Racing Evolution: Forza 3 And The Changing Driving Sim

Dan Greenawalt, director of Forza Motorsport 3 at Turn 10 Studios, is passionate not only about cars, but the evolution of the racing genre itself. When Sony and Polyphony Digital's Gran Turismo first hit the market, its tagline -- "the real driving simulator" -- marked a change in the way console gamers interacted with racing games.

Now, as he approaches the launch of  the third installment of the Forza series, Greenawalt forsees the baseline for game design in simulations changing irrevocably. The line between arcade and sim is blurring and shifting, he says, and developers will have to change their mentality to capture a wider audience and stay with the times.

Greenawalt also discusses how profound changes to team structure allowed for a better, more productive environment and a sequel which allowed for all of the meaningful evolution beyond its predecessors.

How has the development been for Forza 3 -- is the team happy to see light at the end of the tunnel? Were there some challenges there that you're glad to put behind you?

Dan Greenawalt: I think in a lot of ways, we went through some crazy growing pains on Forza 2. On Forza 1, we were trying to just build a team sort of from scratch and from the ground up. That was pretty difficult. We slipped a few dates. We made a game we were really proud of, but we were so unpredictable. And then on Forza 2, we tried to shoot for the stars. We felt we had a team, and we're going to really crank it up and see how quickly we could get a great game out, and we slipped again [laughs].

For Forza 3, we actually went back to the drawing board, process-wise, and we said, "You know what?" We looked at SCRUM. We started doing sprints. We broke out our team. Rather than functional groups -- like design versus dev versus art -- we broke them into strike teams that were just there to develop features. And that allowed us to get far more predictable and actually grow a bunch of leaders out of the team.

So, we've been in development for two years, and the funny thing is in that amount of time, we got a lot more done because we had a lot more leadership down in the ranks. And we also came out with a much better game, and we hit our schedule. So, we actually went in front of management about a year ago, and said we were going to go into cert on a specific date, and within one year, we actually hit the date on the day, which I've never done in 12 years of game development.

So, it wasn't just a matter of you having a basis to build off of with the prior Forza games, but rather an overhaul in structure at the studio?

DG: Yeah, I'd say that was actually the biggest contributing factor. And we did have existing tools, but we rebuilt our pipelines from scratch, so we didn't even have that to go off of. We stripped out a huge layer from a rendering engine, which allowed us to rewrite large sections of our rendering engine. The game got a huge overhaul. But honestly, it can't be attributed to anything besides great producer work -- we hired a bunch more of those -- and great leadership from within the team.

How has the development team size grown from Forza 1 on Xbox to Forza 3?

DG: I'll give you a little bit of nitty-gritty here. So, we had, I think about 24 full-time employees on Forza 1. And we then bolstered that staff with probably 70 contract staff that were on for most of the project -- a couple of years. And then we outsourced a lot of our artwork to a group in India and a group in Vietnam. That made it so that all-in, we were probably 200 people, that's my guess.

Then in Forza 2, we got up to about 250. We increased our [full-time] staff, increased our [contract] staff, and increased the number of people in India and Vietnam. And now, for Forza 3, we're just over 300, all-in, with again [contract], [full-time], and a fully-burdened outsource group. And that's been pretty much since the beginning.

We grew immediately, so as soon as we shipped Forza 2, all of the [full-timers] took some time off, the [contract staff] kind of went away for a little bit, we disengaged with our outsourcers, and then within three months, we were back fully-burdened. We were up about 200 people, and 300 within a few months after that.

Was there any kind of fear on your part, as one of the leaders in this studio, that things were just getting too big? Is it just a matter of managing it well?

DG: Well, that was actually why we restructured ourselves so much. You've got two options if you start fearing you're too big. You either need to scale back your ambitions, or you need to come up with a new approach to steering such a large vessel. So, by driving more of the ownership and more of the leadership down into the ranks, we were able to do more "divide and conquer," as well as grow our bench and grow the future leaders of the studio. And you have to grow those next leaders. For me, I need to grow the person that's going to come and take my spot.

Can you cite specific problems with the Forza 3 development that you might have run into that you eventually had to conquer? It sounds like the strike teams solved a lot of problems before they even happened, but there's inevitably some problems that do come along.

DG: Yeah. For Forza 2 in particular, I was a bottleneck, as was our executive producer. We simply were involved in too many decisions. And that just means there's a large army of people who can't operate without you at the head, and that causes tremendous inefficiencies.

Now, the flipside happened when we went to strike teams. We gave ownership to people that weren't quite as senior and were not quite as experienced in a leadership role, and so they didn't always make the same decisions that a more seasoned veteran would make. Some of those decisions, we lived with, because it was kind of six of one, half dozen the other. You know, "I would do that differently, but no harm, no foul. It doesn't really matter."

