• "Grand Theft Auto 4"
    CGSociety Production Focus, 29 April 2008

    Digging into the Vault, I thought the Rockstar advice here was still worth taking for a spin. Grand Theft Auto series art director Aaron Garbut talks about GTA IV's development.

    When Rockstar makes a 'Grand Theft Auto' game, what is the very first thing that happens, and how significant are those early decisions in terms of the game's overall artistic direction?

    Firstly we just start collecting ideas. Locations, technology, gameplay, missions, basically everything and anything we want to include or do. We think about what we are trying to achieve in basic terms, we spend a bit of time sorting and discarding some of the initial ideas. Then we just start...

    In terms of the art department, the character artists will start playing about with concepts, trying various main characters in the game, playing about with pushing the style a little, basically experimenting. The vehicle department will begin a first pass of every vehicle, the environment artists will lay out a road network and once we've all driven around on it for a bit will block in each city block roughly so we can start to see the skyline. The intention is that as soon as possible we have a very, very rough version of the game and then we begin to refine it. Like every other aspect of the game the artistic direction grows organically, we try stuff and things that work pull us in their direction and things that don't are changed.

    At what stage do the characters take on form, and do they change significantly over the course of development? What impacts that?

    Right at the start the ambient characters are blocked in, there is a first pass that gives us an initial version of them and as the style tightens and the artists become more confident with the tools we go over them again and again and tighten them up, add varieties and consistency. The main characters all wait for the script. We work from some initial biographies while the script is still in its early stages and then again push these further when the script evolves.

    Did the exponential increase in hardware power on PS3 and 360 relative to PS2 have any influence on the kind of New York reconnaissance research you had to perform?

    Not really. We always aim to get as much reference as possible regardless of the platform. It's always going to be reusable in some way and building up a library of this stuff is really useful. The first reference trip happens fairly early in the project, not long after we have the initial block in. At this stage in the project we were still fairly unsure of what the power of each system was. When we did the second reference trip though it was much more focused. We had a fairly evolved game by this point and were able to get exactly what we needed to help us in areas we were struggling with.

    Character interaction, but in particular simple gestures and mannerisms - the way people move while they speak, for instance - have always seemed like the GTA cut-scenes' best weapons in the fight to define characters visually in spite of system limitations. But now that you have the ability to pose characters convincingly (like Niko leaning on Roman's kitchen table in the opening cut-scene) and do things like eye contact convincingly, did you find yourselves having to think differently about how you approached character design and animation?

    I think the key difference in approach this time was that we just didn't take the easy way out, not that it was ever easy before. We just shot the action we wanted and then dealt with it when it arrived. Having any sort of interaction like this always adds to your problems, and now that you can see those problems in HD we couldn't really hide anything. On this project though, there was a decision to make as few compromises in all aspects as possible. Just to try stuff we might have shied away from in the past and see if we could manage.

    I think the level of detail the cut-scene guys have achieved is intense. They have the characters interacting with the world, which is hard enough, but they also have added a lot of feedback onto the world. Some of it is so subtle you sometimes don't even notice (though you would if they hadn't done it). Things like pillows and mattresses bending or bouncing a little under the weight of characters sitting on it, phone cords dangling and following the phones movements, liquid moving around in glasses. It's pretty amazing.

    There's so much of this stuff, all the characters just feel part of the world, they lean on or against things, interact with each other, push things about, knock things over. I think we have more of this stuff in one scene than we had in the entirety of previous GTAs. To get to your question though, I think it was more of a mind-shift than anything else; we just decided to go for it.

    I've read about how you sought strong brow lines and a face that could convey the emotions you wanted for Niko. Were there any other directions you could have taken, or was Niko the exact fit for your goals?

    There were a few other versions of Niko that went down slightly different routes, but he evolved into more or less his final look pretty early in the day. I'm sure he would have worked in a number of ways, but when we got to what was essentially the version you've seen, he just seemed right. He had a good feel about him, he looked like he had a history, and he looked different.

    In creating Niko, what was it about his face that allowed you to achieve what you wanted? And how do you even begin to search for that kind of thing? And, moreover, what comes first - the emotional core of the character or the framework of the journey you will be laying out for them to take?

    We don't do anything as pre-planned as designing the character around his journey through the story. The story always comes a little later in the day. At the point when the characters are developed we tend to be working more from feel and a short biography. We know the kind of character the player is supposed to be and we know a little of his background but we still don't know what he is going to do and say. So in that sense we have to work with broad strokes. Niko's face looks like he's had a past, that he has done and seen things most people haven't and that this has affected him. That's a good basis to build on.
  • Living next to the sea myself, my very first impression of GTA IV was that you had captured the character of water under sun, metal and glass - the range of colours and their interaction with dynamic surroundings. Can you tell us a little bit about that process?

    A large part of this is down to programmers with a great eye. Most of what is going on with the water is very clever and creative code. The only influence we have as artists is obviously creating the world that's being reflected and then setting the lighting up to get this looking as nice as possible.

