Monday, May 26, 2014
Codemasters has a lot of racing on its plate right now, what with GRID Autosport's release just around the corner, a new, yet-to-be officially announced DiRT in the pipeline and no doubt another Formula One title planned for the latter part of the year.
It was perhaps a brave decision to keep development rolling on the GRID series having only just finished GRID 2 last year, but the racing studio took the leap anyway, capitalising on the momentum and excess of ideas left over after getting the second game out of the door.
Having been afforded a few hours with the game at a recent press event we also had the chance to talk to GRID Autosport's Producer Toby Evan-Jones about what could have been a risk, Codies' plans for PlayStation 4 and Xbox One, and the many trials and tribulations that come with creating a racing game.
Plus some stuff on DiRT 4.
GRID Autosport is made up of five different racing disciplines. Has that essentially been like developing five racing games?
I wouldn't say so. The main differentiator would be the way that the cars handle in-game, and our handling team have obviously been working on cars for donkeys of years now; they know them like the back of their hand. What we tend to do is give a lot of attention to each individual car in the game, and we spend months giving each car its own character and real-world personality, then polishing it up. They've [the handling team] spent the same amount of time ultimately working on the cars, but they've had more freedom rather than having to make a one-size fits all handling model, the five disciplines have meant that they can have the freedom to focus on how the individual cars handle.
For example, the drift cars, they compete in drift events, so there's no need to worry about getting them around a track in open-wheel or having them compete with touring cars or anything like that. It means that they can be really representative of real-world handling. Drift cars, for instance, often have completely welded rear differentials, which means the rear wheels turn at the same rate, so they can't grip drive for anything. They just break traction very, very easily. That's been factored in, because they're only ever going to be used in a drift event.
Similarly with touring cars and open-wheel, we've had that freedom to really play to the strengths of the vehicles. I wouldn't say that it's resulted in extra work; it's just really opened the doors and given the handling team the freedom to truly express the characteristics of the cars.
There's still presumably a lot of track days researching and actually driving the cars though, right?
Yeah, that's really important to us actually. I think that's one of the areas we actually lead in as a studio; just the passion and knowledge of the team and the first-hand experiences that the guys have. In every game that we do, we try to get as much time behind the wheels of the cars as possible. For example, about a month ago we were down at Brand's Hatch with Drift All-Stars doing one-to-one driver training, going out in a variety of front engine rear-wheel drive cars from a Nissan 350Z, Toyota Soarer to the BMW E30; just a real range of cars, so we can understand even just within the drift discipline, how real cars handle. Like a 350Z with a short wheel base would be very snappy, so you'd have to be very quick to counter-steer, whereas a Soarer or BMW E30 have much longer wheel bases, much smoother and more progressive in the drifts. We've really built that first-hand knowledge within the team.
On the flipside as well, the drivers come into the studio and work very closely with us. So James Dean (Drift All-Stars) and other drift drivers have been in reviewing all of the drift cars, sitting with the handling guys and tweaking them. On the touring car front we've had British Touring Car champion Matt Neal. We've been working quite closely with him at the studio a lot, looking at the handling. So both ways there's a really big exchange of information, communication and experience, and that's something we think is really important.
And it all means you can drift your car around a ring road now.
That's it, yeah! The car I personally own is rear wheel drive and rear engine too, so it's really snappy. I'd have to be quick to grab it back!
Was the decision to roll into the development of GRID Autosport straight out of GRID 2 to maintain momentum going in?
Exactly. It was a natural decision that sort of evolved. It was not something that was planned from the outset at all. Often we alternate between projects, so we'll make a GRID and then start to look at DiRT while having a skeleton team on other projects, as we kind of prepare. This time around it sort of evolved such that we put GRID 2 out and the development team had a lot of ideas as to how we could eke out a little more from the current generation of hardware to get more cars on the track and have other areas of optimisation, like texture quality and things like that.
