All this year, world's largest publisher Electronic Arts has been messaging that it has changed -- but the proof is in the words of its developers, and the quality of its upcoming games. While the jury is still out on that, as most of them aren't due until later this year, we do have the words of the creative teams.

Here, we present a discussion with DICE's Owen O'Brien, senior producer on Mirror's Edge, an unusual title for the Swedish-headquartered company best known for its long-running Battlefield team-based combat series.

Mirror's Edge is a first-person adventure title with an intriguing context-sensitive, dynamic action set, as showcased in a recent trailer for the game, and notable inspirations from parkour.

The title is "...set in a conformist dystopia in which communication is heavily monitored by a totalitarian regime, and so a network of runners, including the main character, Faith, are used to transmit messages while evading government surveillance."

In this in-depth Gamasutra interview, he discusses everything from its creative inspiration, to designing a HUD-free game, to how the team's use of Scrum methodologies has allowed for unexpected gameplay iterations.

On Prototyping

The thing that I want to say is, I'm extremely jaded, and I actually was still impressed by the Mirror's Edge demo.

Owen O'Brien: Well that's good!

I was just talking to Ben [Cousins], who's working on Battlefield Heroes. I was at the DICE Summit, and I saw John Riccitiello do his city-state speech [advocating greater independence for individual EA studios], and Ben said that Battlefield Heroes couldn't have happened, and Mirror's Edge couldn't have happened, without the new EA city-state mentality.

OO: I think the city-state mentality is there to encourage things like Mirror's Edge and Battlefield Heroes; I think it's there to allow studios the freedom to try things. So we've got Mirror's Edge now, which is an amazing game -- of course I would say that -- but trailing behind that, there were lots of other ideas that were killed.

So it was a product of a lot of trial and error, and, you know, trial and error is expensive, and the company's got to be behind you and say, "OK, you know what, that didn't work. We'll kill it and we won't fire you for it."

So, trial and error. Do you do a lot of prototyping? Do you do a lot of just sitting in rooms, coming up with concepts? How does that phase of things go?

OO: Yeah, I mean, basically when I joined DICE two and a half years ago, from EA, I came into a team that had an idea, there was a game in development, or idea in development, we kind of rambled with that for a little while, then we decided it just wasn't going to work. It was too far outside our realm of expertise as well, so we decided to get back to doing what we do best, which is first person.

So the, sort of the genesis of Mirror's Edge is, we had the idea, and then everybody went, "You're crazy." So the first thing we did was, we literally just did pencil sketches of what we would see on the screen.

We did that, then we turned those into an animatic, then we did an animation test -- you know, just running in Maya. After we got that, we kind of looked at it and went, "OK, if we could actually do this, it would work." And then we basically built a rough prototype, and we've been iterating on that ever since.

EA DICE's Mirror's Edge

When it comes to development methods, do you do more agile-style development, where you iterate quickly?

OO: On the team, we've been running agile development for 18 months now? Maybe two years. So yeah, we're using a system called Scrum. It's fast, agile development.

Your meetings, when you work on this project, how do they go?

OO: The sprint reviews? They go pretty well. I mean, I think initially, people were kind of a little bit hesitant about it, kind of like, "Can't you just tell me what to do, and I'll do it, and I'll give it to you?" But then, they've become more and more interactive, and more and more. It's almost like an entertainment system now; all the teams put together movies of their work, and they've started adding music to it, and we've got a whole load of great behind-the-scenes footage of fun stuff that people would be working on. So they have really embraced it.

Designing a First Person Adventure

One thing that you've left out of this game is a lot of weapon combat, and I'm wondering if that coincides somehow with the concept of scrum, where you say, "This isn't working; let's get rid of it quickly," rather than in a more traditional development setting, you would have a milestone where, say, six weapons are meant to be implemented by six months from now. Does that affect these things?

OO: Yeah, it did, to a certain extent. I mean, it did allow me the freedom, as the senior producer, to say, "Ehhh, you know what? I'm just going to take guns out of the game completely." And that was OK, because we hadn't done a lot of planning around it. When the game started, Faith did have a base weapon.

But then we kind of quickly got rid of that. We're having enough trouble trying to make people understand that it's an action-adventure, and if you give somebody a weapon straight out of the box, then they think it's a shooter. There is weapon combat in the game -- you can snatch weapons and use them -- but it's really not the focus. The focus is on the physical human being.

I'm trying to think of an example of a first-person adventure game -- and of course there are games that have toyed with first person combat, like hand-to-hand combat, or first-person running and jumping. I'm failing to think of an example of something that really dialed into that.

OO: Yeah, I don't think there is. I'd like to hear it, if you do think of one, because I don't think there is. That's why a lot of stuff we're doing is cool and it's new, but it's also risky, and hasn't been done before.

I think a lot of people have tried. I think people kept dipping their toes in the water a bit, but they didn't really go for it. I think once we decided we were taking out the gun, then it was like, "OK, so now we really have to make this work. We can't fall back."