    There are some lovely things going on in the water FX. From the way it reflects the surroundings and distorts that reflection, to the real-time physics on the surface (drop a car into the water and actual waves will distort the surface, which will affect nearby boats). There's even foam wherever the water gets shallower and simulations of viscosity around the edge. It's an insane level of detail, but you can see this when the water laps up and down a wooden post or jetties. All that is done in the FX code.

    On top of that, we have global control for each time of day and each weather type to alter the way it fades with depth, which can help to make it look really murky. We can tint it subtly, but mainly it's the time of days, sunlight and sky colouring that has the biggest effect. But as artists we're just tweaking what is already at its base level a bloody nice effect.

    Are there any other effects in the game of which you're particularly proud, or which threw up particular design challenges that you feel others have yet to overcome?

    I think the lighting system in general is pretty amazing. There are no hard limitations on the number of active dynamic lights around the player. The real-time shadows are working across every object and surface in the game with everything self-shadowing and casting onto everything else, there's ambient occlusion and emissive lighting on top of that. And then your standard next-gen shenanigans - light shafts, bloom, depth of field and motion blur, and of course it goes without saying everything's rendered with HDR.

    The net result is a fully dynamic, real-time lighting system that is consistent across every surface in the game and has the subtlety and solidity of prebaked lighting. We've always had to make compromises in GTA's lighting because we had dynamic time of day. You make a trade-off with this essentially between variety and quality. There's a lot of stuff you simply can't do because the lighting needs to gradually fade between hours and weather types. With the system we have now though, there really aren't the same trade-offs. We get amazing almost prebaked quality combined with a constantly changing world. Where you can stand at the same spot and the combination of weather and moving time will mean you will probably never see it look the same twice.

    Given your use, in some cases, of real world locations, do you find that you have to contrive or excise details to skew the game in a fictional direction, or does the process of creating locations from visual material and memory naturally imbue them with their own distinctive aesthetic quality?

    We never reproduce real world locations. We take interesting or representative elements and create something new from them. It's about taking inspiration from real places and producing something that captures the essence of it. We're trying to take our impression of New York and keep it as that, an impression, not a laboured reproduction. I think that gives it more flavour, more intensity and in an odd way makes it feel more real. I've seen it in other games that set out to rebuild a city street by street, not only do compromises get made that favour realism over fun but a lot of the life is lost and all that's left is a hollow representation of a real place. I'd rather have the right vibe than an accurate roadmap.

    I've read before that you weren't seeking realism visually, but you must have discovered that the extra graphical fidelity afforded you the ability to give characters and locations more depth. Can you talk at all about how you filled that technological headroom?

    There were some similarities between GTA III and IV in the initial approach. It's always a lot harder to get your head around the possibilities of a completely new system. With Vice City and San Andreas we had a pretty good idea of what the PS2 was capable of. On each the engine was enhanced or we came up with a new way of doing something and that bought us a little more power, but generally from day one we knew what to expect and we had an engine to try things out in.

    On GTA III and IV though there was a lot of guesswork involved, we had to make things up as we went along. A working engine doesn't appear till a fair ways down the line and even when it does arrive there's still plenty of guesswork since it will not be optimised till nearer the end. So we just make one guess after another and hope we're guessing right. It's always a compromise between memory, detail, lighting, AI, physics, streaming, numbers of characters and vehicles, missions, etc. There are no hard and fast rules since we can bias what's important depending on the area - one area may be lighting heavy, another physics, another memory and rendering intensive. To work successfully with these sort of unknowns, and this number of variables we try to work iteratively as much as we can.

    It's only towards the end once we have an engine that's close to final that we can start to tweak with that in mind. And it's at those stages where we find ourselves getting a real sense of what is possible. Where we have to take things out and make compromises and where we end up adding detail. I think we have already pushed both consoles very hard but I'm really excited to see where we can take it next now we know what works and what doesn't, now we know how to play to the strengths of our new engine and the consoles themselves.

    It seemed to me from playing the game that individual neighbourhoods and streets didn't just have their own character, but that you were using them to direct an emotional reaction. Is that scale of mood something that consciously informs your location design, or is it a natural characteristic of New York that you're simply able to harness to fit the existing parameters of mission scenarios?

    The cities are never built specifically with missions in mind. We always build the cities first and fit the missions and stories into them. There are a few reasons for that. One of the main ones is practical and it's more pronounced on a new engine. The basic rendering parts of an engine tend to come online a lot sooner. The mission designers need a scripting language, fairly evolved physics and vehicle handling, the weapon systems, AI etc before there is much they can play with. Whereas the artists have 3D software from day one and the game can start rendering that quickly so we can get on with building the city right from the start.

    So we've always treated the cities like a real place. We build them, we pack them with interesting things and then we place the missions within them at a later date. Obviously once a mission is placed and working we will tweak the area to work better, but essentially the processes are fairly separate. That's not to say there isn't a deliberate intention to evoke emotional reaction as you say. It's just that if there is one it's happening during the placement and pacing of the missions. I think having this massive environment available first gives a lot of opportunity to play with the missions and find what works best.