And some decisions, we then have to say, "Look, where you guys ended up is not where we need to be as a franchise and deliver the right experience we want to deliver." So, it came back to me a lot of times just reinforcing the vision with those teams rather than sticking at the feature level, evaluating their work and saying how well is it actually accomplishing the vision and saying, "You need to go back to the drawing board because you're not quite hitting the vision yet."

Was it worth the trouble to empower the employees?

DG: Absolutely. There's no way we would have hit our date if we had not driven the leadership down into the ranks more. It's not even possible.

There are a lot of new features in Forza 3 that are particularly meant to ease the new players into the Forza world, like the Livery Editor -- you don't even have to race -- and the storefront. And there's the mid-session rewinds. Do you think the next big step in the racing sim genre -- and I see Forza moving in this direction -- is to make it less scary for the novices?

DG: Yeah, I think that's really part of modern game design in general. I don't know if you've played much of Batman: Arkham Asylum?

Yeah, I actually just finished that a few days ago. I couldn't stop playing it.

DG: Yeah, it's an awesome game. And the thing I love about it is you can see the heritage. You can see going from a standard kind of brawler or adventure game -- call it a Tomb Raider if you like -- all the way through like a God of War, and up to this game, where it's still got a lot of the core mechanics of, like, an Ikari Warriors with some of the bosses you might fight, and yet it's not challenging.

They didn't do the challenge by punishing you. They did the challenge by just making it fun... You know, the checkpoint system is great, the brawl system is great. It's just a really good game and not very punishing.

And I think the problem with simulation in general is it's gotten this reputation -- a well-deserved reputation, I might add -- for being overly punishing. And it's something that frankly a lot of our fans ask for, but I don't think they really, frankly, want it. The hardcore fans of the racing sim genre, in particular, are constantly calling to be punished more. It sounds weird. It sounds like we're in some strange sadomasochistic relationship. But they're always saying, oh, there's a lot of our resistance in our community saying, "Well, this rewind feature is going to ruin the sim."

I could not disagree more. I am a sim racing gamer as well, and I can't go back to games that don't have it now. When I play other racing games and they don't have this rewind, I just feel like, "Are you kidding me? I mean, this is kind of old school." And so, in a lot of ways, it's like when you look at Batman or Prince of Persia. When they had the rewind feature in Prince of Persia or Batman, there's no death falls, there's no big penalties. When you fall off a cliff in Batman, they're just like, "Hit RB," and you just miraculously get saved. That's awesome. That doesn't make the challenge less. It doesn't hurt the tension. The game still has great tension.

You still have to earn all of your victories [in Forza]. [You can] redo the corner, but you still have to do the corner. You still have to make the passes. You still have to beat the AI. So, we're trying to accomplish that challenge. So, do I see more racing games going this way? I sure hope so because I think this is the industry needs to go.

You know, people talk about capturing the casual audience, broadening, and all that, and that's what I want to do as well, but I think part of this is just embracing modern game design and realizing that people are different and what they're looking for is great entertainment, and entertainment does not have to be punishing. It can be, and it has been, but it doesn't have to be.

Were the other sim racing fans that worked at Turn 10 standoffish about putting that rewind feature in there, like the fans?

DG: Absolutely. That's the funny thing that's happened on the team. Yes, there was some resistance on the team, there's resistance in our community, and I felt very strongly that this is an important feature. We actually started planning this before [Codemasters'] GRID came out, but it was actually helpful having GRID come out because more of them played it and said, "Oh, I get it."

But the truth is we implemented [rewind] pretty simultaneously with when GRID came out. All props to Codies. I love Codies. They did beat us to the market with it, so I'll give them all credit. But internally, we were prototyping how we wanted to implement it, and we found that even internally, people were challenging themselves more because of it. So, they would turn up the AI. They would turn the car damage to a higher level of difficulty. And they never would have done that in Forza 2 because, you know, you get six laps in, it's been a long race, you make one mistake, and your car is totaled and you're going to lose. Or the AI passes you because they're just too strong, you can't reel them in, and you're going to lose.

The rewind allowed people to actually have more challenging races, even the sim hardcore guys. They could not upgrade their cars much to have a lot of great wheel-to-wheel action. They know, "If I make a mistake, I haven't lost the race." And that's what I think won over a lot of the guys on my team, that they realized that they are actually now all of the sudden driving harder races. You wouldn't think that the game would get more challenging, but like Batman, it can become more challenging simply by making it less punishing.

It does make sense because if in real life, if a race driver had the ability to rewind a wreck and just practice, or practice a botched corner over and over again to improve it for when the real race day came, they'd be better at it.

