Under the Hood of Forza 4

Eurogamer: You can't get all of Forza 4's Achievements without having Kinect. Some of our readers who like to get all the Achievements in a game reacted strongly to that. Why did you think Kinect specific Achievements were the right thing to do?

Dan Greenawalt: Part of it is, there is the reality of what people say about collecting Achievements, and there's what they say. We have all the stats on which Achievements were achieved in Forza 3. There was an Achievement, for example, for finishing every event in the event list. There were Achievements for collecting some of the most expensive cars in the game. And the number of people who have that Achievement is less than I've seen across all the forums so far, arguing about getting all of them.

So that's the general principle. And then there's the actuality. If you look at how many people got that Achievement in Forza 3, it was a minuscule number of people. Now we have 50 per cent more events, so minuscule minus one will be how many people get it.

I've seen on the forums people saying, I don't like this, I'm going to be forced to get a Kinect. I'm like, you wouldn't have got all the Achievements anyways. Statistically, some people will. But when I look at the number of people who argue about this and that...

I have tremendous respect for gamers. I don't want to say I know you're wrong, I'm just saying, statistically, some people are not quite being truthful.

Eurogamer: The EA Porsche issue struck a chord with our readers. Reading your blog post that brought this issue to light, I detected a genuine sense of disappointment that EA has blocked Porche from being in Forza 4. Is there hope the issue can be reconciled?

Dan Greenawalt: Well, I don't have anything new to announce. I pretty much got everything out on the table on that blog, so there are new comments. Obviously, I see it as my role to deliver the greatest car experience we can possibly do to our customers. I'm disappointed in myself. I would like to be able to deliver the greatest cars to our customers and get people really excited.

As far as what the future holds, honestly, everything I know, I put in that blog, so I don't have anything else to add. The big thing is, and the reason I wrote that, is that I care too.

Eurogamer: The Kinect features are interesting. Were there any features you explored that didn't make it into the game?

Dan Greenawalt: A huge number of features.

Eurogamer: Can you give us any examples?

Dan Greenawalt: No. Part of the reason I say no is because of the misunderstandings about how innovation and game design on large scale, triple-A franchises works. There are two aspects people don't commonly understand. It is the job of a designer to make throwaway work. That is our job; to come up not with just one great idea. It's getting a thousand great ideas by breakfast. I don't mean me personally, I mean designers. That is our job.

The right idea at the wrong time is the wrong idea. Often people say, oh, I thought of that 10 years ago. It's like, if it wasn't the right idea, trust me, 40 other people thought of it. There are very few new ideas I hear, because we've been working on this for 10 years. What it comes down to is delivering the right idea at the right time.

When it came to Kinect, it's a brand new model. It's a brand new paradigm. The idea of stuffing controller gameplay, or mouse and keyboard gameplay or wheel gameplay or whatever into it is just a fool's errand. You've got to start with a blank slate. And a blank slate means you don't just have to do a little throwaway work - you have to do a ton of throwaway work.

We have four main integrations of Kinect we covered, but that was born out of thousands and thousands of ideas and months and months of just brainstorming ideation. To be a designer, the work is not ideas. It's shotgun really, is what it comes down to. If you're shooting at a bird, a .22 is much less likely to hit it than a shotgun. We're using that to make sure we've covered all bases and thought of it from all angles.

Eurogamer: What are the issues associated with using Kinect to steer an imaginary steering wheel?

Dan Greenawalt: It's a brain-dead idea. But we did prototype it and threw quite a bit of work away. We're core gamers. Part of the job is to strip away our own bent, our own predisposition so we're designing it for the type of player we're trying to delight and surprise. We knew from the beginning what we wanted this to be was not a replacement of the wheel. If you want to do the 24 hour Le Mans, go get a damn wheel. If you want to do hardcore gaming and drifting and things like that, use a controller, if you're a good gamer.

But when I look at my kids, they're intimidated by controllers. I look at my father-in-law, intimidated by a controller. So the first thing we did, because we're core gamers, is we looked at hooking up the feet to brake and accelerate. Should we do shifting and handbrake and all these other gestures? Those players we were trying to get to think about cars in a new way, they didn't want to do it.

All of the core players, like us, we go, oh, this is neat. I can do all this other stuff... and now give me a controller. All we wanted to do is try it and go back to our normal way of playing. It's like, why would we deliver it if we're not going to use it anyway. A lot of it came down to really understanding, we're not trying to change you. If you're a core gamer, be a core gamer. I don't want you to change. What I do want to do is get people who don't think about gaming to get into our culture, because I love gaming, and I want more gamers in our world. But I'm not going to be able to get them all in if I've got a 16 button controller, and people look at it and go, dude, what the f**k?

Eurogamer: What can gamers expect in terms of post-launch support?

Dan Greenawalt: 10 cars a month with DLC. Plus, we do listen to our community. Just like with Forza 3, it's not that we do everything everybody asks for, but we do look at how they play data wise. Our forums are very active. We try and nurture an active forum community. We also read other forums on other community sites. But we know from our data that the forums are not representing our larger community extremely well. There are a lot of people playing in a way that is totally different than what you'd think based on reading.

We have three primary lenses for deciding not only what features we do in a version, but also what title updates and things we change. We look at what people talk about, how they actually play, and our long term vision for where we want to go.

Eurogamer: Is too much made of the Gran Turismo Forza rivalry?

