two Towers, we had grown to 350 staff. Almost 290 of them were digital artists working directly on shots, and the digital infrastructure had grown to almost 24 terabytes of storage, the largest gigabit ethernet and campus network in New Zealand and almost 500 workstations. There were a number of artists who worked with two workstations at their desks. The primary software tool we used for modeling and animation was Maya, and the primary hero-rendering tool for the creature pipeline was Renderman -- although tools like Mirai were used for significant tasks like building the Gollum subdivision service space and the control system. Houdini was used for some particle effects in the first and second films. The primary compositing tool was Nothing Too Real / Apple's Shake system, and various other tools were used around the office to fill in as necessary.

What do you consider your main contribution to how special effects are created for movies?

That's a difficult question to answer. Certainly when somebody asks you what your favorite effect is, or asks you about your contributions, you tend to think about the technology that you put in front of the viewers. In any facility, you start to develop the technology along the lines of the current production. For instance, with The Lord of the Rings, we knew we had specific technical challenges. We knew we had to have techniques to do things like achieve scale with small hobbits, so we developed both practical and digital techniques to do that. We also knew were going to do scenes that involved large crowds, so we had to do development with massive crowd animation systems. There were lots of digital creatures who were fighting, screaming, dying and running all over the place, so we had to develop pipelines for that. But I think we're proudest that the effects in The Lord of the Rings served the story. There's not a lot of "Look-at-me!" special effects.

We're fortunate to have been involved in a production that requires a lot of effects to tell the story because of the graphical nature of the work. I think the reason that the effects work resonates with the people is because it's part of the story - the effects are integrated entirely into the story and help to further tell the tale. So the thing we're most proud of is the integration of all the technology into the story.

What is your take on the relationship between film and games? What can the game industry learn from the movie industry, and vice versa?

To my mind, they have been fundamentally different industries. I think that the game industry has done a good job of leveraging some of the things that work with linear storytelling to create a more immersive experience.

But games are limited in using stories to drive gameplay, because telling a linear story in a movie is mostly about a director's point of view. It's about a director saying, "I want to tell the story in this fashion, I want you to look at these things in this order, and by doing this I'm going to create meaning and show you the way I view the world." With games, you're limited in your ability to do that because when you put a controller in the hands of the player, meaning becomes blurred. So storylines in games are often offered either as context for the gameplay or they're offered as reward for the gameplay or a combination thereof; you get a cutscene. You learn more about what's going with the story because you completed a level.

But I think as the artificial intelligence of non-player-characters increases, and the verisimilitude of the gaming engine improves, our ability to provide a visual immersive experience will improve. We in the game industry will turn more toward improving the emotional and the NPC engagement aspect of gameplay. We'll begin to resonate more with the interchanges between NPCs and gamers. Grand Theft Auto and Grand Theft Auto: Vice City are good early indicators of this shift, as is Deus Ex. If you're clever with the way you design your levels, you can have immersive gameplay, and you can have a player seize control of the environment and do things that couldn't possibly have been predicted by (or scripted by) the game designers. I think that as computing power improves, as the game development tools improve, you will see more of that kind of thing.

What the film industry has contributed to games is an understanding of the way to tell a story visually: knowing where to put the camera, when to use a close-up effectively, why to use cameras placed over the player's shoulder. Look at Gran Turismo for the Playstation 2. The playback of the races are incredibly impressive -- those camera points on the raceway are somewhat chosen the same way as people in the real world choose to put cameras on real tracks. Because of that, they're speaking in a visual language that we intuitively know and understand. The movie industry has been demonstrating how to tell a story visually for years, and there's a lot of skill there we have leveraged in games. I'm not surprised that camera technology and understanding where and how to place a camera continue to concern game players. That's how you keep players involved.

One of the main things that struck the audience in The Lord of the Rings was the realism of the creatures, especially Gollum. What is your opinion of the state of the game industry concerning characters compared to the way characters are done in movies?

I think there are some standouts in terms of character design. One of the luxuries in working on The Lord of the Rings is that we had significant budgets to devote to character design, and because of that we could iterate those creatures a number of times. There's no reason you can't have something like Weta Digital workshop's Richard Taylor working on character designs for a game. As a matter of fact, if you look at The Lord of the Rings game, those creatures draw directly on the work that was done for the film, and the quality of those creatures was directly formed by what we do.