I think a lot of people in the past have made first person shooters, and then they've decided to try to add in movements at the last minute, and that's just not going to work; you need to be committed to it from the get-go.

I would say that, in general, HUD-free games have not worked out. Basically. Most games that ditched the HUD have actually felt just as contrived, or in some cases more contrived, than games with the HUD on.

OO: Interesting thought. I mean, for us it works because we've got a lot of stuff in the world that tells you the information you need to know, like where to go; and you don't need a health indicator because it's very obvious how healthy you are, and your health regenerates.

You don't need a mini-map because the Faith-vision sort of guides you through the world. Bullets are not important... So, I don't know. So far I think it's working out well for us. We haven't suffered from it.

Something that I'm interested in from a design perspective, that seemed forward-thinking to me, is how interactive objects become red. Could you talk about where you got that idea? Why do you want to keep blinking arrows out of the game?

OO: Well, it was built out of a desire, really. We want the player to be able to move quickly through the world, and read the world as quickly as a runner would be able to read it. You know, to use the analogy of Jason Bourne, he looked the world differently to everybody else, and I wanted to get that sort of a sense through it.

So that's really where the idea came from. Initially we had it sort of non-dynamic, so it was just red marker-posts through the world, and it was almost like there were red objects that the runners left for other runners, to show them --

You mean almost like checkpoints, right? Like in a race.

OO: Yeah, exactly. So we experimented with that initially, but that as well felt a little bit contrived. And it was also limited, because you can't do the stuff that we can do now with this dynamic system; so we can change pipes, or telegraph poles, or we can change anything.

So it's a very flexible system. And it's something we can dial up or down depending on skill level, or work with how quickly it fades in, or how far away it fades in; so those are variables that we can play with to really tweak and guide the player, without feeling like it's a trail of breadcrumbs through the world.

Zeitgeist and Inspiration

It's got that parkour, free-running vibe, and that's actually informed two very successful, and I think also quite good titles in this generation: Assassin's Creed, and Crackdown. Do you think that's a function of the zeitgeist -- it just feels right to draw from that?

OO: It was kind of coincidental, but basically again it was born of a desire to -- in a city, you want people to move in the vertical plane very quickly, so we needed a movement system that would allow you that real agility. So that's where the desire came from.

Because we've been working on this for couple of years now -- I think recently, parkour has kind of ended up everywhere. It's in Casino Royale, it's in the Bourne films, it's in Madonna videos, so it's started to permeate into more sort of a mainstream.

But at the same time, we don't want to say that we're a parkour game -- which is probably why I didn't say it. Because then it becomes too niche. I mean, we're parkour in the sense that Prince of Persia, or Assassin's Creed are parkour.

Ubisoft Montreal's Assassin's Creed

I'm not trying to imply anything about the way that things work at EA, but it's like, if you give that word out, it'll become a bullet point. Whereas, if you can draw from something cultural, it becomes an inspiration. And it doesn't really matter whether or not you tie into the culture that you're working with.

OO: No, I think that's fair. And I think as well, the other name for parkour is "free running," and that's very much -- that even more fits in the storyline of the game. This game is about freedom. About freedom of movement, but it's also about freedom of thought, and freedom of action. So "free running" really worked for us, in that respect.

We've been talking to a lot of developers recently, and oddly enough, on shooters, who are very politically aware. A lot of politically aware shooters have come out -- are coming out. Like, I mean, granted, quality varies, but we're talking everything from BlackSite: Area 51, which was very aware of the Iraq conflict; we've got Turning Point from Spark; we have Metal Gear Solid 4 that's about to come out -- that's very much about the PMC situation. As is Army of Two, which is another EA title. I mean, when you talked about tapping into what's culturally relevant to gamers right now, during your presentation, what do you think?

OO: Yeah. I think what I want to do is... It's very easy to look at this game, to misunderstand this game, and say, it's one girl against this police-state dictatorship. It's not. It's more subtle than that.

One of the core questions that the game asks you is, how much of your personal freedom are you willing to give up for a comfortable life? And the other sort of theme for the game is, you can't force other people to live by your rules and your society, even if your society is better -- even if you have got better health care...

But you live in Scandinavia, so you've got better health care.

OO: We have, yeah. We have. I'm quite happy with it. (laughs) But, to be very honest, I'm a big Joss Whedon fan, and a lot of the things in the story of the game came from Firefly and Serenity. I was listening to this director's commentary, and...

So, actually, that sentence that I just used is actually a direct quote from Joss Whedon; the basis of Firefly and Serenity is, you can't force other people to live by your system, even if your system is better. These people want to live on the edge of that society.

Again, in Serenity, The Operative actually says, "This is not an evil empire. We just don't understand why you don't want to be part of our happy club." Obviously, they take it too far, and similarly, that's kind of what happens in our game as well: the mayor of the city decides to take things a step too far.

EA DICE's Battlefield Heroes

This goes, to an extent, for Battlefield Heroes as well -- it seems like right as Japan is descending as a developing force, it seems like the games that are influenced by Japanese culture to a greater or lesser extent are ascendant in the western studios. This game reminds me of the big shining metropolis, which is an icon in anime for years, and it's something that you're feeding on. Am I projecting there?