    There are essentially two routes you can go down in making a game: you can do a load of pre-production upfront and plan it all out in advance or you can just dive in and be a bit more organic. The first option is the safest, it lets everyone know where they are from day one, it lets everyone know what needs done and it's the easiest to organise. But I think it tends to lead to fairly lifeless, soulless games, particularly when the games are more open like ours. We are a lot more organic, this is a conscious choice and it does lead to more difficulties along the way, it's harder on the team and it's trickier to keep track of but it leads to better games. It works because the core team know each other well and have worked together for a long time, we trust each other and know what to expect. Our whole ethos is to try things out, play with them, find what works best and move in that direction.

    The entire game in some respects starts blurry and just slowly comes in to focus over the project. Some areas work better than others and the worst areas are always looked at and pushed forward. Missions start as experiments and are moved around the map until they work. The story drives some of this and weaves its way around the rest.

    No one aspect of the game is the driving factor, we don't create a list of missions, build levels around it and stick a story on top, and we don't create a story and hang everything off of it. Instead we have a bunch of ideas, elements of the story, the characters, locations, the general tone, gameplay elements, technology, mission ideas, and we just mix it all up and see where it goes trying to steer it along the way. It's all a big scary tangled web. But it works.

    Previous Page next
  • I'm very interested in the process of creating the game's vehicles. It's easy to imagine a number of scenarios, but most people I've spoken to about it seem to assume one of two things: that you adopt a real-world vehicle as your model and subtract and distort until you have a GTA car, or that you decide on a range of characteristics that fit the requirements of a particular area on the game's curve, and prototype something before sliding it in the direction of an existing archetype, like a Porsche or Lambo.

    It's a little of both actually. We make some basic decisions early on about the spread of vehicles. This is based on generally what we want from the game - the basic number of sports cars, the general percentage split of luxury cars, four-doors, four-doors, vans, lorries etc. Then the vehicle department start putting images to the percentages.

    We don't specifically go for one particular car but take elements of many cars and put them together into something new. I think if you know a lot about cars you could break the vehicles in the game down into the real world inspirations. But you would find many for each vehicle and you'd find a lot of our own ideas in there too.

    I think if our guys actually designed real cars I'd be driving one. Some of the things they have come up with are beautiful.

    As an entertainment series, with humorous and fantastical elements, do you have clearly defined structural, narrative and interactive parameters - a kind of GTA World Bible - or do you employ more of an organic approach informed by your experience of what worked in past titles?

    Every part of our process is pretty organic as I was saying earlier. We know what we are doing, we've been doing it for a while. We split up the responsibilities for the various elements between departments and have a fair amount of crossover. With a project of this scale it's increasingly difficult to have an overview of every element so we just have to have people take control of certain aspects and drive this.

    A lot of this works because of the iterative process we use. Since things are "blurrier" to begin with - less well-defined - we are able to tighten and cross-reference more and more as elements come into focus. So we end up with something cohesive and working as a whole because the smaller elements that really pull things together happen later once the bigger picture is more defined. The more time we have, the more interconnected and defined the experience becomes as layers of complexity are added.

    How do you go about creating the fictional brands and adverts that you use in the game?

    These things come from all over the place. We have a design department that heads this stuff up, organises it, dishes it out to the artists, and works with everyone and anyone to produce it. A lot of it they come up with themselves, a lot comes from the radio ads and shows, every so often an email will be sent out looking for ideas for businesses or slogans which always ends up degenerating into pictures of cats and diarrhoea for some reason. But ideas come from all those places.

    One of GTA IV's most written-about additions is the Euphoria physics, and we've seen plenty of examples of how that can be used to capture things like drunkenness, and compliment the game's freeform structure and emergent gameplay with humour. Was including Euphoria a hard decision to take? Did you perhaps feel you were - and was there perhaps any resistance to the idea of - ceding artistic control of certain aspects of physical behaviour to procedural animation?

    No, the decision to use Euphoria was pretty straightforward. I don't think we ever felt we had to compromise to achieve what we wanted. It is more a layer that happens on top of other animation whenever we decide we want it to. Basically we are always in control of it. In that respect we aren't losing any artistic control at all and what we gain is huge. We gain another level of interaction with the world and that adds a lot to its believability and solidity.

    I think the basic fact is it's almost impossible to create predefined animation that deals with physical interaction with a world in as many possible varieties as we would need. It makes much more sense to let physics do what it does and just make things happen correctly. At its simplest that would have been adding a ragdoll, but Euphoria is so much more than this even at its most basic level.

    Finally, do you have a favourite character in GTA IV, and what is it about he or she that you particularly adore?

    For me it has to be Niko. He's just a breath of fresh air in what can be such a dull, cliché-ridden industry. He's got real depth and soul, and just seems so unique. He's likeable and he's got a pretty dark history. He has done some pretty bad things in his past, but he still feels like a good guy.

    Tom Bramwell is Editor of eurogamer.net.
    Grand Theft Auto IV is due out - on PS3 and 360 - 29th April.

    Related links:
    Rockstar Games

    Discuss this article on CGTalk
    Previous Page more

blog comments powered by Disqus