The Art Of: Just Cause 2

Though far enough apart to offer an important sense of travel, Just Cause 2’s ecosystems are also close enough for some gorgeous transitions of earth and sky, both taken in the stride of the studio’s renderer, Avalanche Engine 2.0

Postcard-perfect ground-to-air-and-back-again action staged across three climates, 32 square kilometres and 24 hours of simulated day and night: Just Cause 2 is a blockbuster all right. Art director Stefan Ljungqvist recalls some of the highs and lows of creating it, such as finessing grenade-flung bodies.

Images courtesy of Dead End Thrills

Is a despotic island nation the perfect place for an action sandbox?
We chose a banana republic for the first game – it lent itself very well to what we wanted to do. A dictatorship seemed like fundamental decor in those terms. And the lawlessness… A vast part of the western world doesn’t live in a dictatorship, so it feels kind of exotic. For the second game, I wanted to have a much greater variety in the game world. In the first [game], we pretty much had one climate zone and tricked you into thinking it had more variation than it actually had. So that was a huge thing for the second game. We all love Southeast Asia, and felt it was a great place to start. I’ve been boat hiking in Thailand and Vietnam, so I had loads of reference material.

So it’s important to treat a place as foreign, then? To play up aspects that seem exotic?
We’re not making Grand Theft Auto, right, so we’re trying to superimpose some of the elements that make things interesting. It becomes a caricature – and it’s all with intent. Compare it to Saints Row where with the latest instalment they try to push through even that boundary, maybe over-pushing it. It’s a game. We want to break away from reality and ordinary life. It’s distilled from several countries. We’re not trying to say we’re portraying Thailand, for instance. There’s all kinds of stereotypical ideas that people have about countries like that.

A lot of console games go overboard with their assets and end up violently compressed. Not this one.
There’s a lot of people on the art side at Avalanche who come from a technical background, who started in the demoscene. If you start there, you realise you can’t just say, ‘Oh well, it looks beautiful in Maya,’ because nobody will see that unless you put it on your portfolio. If you approach it from the angle of, ‘I’m an artist and I do great art assets,’ I don’t think that’s the right one. It’s about getting your hands dirty, right? We never gave up [on] that idea. We have to make it look really beautiful, and the only way you can do that – to understand how to do that – is to really get into the technical aspects as an artist.

And that’s also how you earn confidence from the programmers: prove that you can be part of that process. It’s also the other way around; we’re really fortunate at Avalanche to have programmers who have an artistic eye. That collaboration has been fantastic. The one thing we all share is that we’re never satisfied, we’ll try to push the game as far as we can. We have a saying: if the world needs another corridor shooter on Mars, [Avalanche] shouldn’t do that because then you can just use Unreal and – I’m not saying that’s bad, I love that stuff, but that’s not what we’re supposed to do.

And if you compare a game like Just Cause to Just Cause, there’s great deal of technical, graphical and game design innovation. We made Just Cause actually work on PlayStation 2, which was ‘impossible’ to do. The collaborative part is essential, and as a game artist you must realise that you have to look at your assets on the actual platform. The Xbox 360 and PS3 versions [of Just Cause 2] looked quite different for a long time, and we weren’t satisfied with that. We really pushed to understand how we could make them look good on both but then also use the same baseline assets, and then use the same approach for PC which is a totally different platform. And you try to accommodate the publisher’s needs, too.

How satisfied are you with the multiplatform landscape: the expectations, challenges, and economies?
The bottom line is that the publisher has a plan for how to commercialise the product, and at the end of the day we are making products. For some of the developments we’re fortunate enough to be working with publishers who really believe in what we do and support it. Usually our titles start out with all of the available platforms. We’ve had scenarios in the past – Just Cause springs to mind – where they tagged on additional platforms right at the end. We’re primarily a console developer, even though we’ve made efforts on the PC scene from a studio perspective – there’s a lot of love for the PC. There’s a lot of interest in PC games now that maybe, two to four years ago, was in decline. But I think most of our partners are looking at it still from a console-first perspective. Our tech is truly independent of that, but we’ve come so far now that we should primarily focus on making console games. That said, we’re never trying to make slap-on ports.

From a personal point of view, that’s mostly about the controls. In the kinds of games we’re making it can become a little bit difficult to make a completely justified set of controls in the timeframe we’re given to complete that part of development: the PC part. But maybe that will change in the future. It has been like that, though: it hasn’t been a significant part of the development contract to make the PC version. That’s not the significant part of the work that we’ve been hired to do.

Square Enix is seen as something of a champion of PC at the moment, much as Eidos were before…
I think we’ve been fortunate enough to have good partners. When we start talking about the game we’re about to develop, we discussed how much effort we are going to bring to that specific platform. It’s an important thing for the studio and it’ll continue to be an important thing. We have a few extremely talented programmers who believe really strongly about the future of PC and PC graphics. Naturally, because the hardware is there first, they can try out and flex their muscles on that. We all try to push, and the easiest platform to do that quickly on is still the PC. It’s a good showcase scenario.

That’s how The Hunter started, right? It started as a prototype for some of our new landscape technology, and eventually we were really interested in figuring out how we could apply that to learn more about where the market would go, how people actually engaged in F2P or subscription-based games. That’s not really Avalanche, that’s not our core. We’re console developers who make action games, so that was a huge step for the studio. But the knowledge that we still get from that affiliated company is great; there’s a lot of valuable information we can use and seep into the bigger games we’re making.

The animation is very sophisticated.
I love animation. It was a huge battle internally between the programmers, game designers, animators and myself. One of the things with traditional animations in games is that it requires a lot of memory. The ideas we had didn’t really support the traditional method of moving forward with just keyframe or motion-captured animation. We wanted to bring in more physicalised animation, if you will. That’s also expensive but it doesn’t [expand] the memory footprint, and that’s the huge thing in console games. So we came up with this solution called constraint-based animation, a kind of controlled physics.

When you throw a grenade in Just Cause 2, the way people fly through the air is very orchestrated. It’s not like we apply a random ragdoll physical impulse and see what happens. I’m talking about weeks, maybe months, of figuring out the coolest way for portraying both how a solider would fly through the air, and what’s most interesting to watch. Another example is when you’re hanging beneath a helicopter in the game and firing. Rico [Rodriguez, the protagonist] feels very alive, and that’s because there are five different systems animating him.

Continue >>