The dev team actually wanted to continue straight on, and there was a lot of feedback from the community that there was content they wanted to see in addition to GRID 2. A lot of people wanted touring car content in there, they wanted us to bring back the cockpit camera - which we needed to take out [in GRID 2] to pull back processing power for other systems - but with the additional time the guys at the studio had ideas about how we could further optimise to bring it back in and do something special with it.
After GRID 2 we just kind of took a step back, looked at what we had in the pipeline, as well as ideas that the dev team had and what the community was asking for, and thought, “you know what, we could just roll straight on here with Autosport and do something we want to do as a studio that the community also want.” And we could achieve it all on the current hardware, so we just carried on and rolled straight into it, cracked on and carried on.
On PC we can't imagine reinstating the cockpit cam was too difficult, but on consoles, did you have to make concessions to get it back into the game for GRID Autosport?
Not really. Just time and effort, definitely. Where we are with the console generation on Xbox 360 and PS3; we're at the point where we're getting every inch of processing from GPU, CPU: we know the architecture like the back of our hands, so it's a lot of work to bring out additional features. For GRID 2 we wanted to push on track streaming technology that streams in textures in real-time, allowing us to do longer tracks and more detailed textures. Rather than holding the whole track in memory, you can just stream in what you can see, essentially.
For GRID 2 we had to claw back [processing power] elsewhere, so head cam was one of the things we pulled out to do that. With more time we've been able to make optimisations to other systems allowing us to pull that back in. Another big improvement we've made is freeing up enough memory now to have 16 cars on track as opposed to 12, which is what we had previously. That makes a big difference for say touring car events, where it's all about a tight-knit pack. It feels really hectic and exciting with 16 cars in there jostling.
Was it seen as something of a gamble to go next-gen with the studio's next racing project then?
I think it was more that we didn't want to wait. We're at the point where we're looking at next-gen now, and we have a team dedicated to taking the EGO Engine and making it work on next-gen and making it play to the strengths of that architecture. It's going to take time to get it to a stage where we're happy that what we put out on next-gen is something really special that plays to the strengths of those consoles.
What we didn't want to do – and there were discussions – was take Autosport and everything that we wanted to do and what the community wants, and make a quick port over to next-gen just for the sake of being early to market and trying to go for sales. That's not really how we work. The original GRID back in 2008 wasn't first out on Xbox 360 and PS3 [at launch], but when it hit, it really made an impact, because it was head and shoulders above the rest of the stuff available at the time. That's what we want to do on next-gen: plough the effort into really looking at what those consoles enable us to do that we couldn't do on a previous console, so when we put something fresh out, it's really heavy hitting. It was a natural choice from our point of view to do that.
Should we naturally assume therefore that Codemasters' first next-gen foray will be DiRT 4?
That's an interesting question, because we're looking at DiRT, we've got the franchise there, which I'd say is probably within the Racing Studio at Southam, GRID and DiRT are our two primary franchises that we've got close to heart. We're looking at DiRT at the moment, it's in progress, we've got some really good ideas. The GRID franchise too; we've got some really good ideas about what to do next with that. It's going to really be a case of looking at what the EGO Engine guys are doing on next-gen, what they're managing to achieve on those consoles and then thinking about which game would be best to employ that and which is a natural fit for that, so that we do something very special when we hit next-gen.
Do you think GRID will forever stay its own thing and DiRT will stay its own thing or could they one day come together?
That would be something that would be silly for us not to think about, because we've got some very strong IPs there. We'd probably say that DiRT is leading the way in terms of that off-road experience, whereas GRID is very much about that on-track, tarmac racing experience. But there's definitely two prongs of attack going on there, and of course we've got the F1 license, so it'd be silly not to think, how can we bring all of this together to do something really interesting that hasn't been done before in terms of crossover. I couldn't speak about that too much, but I'd say that next-gen is going to offer some really interesting opportunities for us as a studio.
When Codemasters makes the jump to next-gen, what's that going to mean in terms of handling?