OO: No, no, I don't think so. I mean, the city is a very definite blend of east and west. It's taking you to Singapore, or Dubai, or any of these sort of very modern cities, and if you look closely at the signage, it's all in dual language.

It's the wrong look, the wrong atmosphere, but I kind of envisage the city in the way the city of Blade Runner would be if it was today. Heading in that direction, you know: a clash of east and west. Obviously ours isn't the dirty, rainy, depressing sort of Blade Runner thing, but it has that mixture of cultures.

You mentioned that the main character is atypical for a videogame lead, and that's something I think we're struggling with as an industry -- especially with female characters. It's known now that in Uncharted, they wanted to have the female sidekick character to be a little bit less sexy than she was. I have a friend who's working on a title where the marketing team is telling him, "Don't make her average-looking. Make her Hollywood ugly." Which means make her gorgeous, but with glasses. Your character is attractive, in a conventional way, but at the same time, can you speak to that?

OO: Yeah, sure. I mean, again, to go back to my Joss Whedon fetish, I like that we have strong female characters. And I wanted a female character that females would like, too. So Faith, we've done a lot of tests, focus tests, on her and with her, as well.

So she resonates with females as well, because she's aspirational without having pneumatic breasts, or ridiculous body proportions. So she looks fit, and healthy, and agile, without being silly. So, she's... I think, again, by accident or design, we've designed a character that appeals equally to men and women. Which is really good.

"The Biggest Outsourcing Project at EA"

Technology-wise, are you using your own engine technology on this game?

OO: No, we're actually using Unreal 3, but it doesn't look like Unreal 3.

It certainly doesn't. It's funny -- we talked to Mark Rein, and we sort of accused him of the saminess of Unreal 3 games. He said, "You're totally wrong!"

OO: Yeah, I think we wanted a very distinct art style, and we needed -- lighting is very important in our game, so we actually worked with a company in the south of Sweden, called Illuminate Labs.

They've developed a lighting solution with us, called Beast, which gives us a lot of very soft shadows, and color bouncing, and it's what makes the world feel solid and real. And, also, it gives the game its unique look.

So you're outsourcing some elements of the development.

OO: We're outsourcing it to the extent that... I think there's a figure bandied around that we're the biggest outsourcing project at EA, so far. We've outsourced a lot of the art.

I wouldn't say the lighting system is outsourced, so much as co-developed. We're [also] using an external agency for the 2D artwork/animation we showed here -- we'll actually be using that in the game, to tell the story.

It's interesting, because of the size of the EA organization, would suggest, at the face of it, "Do they really need to find people who don't work for EA to do things?"

OO: I think you do. I mean, EA and, particularly, DICE, we focus on what we're good at, which is making games. Then hire professionals to do the other stuff, that maybe we haven't done so much.

Certainly at DICE we haven't done [cinematics]. Particularly the 2D style of cutscenes. We have no experience on that, and I wanted a very distinct look, so I went to people who know what they're talking about, and they can give me what I need. And, in general, it works out cheaper than having to hack away at it internally, and not get as good a result.

Do you think -- are all the companies you're working with externally, are they in Europe? Are they only in Europe, or are you working with companies in Asia and North America and all?

OO: Some of the outsourcing is -- or was; we've finished it now -- was in Asia.

Oh yeah? For art assets?

OO: For art assets.

Got in-game art assets, like in-engine art assets?

OO: Yes.

Did you work with companies in China?

OO: Shanghai.

How did you feel about what you got back from them? Was that a difficult process, to get back what you wanted?

OO: No. I think it worked out pretty well in the end. We didn't make their job easier by... you know, we did this very unique art direction, kind of mid-flow. So that kind of screwed things up.

Like, "Oh, sorry! Can you make everything white?" And they're like, "What?" So, yeah, that worked out. We're very happy with the results. It worked out pretty good for us.

In discussions of outsourcing, what keeps coming up is that communication and relationship-building are the only ways to get good results out of outsourcing companies.

OO: Absolutely. I think that what really helped to make it work for us is that we sent a member of the team out there, and he actually lived there for, you know, six months.

A lead artist?

OO: Yeah.

And did he just work with the art team there, to get everyone on the same page, and get everyone working?

OO: Yeah. Absolutely. And he was kind of the gatekeeper, as well, we knew that when our stuff did come to us, that it would be through him. So that just saved us no end of hassle, I think.

That might be a difficult proposition, though, to say to someone, "Go live in Shanghai for six months." I mean, I guess if people who were -- you know, at the lead level, you're talking about people who have, wives, or husbands, and children, and other responsibilities.

OO: Yeah, it's a... I mean, the guy we sent is 22 and single, so he loved it.

That's pretty young for a lead, isn't it?

OO: Yeah. He's very talented. We've got a very young team. We've got a mixture of old people, like me, but DICE is a very young company. I think it benefits from that.