Dan Greenawalt: It makes for great stories for people. It's not what drives our team. Our team is driven towards that vision of car passion and car culture. And that's what gets us to accelerate. Now, obviously in my job it's important I play all the competition. I play everything. I play Grid, Need for Speed, Dirt, F1, everything. I play a lot of games in general. Part of it is just understanding what's going on. But that's not where our innovation, our ideas come from. That's just knowing where the landscape is. You can't navigate without knowing the landscape.

The other thing that's interesting is, with any of these giant franchises - we have over 350 people working on our team across the globe - when you think about that many people over two years, that's a lot of man hours. You can't maintain incredible agility at that size. We've restructured our team over and over again to give autonomy and freedom to leaders within the team to innovate within different areas. They can innovate in Kinect. We can hire specialists. We can innovate in graphics and hire specialists. We can innovate in physics.

But even so, even with the agility we get from that, we have to fire this artillery way in advance of trying to hit the hillside. So when a game comes out a year, six months, before we launch, we can't do anything about that. There's no reaction we can do at all. I can play it. I can appreciate what they do. We can't react. Which is why our stress is on innovation long term. If we don't innovate long term, we can't accomplish it.

Sometimes games come out with features at the same time. First-person shooters do and you go, oh, those guys copied that guy. I know that, just with the size of teams today, no they didn't. They came up with the same idea around the same time about a year ago and one game beat the other one to market. The ability to copy other games is extremely limited, honestly.

Eurogamer: Gran Turismo has weather effects and night racing. Why doesn't Forza 4? Have you explored the possibility?

Dan Greenawalt: Every version we explore night and weather. We whittle down thousands of ideas, and we have these autonomous teams. We hire specialists. We build the game like an onion. At its core is a drop-dead amazing simulation engine. But we always assume we're wrong, and then find the best way to make it better. It's that commitment that makes me confident we have the strongest simulation, because we build partnerships other companies just can't, and we have a commitment to getting it right at our core.

We then build assists and fun gameplay and things on top of it without ever sacrificing that. Well, part of that is a solid 60 frames per second. And I mean solid. When you start doing multiple projected shadows off of the front of the car... we have 16 players, so 16 cars times two, so 32 projected shadows. That is a very graphically intense thing.

Now, that's totally possible on the Xbox 360, but it means the specialists we have in graphics would need to work on that problem, and it would be a hard problem. The hard problem we chose to take on with our graphics this version was Image Based Lighting, and working with Hollywood. Every version, we look at this long list of features we would do, and they are divided into these different autonomous groups that have the specialists that could actually do it. We just can't take a network developer and say, hey, why don't you do a new particle system. It's not that they're smart guys, but we hired them because they are so good at delivering network code.

Graphically, doing huge particle effects, we would have to have that group that was developing IBL, and that took a long time for us to get it right. We had to implement a whole new way of thinking through the partnership with Hollywood. We'd have to have them working on that from the ground up. So that's the type of thing where we could do it. We could do it on this hardware. But it didn't prioritise higher than IBL, because the IBL and the way the game looks now is stunning, and it's universal. It helps everybody, no matter what type of player you are, having a beautiful looking game that runs at 60 frames per second is awesome.

The things I've read on forums, people saying, well, I wish they didn't do this feature - first, that shows a lack of respect for the people who are going to love that feature, which is fine. I don't expect gamers to respect people the way I do. But, the assumption in that statement is, we have 350 people, why don't you just move them all on to this other thing I want?

It's like, I can't move artists onto it. That's not going to help. I can't move networking devs. I can't move my physics developers. I can't move my AI developers. To hire the best in the industry takes years. Years. To hire the type of guys we get, you have to take people from Hollywood and other game companies. It's the only way you get senior talent. So you can't just decide, we want to do more, let's just staff up.

Eurogamer: Is Forza 4 the last Forza game on the Xbox 360?

Dan Greenawalt: There's no reason it has to be. There's more than one way to skin a cat. A lot of how we get our improvements, such as the graphics, was not with more hardware. It's actually through a knowledge of the hardware. It's actually easier to be on a platform after it's been several years out, because we can borrow from Epic, Rare, Lionhead, 343, and we give back to them.

Plus, we borrow from the rest of the industry: SIGRAPH, DICE, GDC. A lot of people think, what about the rivalry with this company? It's like, we all work in game development. Marketing teams rival. PR teams rival. Platforms rival. We're game developers, and we want to deliver a vision, so we're not really looking at the rivalry.

Eurogamer: But more RAM would help, surely?

Dan Greenawalt: Hardware can always help, but only after you've had some time to trick the hardware. The reason we're able to do what we do is not by using the hardware the way it was supposed to be used. That's what happens on every version. Every game that looks great, the way it does it is by getting the hardware, learning every nook and cranny of it and tricking it into doing things more efficiently.

It's like packing up a moving van. When you unpack a van and you put it on your pavement and repack it, you get more space. Well also it's about knowing every nook and cranny of that van, so you're putting things in the glove box, and you're putting things where you were never supposed to store them. You've tricked it into doing more.

When you get new hardware, it's actually really hard to trick it because you need a couple of years to develop that. It's nice to have other developers like Epic, who are like, hey, we've found a spot, if you rip the door panel open you can shove some stuff underneath the window. The window doesn't work any more but that's OK, you've got more room in here.

There are a lot of PCs that are really powerful now and yet a lot of the games don't look much better because it's an unstable platform. You have to make it for the lowest common denominator. So, would more power and more RAM help? Always. But it wouldn't help immediately. And there's still plenty of things we can do on this hardware because it's not always the hardware that delivers the experience. Auto Vista, the new IBL, new physics engine, new AI... there's a lot of new innovation we have in this game. And this is our third version on this box. So innovation is a lot more than hardware.