We've got a lot of processing power available on next-gen and that could allow us to do a lot in all areas, and it leads to a bit of fisticuffs in the office to be honest. Because you've got all the artists and the graphics programmers saying they can make the game looking amazing, they scale the visuals up to this resolution, boost the geometry and all that stuff. Then you've got all the handling and physics guys going “forget about all that; we can take the physics systems even further! Massively increase the sample rate on the surface textures to show every undulation, bump and dip.” So they want to do things there, but then you have the audio team thinking about 3D dynamic audio and the resolution of what's going on there, so everyone's fighting for that additional processing power.
It's a really interesting balancing act of where we allow the teams to have (that processing power) and what we do with it. And that's what's going on at the moment.
Is that essentially what developing a racing game comes down to then: striking that perfect balance?
Definitely. And you see different racing games taking different approaches to what they spend the money on. You have things like your Forzas and Gran Turismos that are very much more a simulation versus our games and some racing games that take that simulation approach, opt to go for 60-frames per second, but they need to pay for that by reducing texture quality and maybe dropping shadows, certain other objects and graphical fidelity maybe drops a bit. Whereas we opt to go for the 30-frames a second route, but make the environments beautiful and put lots of detail into the cars and all the rest of that. Different studios take different trade-offs, and spend their money in different places in terms of the processing, but I think with GRID, we've really carved a nice slot for ourselves.
Do you see a next-gen racing title like Forza or DriveClub with such a high level of visual fidelity as a target to aspire to?
I wouldn't say we see it as a target. I think what those guys are doing, they're doing very well. Hats off to them, but we kind of take a different approach to our games where we tend to spend the money and processing power on pushing the appearance and visual quality, but it's a different style we go for. (Our style is) a very cinematic style and to pay for that we run at 30fps, but we think that's a fair trade-off for our games and where we're going in terms of that cinematic experience as opposed to being a raw simulation.
We're bored of the resolution debate, but is being 60fps at 1080p on next-gen a concern for you?
Again, it's one of the questions that goes into the massive melting pot of where we spend the processing power that's now available to us. Do we make it 60 frames and lose a huge chunk of processing power that we could otherwise put towards graphical fidelity, physics systems or other areas of the game? That argument, 30 vs. 60; its power we could put into shadows, textures, post-processing... It's just one huge melting pot where you've just got to portion out everything very carefully.
It all seems like very negligible differences anyway, right?
I'd say the differences are subtle. When you go down the 60-frames per second route it feels smoother to the trained eye, people who are very experienced with racers get a smoother feel out of it. If you go down the 30-frames route, it means more processing power to push that graphical fidelity that bit further with texture quality, shadow detail, post-processing effects; it gives us more of a cinematic quality rather than a flat look at 60. It's very subtle differences, but people who are really into their games can probably tell the difference, but it's often down to personal preference.
With GRID Autosport taking the series back to its roots, do you think the next DiRT will return to pure point-to-point rally racing?
I wouldn't want to speak for the DiRT guys, because I don't want to put my foot in it and get a kicking when I get back to the studio!
Aw, go on...
(Laughs) I'd say the fact that we chose to roll into GRID Autosport following GRID 2 gave them more time and more space, because there's always a skeleton team working in the background with what we're going to do next. They've had even more time to look at what the essence of a DiRT game is, looking at the original audience, the core community and the essence that they look for. Things like point-to-point, track lengths and the style of the game, the team are really trying to play to that grassroots rally experience and what people expect from that kind of game. They're doing some really cool stuff. But as I say, I don't want to get a kicking, so I'm not going to say any more.
Can you see DiRT Showdown ever making a comeback?
I think that was probably something that at the time came from the studio wanting to do something more slanted towards arcade-style gameplay and it was really fun, what came out of it. I would say it definitely has its own place within the DiRT franchise, but slightly set aside from the 'core' DiRTs, as it were. I don't think you should look at that and expect that's what future DiRT games are going to be. It was more something that made sense at the time, and people wanted to make it because it'd be fun and it was a fun title. It won't dictate at all the future of DiRT.
GRID Autosport is out on June 24th in North America and June 27th in Europe for PS3, Xbox 360 and PC.
Monday, May 26, 2014 @ 07:44 PM