As far as I know, the triangle count used by Massive, and the budgets we use for games are in the same ballpark, so really there's no difference from a technical standpoint, right?

Yes, absolutely. Massive has between 1,000 and 3,000 polygons and when those polygons are used intelligently, you can get fairly convincing creatures until they get very close to the camera. So there's no reason to think that with the current and next generation consoles, you can't achieve a similar level of visual detail to maintain the illusion of reality. That's the thing I found most impressive about the GDC: the advances that have been made by the chip manufacturers. Look at what's happening with Nvidia and ATI -- they're showing very, very impressive visual content on boards that you can drop into your PC. I think it's extraordinary. I think as soon as we have artificial intelligence, physics engines, and the other bits matching the level of detail that we can achieve visually, and we start to make those tools more readily available to the casual user so that you don't have to be a rocket scientist to manipulate and build environments or create characters, we will start to see some extraordinary output. We will see a lot of crap too, but we'll see some really extraordinary work.

Do you think there's a need for some kind of uniform toolset that can be used by both the movie and game industry? For example, today it looks like Renderman is the standard for film shading effects, but there are at least three proposed shading language standards for games. So what do you think of this?

Well, it's a problem. I would personally say that Renderman is a standard, but it's not the only standard. There are others, like mental ray. Even Renderman doesn't work in certain situations. Being new to the game industry, some people say I don't know a lot of the standards that are out there, but I understand that there's factionalization both in terms of the developing core and the SDKs that you're using.

I think the availability of middleware is good. You can put the game development tools in the hands of a larger number of game designers who have an idea rattling around in their heads but can't get a team together to actually deliver a gameplay idea. The next thing you know, the coolest game engine is available for ten bucks to the guys that are doing fresh garage-development kind of stuff.

As for standards, yeah they're a problem. But I think the market will decide how it's actually going to fall out. It's difficult to try and force standards on people. It's difficult to force standards on companies, especially in an industry like the game industry that is technological to its core. A lot of very talented game programmers are saying "We have an interesting way to solve this problem…" and they spend 18 months and a couple million dollars to solve it. When that happens, it's pretty hard to pry that technology away from them. Actually, what they should be thinking is, "We only get 80% of the game's profits, but get the game out to a lot more people." It's a tough battle. I don't envy the people that are caught up in it.

My final question: I heard that you are entering the game industry. Could you elaborate on that? What are we going to see from Jon LaBrie in the near future?

I left Weta Digital in the middle of 2002 to start a game development company, and I remember talking to [Lord Of The Rings director] Peter Jackson about that and saying, "Peter, I'm bailing out to start my own company." At first he was disappointed, but later he was actually very encouraging. He said, "Life is not a rehearsal. You want to get out and check it out on you own, now's the time and if you're interested in console games, then I'm interested in talking to you about that." I said, "Actually, what I'm really interested in is wireless games. I like the idea of ubiquitous handheld mobile devices and ubiquitous connectivity. Jackson replied, "Well, I'm a console guy and I don't really see that happening, so good luck to you." And I took off.

Then I began a six month investigation into the state of the industry, and I came to a couple of conclusions. First, you see a lot of people try to cram games that we played 15 years ago into mobile entertainment devices, hoping to capture a facet of the game market, to get more cash and attract some gamers. I don't think that is ultimately a successful strategy. I think to capture gamers, to capture people who have these devices but aren't gamers, you have to offer them something they haven't seen before. That means leveraging the capabilities of these devices: ubiquitous, wireless connectivity. So my focus has been fundamentally on developing massively multiplayer games that are very simple, very fast to play. My ultimate experience is a massively multiplayer game that you can get in and out of in two minutes. My thinking of late has been about gameplay structures that are optimized for time in and time out. We have a couple of them in development in New Zealand and we're looking to bring those games to the world in six to nine months. So be watching for games from Blister that are MMP and very, very fast.

And the website for Blister is…?

www.blister.co.nz and we'll probably be publishing in the United States under Interval Entertainment at www.gameinterval.com.

______________________________________________________