DG: Absolutely. I'm a much better racer. It's funny, too, because I've now worked on racing games for 10 years, and I'm actually becoming a better racer in Forza 3 than I have been from Forza 2.

On Forza 2, when you're racing and you're on the final lap, and then you screw up, you just have to do the whole race over again. That may be realistic; that's what it would be like in real life if you were racing. You'd mess up, and you'd have to completely start over. But it goes back to the whole entertainment versus punishment thing.

DG: Yeah, and on the other side of this, which is interesting, I worked with some European developers a decade ago on a racing game that didn't even want to put a "Restart Race" option in a game -- because you can't restart a race in the real world. And they thought that was going to take away from the tension. And today, you can't make a racing game without "Restart Race." That's simply what racing games are.

And I feel very strongly that we're years away, maybe three, maybe five, from this just being the cost of entry. If you make a racing game, you're going to have a rewind. We're already seeing the green line from Forza 1 getting adopted all over the place. Frankly, it's not in the real world, but it makes you so much more successful, and it's more fun and entertaining.

We've talked a little bit about where you're going from here with the racing sim genre. Are you guys trying to steer away -- pun not intended -- from the term "sim"?

DG: No, not really. I think there may be other things on the kind of marketing side as far as how we keep from intimidating people that really love cars, and Forza is a great game for car lovers because we have so many and we treat them with such reverence. So, I think maybe on the more global marketing side, we're looking to shy away from that term "sim" because it just has connotations of "too difficult," and what have you.

Certainly internally, as well as just when I talk more on the PR side, I'm not afraid of the word "sim." We just simply need to get people to understand that the old idea of what a sim was and what an arcade was are simply gone. It really comes down to licensed cars versus not licensed cars. When you've got licensed cars, it's very difficult to do big explosions and things like that. In fact, Blur is one of the few games that I've seen that's able to do that, from Bizarre Creations. But they had to do it by working with a very small group of manufacturers. They can't do Ferrari, they can't do Porsche, and they can't do a bunch. And that's a choice they made by being licensed.

I think as graphics have gotten better and better and better -- you've got games like Ridge Racer, which looks beautiful -- it's basically meeting the sim versus arcade line, it's getting very, very blurred. We playtested the hell out of Forza 3, and we basically jumped over GRID, Need for Speed, Forza 2, PGR, and Gran Turismo as far as ease of use, so we had kids coming playing all these games that are metrics.

We're trying to figure out how we can make the game fun and rewarding, increase our playability scores, and we never dumbed down our physics at all. So, if our game is defined as easy to pick up, and quick to get into and play, then Forza is an arcade game. And yet, when you turn all the assists off -- we just had that pro racer in last week, and we got one of the fastest guys in our studio, really fast, top of our scoreboards in Forza 2, and he was getting his ass handed to him by a pro rice driver.

I think that speaks to the simulation level, that [the pro driver] got in and he knows these tracks, he knows these cars, they break where they should break, they're hitting the right top speeds, he shifts at the same point in the real world. So, he was just killing people. And there you go with basically the best sim that people are going to find, and yet easy to pick up to play for even a kid. How do you define arcade and sim now?

So, we've got accessibility, and that's a big step, but beyond that, we've got online, and then Microsoft has Natal coming out with new control methods. Where does racing go from here beyond just making the games bigger, prettier, and more accessible? Or is that it?

DG: No, no. We didn't even get into the accessibility message of how we changed the career mode. We've actually reworked the career entirely using the same design ethos that we did in-game. That makes the game a lot more approachable.

I don't know if I can speak to the racing genre. What I can say is that the vision we have for Forza is to turn gamers into car lovers, and car lovers into gamers. And so, for me, it comes down to being more beautiful because that helps you connect people with that real looking car. You can smell the car, you can feel the car. It's why we're always going to be a sim because that's just part of really experiencing a car. So, in Forza, what I'm most interested in is new tools and technology that allow us to turn around cars faster.

Jalopnik just had an article about Fiat and Ferrari working together to make a Fiat 500 that was tuned by Ferrari. It's just a cool car, and you read an article like that, and you go, "God, I want to drive that car." Well, for us, it takes use three months to make a car, and that's a lot of people and a huge investment.

What I'd really like to be able to do is turn that around so we can deliver a car in like a week. So, you read about it in Jalopnik, and then a week later, you go to Xbox Live, and bam, I can go drive that really cool car. So, in a way, getting rid of Road and Track -- that sounds bad. [laughs] I'm describing a new experience that people get from Road and Track. I'm still going to read Road and Track when I'm on an airplane, but in my living room, rather than getting a subscription to a magazine or even going to the internet to find out about cool new cars or even if you're looking to buy a car, what I'd really love to do is have all of the new model lines just there. You can drive them, experience them, and get excited about them. Technology-wise, we're years and years away from that.