Adolph Lusinsky interview: Wreck-It Ralph, software and Disney

Interview Simon Brew 8 Feb 2013 - 08:25

Wreck-It Ralph's director of look and lighting explains how a new system was needed to get something so old school looking so good...

Adolph Lusinsky, Wreck-It Ralph's director of look and lighting, actually came from a videogame background, working for Disney Interactive. "There were very few pixels on a really bad TV screen", he laughed, remembering his days developing for older Sega machines. From there, he went off to work for a visual effects house, before that got bought up by Disney, bringing him back to the studio.

Which seemed like a good place to start....

Was your background crucial here, in that you got a schooling first in as much of what technology couldn't do, as well as what it could?

Yes. It's funny, because I came from a painting background, and the way the Sega games used to work, you'd be on your PC, you'd do the artwork, and then you'd have a little dongle that popped it over to the TV screen. And it looked really blurry and mushy, and you move a pixel, and it looks like a different character almost. It was a great experience though. When we were developing some of the earlier look of the 8-bit version of Ralph, it had too many pixels. It was too complicated. Just having some of that background could really help us.

Appreciating that we're at home watching material in 1080p resolution, everything has to stand up to such insane levels of visual scrutiny now. In the old days, you had 480 lines to work with. Moving one pixel around on there would make a dramatic difference? When you scale everything up further, is it the same level of impact for you?

If you scale something up to a 4K projection system in a modern cinema, you're going to have to move a lot more pixels just to put across the impression of moving one?

Yeah.

So where do you pitch it? You talk about taking pixels out of it, but how do you balance that when you consider how the film is going to be projected?

You know, to get that real authentic feel, it's always seducing when you have more to work with than you're supposed to work with. We really did try to limit ourselves. Those old 80s videogames, at the maximum, are just 320 pixels tall, often more like 255 pixels. Which means those characters are like, maybe maximum, 80 pixels tall. A whole character. We really did design them with that in mind. But what really makes it feel good on the big screen is the research we did in the way a CRT looks. It gives it a believability that's really satisfying.

That you had this artwork and projected it on a CRT screen, and when you see it in the movie, you really feel like you're zoomed in close onto a TV monitor. You see the little RGB pixels through the screen. We really tried to be as accurate as possible in how that worked. Everything is getting processed through a red green or blue pixel. Remember when you're a kid, and you get close to a TV screen, and you see the little RGB pixels? We tried to capture that. 

The best very animated work is generally about brilliantly chosen imperfections I think...

Yes, yes.

This seems to epitomise that quite a lot. Digging into the imperfections of the past and turning them into something so vast?

Yeah, yeah. It's funny though. At the end of the day, it's still the same design problems: value structure, colour structure. To get things and characters to read appropriately. That limitation provides a certain sense of humour to the image. When you see those little pixels, and the animation so limited, and yet it's still funny, it's really satisfying.

So how much work did you do with some of the third party characters you had to work with?

The relationship was usually okay, you can use our character, and there was a sheet with Pantone colours. Pac-Man needs to be this Pantone colour, for instance. A lot of what we did was trying to match that stuff. With Bowser, the tendency is you've got this character that's fully projected onto a theatrical screen, and your intuition would be to add more texture to him. But we really tried to be as authentic as possible to the design sheets we were provided with. So it looked like what these characters looked like in the game. We'd send shots to the various companies, so they could be satisfied. There's a little back and forth that happened, but it went pretty well. We really tried to be as respectful and authentic as we could.

Presumably you have to frontload that with a list of what we're exactly intending to do, which must in turn affect your planning process?

It's definitely a challenge, because things do drop in and out. We're not going to get this character one week, then hey, we're going to get it! It took a while to land on what characters would be in the movie, what arcade games would be in the arcade. 

You're not right at the start of the pipeline, and yet you've said that you needed a new shader system for this movie. If my chronology is right, the research work started about four years ago...?

It started about two and a half, three years ago.

And did you come in at that point?

Yes. It's one of those things where you get onto a movie, you get a script, and the director pitches what the movie is going to be. Maybe images that are indicative of the imagery they won't in the film. Then we start brainstorming about what the movie is going to be, and how we're going to achieve it. At that point, we decide what we need to do to make this movie. For this one, the shading system we had just wasn't sophisticated we had wasn't enough to get the results we wanted.

What kind of challenges was your software having to address?

It really is getting that sense of light. With food lighting, a lot of it is about these really broad big reflections. They use a lot of big area boxes in lighting food and things like that. That stuff is really important to make food look appealing. This new shader system has big area lights, and surface shaders that have real and broad reflections based off the size of those area lights. And they're physically accurate, so they look really believable. We really needed that to make the food look as appealing and believable as it is in real life. 

When most of us outside of Disney first heard of this project, there was a real glee at the idea. But that's a double-edged sword for you. There's clearly passion for the project, but this is a film that's failed to get through the Disney process two or three times before. Where did you find the film this time?

It's a good question, because when we started the movie, this was a couple of years ago, I put a look board together. Based off the script we had at time, it was here's Fix-It Felix's Niceland, here's a first person shooter that we didn't know the name of at the time. Here's a game that's going to be like a racing game. And it ran like that. And it's like, how do we make those things look as authentic and unique as possible.

We look at Fix-It Felix, an 8-bit game. So your intuition is to make it as simple as you can: flat shading models for that. There's an academic process you go through, which is great, because all our brains are going. Then you think we know what we're going to do. Then we start producing R&D shots, and half of it sticks, and half goes out the window. For instance, with Niceland, we started with a flat shaded world, but so much of the movie was taking place in that world, that it wasn't cinematic. It wasn't appealing enough. But on our lookboard, that sounded like the best contrast between the first person shooter. But when we got to the movie, it didn't work.

So you have this kind of process where you have an idea, and it's an instinctual 'this feels good' or 'it doesn't feel good'. And at the end of the day, as Disney, we have to put images on the screen that feel appealing. There's an academic logic side to it, but then the intuitive side of whether it was going to work or not.

You worked on Armageddon too, I understand. Was this one harder then?

Yes! I think it is. The visual effects, especially back when I was in it ten years ago, there was a little bit more focus to it. You knew what the problem was that you had to solve, and then you solved it. I think really there the problem was with the computing power: how are we going to do this? With an animated movie, you have access to the director all day long. It's a much more creative process than visual effects. We're not just going to put in a plate that's already been shot. We're going to make every pixel that you see on screen in this studio. Every single thing happens on the computers in this building.

Has it whetted your appetite to go off and make a movie of your own?

No. My background is painting. I really like working with value structures and colour structures and trying to tell stories. That's my passion. That's what I enjoy doing. So I feel fortunate that I get to do that.

Adolph Lusinsky, thank you very much.

Wreck-It Ralph is out in the UK on February 8th.

Follow our Twitter feed for faster news and bad jokes right here. And be our Facebook chum here.

Sponsored Links