Fracture

Platform(s): PlayStation 3, Xbox 360
Genre: Action
Publisher: LucasArts
Developer: Day 1 Studios

Advertising





'Fracture' (PS3/X360) Developer Interview

by Adam Pavlacka on Sept. 30, 2008 @ 9:00 a.m. PDT

In addition to extraordinary weaponry that allows players to do things previously only imagined, each side of the conflict, Pacifica and the Atlantic Alliance, boast soldiers with powers beyond those of ordinary men. Genetic augmentations provide Pacifican forces with amazing abilities, while Atlantic Alliance soldiers like Briggs counter the threat with the more "traditional" method: cybernetics. The differing states of superhumanity result in balanced yet stylistically different combat tactics that have never been seen before.

The East, now known as the Atlantic Alliance, symbolic of their union with Europe, puts its faith in surviving this new world in cybernetics, an established yet evolving technology now more than 150 years old. On the other side of the flooded continent, the Western states, now called the Republic of Pacifica and having allied with Asia, resort to solving their problems at the genetic level, effectively restructuring the DNA of its inhabitants - a method the Atlantic Alliance finds morally reprehensible.

By 2161, it only takes a hint of unauthorized military preparation in Pacifica for the newly restored president to order a strike in the heart of Pacifican territory - an outpost in the now dry San Francisco Bay. This strike leads to the unthinkable: an epic conflict with global implications fought on U.S. soil. As a soldier in this struggle, Mason Briggs uses explosive, terrain-deforming weaponry to change the face of battle: He not only destroys the land in his path, he outright transforms it to gain the strategic advantage in completely unscripted ways no game has ever seen. With such a devastating arsenal at hand, Briggs never leaves any battlefield the way he found it.

WP: Who has the honor to speak with us? State your name, rank, and occupation!

I'm Dan Hay, senior producer and art director at Day 1 Studios. I'm Deke Waters, associate producer working on the single-player campaign in Fracture.

WP: Fracture is a third-person shooter with a unique twist. It's the first one where you can manipulate the environment on the fly. How do you design a level for a game that essentially has a level editor built in as a core part of the gameplay?

DW: Whenever you start off with a game where the core fundamental thing that a lot of games take for granted, the ground, is a movable, almost a character in the game, you really have to take a lot of things into account. You have to take into account where the enemy AI are going to be in the world, the enemy AI has to take into account where you're going to be, and it's not always a flat plane. It could be anywhere. It could be behind the hills, it could be 30-40 feet in the air, it could almost be anywhere, so what we really went back and forth on is exactly how to make the AI interact with that in interesting ways. So what's going to happen when you play the game is, you're going to go into the level and throw a mound up. You're going to hide behind it. The enemies that are in that world are going to react to that mound. They're going to give you a different kind of scenario each time you throw up a mound. They'll come over the top, they'll hide behind it, actually use it as protection from you, they'll come around the other side — it really gives a lot of variety to the game.

WP: And what about from an art and engine perspective? Obviously if you start off with a level that's relatively flat, and you start throwing hills and valleys and spikes into it, you're going to have a whole lot more polygons that were suddenly produced. How did you guys manage to ensure that the engine kept running at a solid clip and didn't chug, as is seen in a lot of other titles?

DH: That's a great question. I think the big thing for us was to create systems. You think about all the different things that that's going to affect. I mean, it affects everything. The ground can move, so you can't light it in the conventional sense. You can't use baked-in lighting because you need everything to be dynamic. The ground can move, so it affects animation. Other games, the ground is flat, or a character needs to go up a flight of stairs. In this, the interpolation of an animation can happen at almost any degree, and the AI has to able to respond to that. Like you said, we've got to have a strong enough mesh on the ground so that we can make the topography look interesting. Plus, you've got different types of mesh. And then we have lava and XP61, and we have game types that require you to lower the ground for that. It gets expensive, and what we tried to do was focus on lighting in advance because that's where you get into a lot of trouble with rendering. You come up with a usable lighting solution in advance and make sure it did everything that we wanted it to do from the point of view of when we're in that playground. Can we light characters successfully? Can we make characters that look great in that environment? Can we put a proper shadow density? Can we get the polygon count that we want? Can we get the population and the number of dynamic assets? You get into some scenarios where you've got rocks that are on a wall with all the different crates and barrels and different assets that are out there. You shoot that off the wall, and you drop it in, and you've got 50 to 100 assets out there, and you throw a vortex, and you pull in a bunch of AI, and all of that stuff is running real-time and being paid for in real-time. Really, when you think about it, you think about the great stable of engineers that we have in our Hunt Valley section of the studio. Those guys did a fantastic job, and it really is due to us sitting down in a room and talking about the game in advance and really thinking about the structure and how we were going to make that work.

DW: Something I'd add to that is you have all of the stuff that Dan's talking about in the single-player campaign. When you get into a multiplayer experience, you quickly realize that not only is it the same level of destructibility, but it's also more frenetic. Things are happening much faster, and this is going over the Internet. You're actually having real-time explosions and the vortex grenades and all of that stuff that happens in single-player — over the network. It's pretty amazing what they're able to do with that side of things as well.

DH: The key thing that we wanted to do with Fracture — and what Deke was talking about is — we wanted to make sure that we really tried and tax what people consider next-gen to be. It's not a gimmick, it's a feature, it's a full-service game, it's got depth to it, and when you look at it in multiplayer and you see all that stuff reacting and moving real-time, it's unbelievable the amount of freneticism that's on-screen.

WP: With everything going on in the single-player and multiplayer game, how do you go about essentially "training" the new player? A lot of people may come into this thinking that it's futuristic and sci-fi, kind of like Halo, so I'm going to play it like Halo. How do you keep them from getting lost and not being able to find an exit, or realizing the different actions in multiplayer? How do you ease them into things?

DH: Deke and I were smiling because how we do this is, Deke, Jeff, myself, and a couple of other people, locked ourselves in a room and we focused on a tutorial that we — I don't know, is "argue" a good word to describe it? —

DW: (laughs) We had a lot of different viewpoints.

DH: Yes. It was a really, really tough one with Fracture, since it's an entirely new mechanic. Deke had ownership over a lot of the beginnings of the tutorial, and he did a great job in saying, "Look, these are the mechanics that we have to teach, this is how we need to dollup this out to the player," and then soliciting responses and putting a series of situations in front of the player. We had to make sure that it's not too heavy-handed but at the same time it's not so easy that you can blow right through it because the dangerous part is getting all the way through the tutorial and not really needing to use TD or not feeling like you were confident with it. The game's barrier to entry — we wanted somebody who was a shooter player to be able to go into this game and go, "This is a really solid shooter, and I really enjoy this, from the AI, from the weapons and everything," but we also wanted someone who isn't really proficient at shooters, to be able to go through using TD and apply it in a multiplayer match against somebody who is proficient, and be relatively successful. And it opened up that barrier to entry. As for the specifics of how we craft that experience in the tutorial, Deke will do a much better job of explaining that.

DW: One of the things that LucasArts and focus testing brought to the table is the notion that we really want to open this up to newer players without making it a "Playskool" version of a shooter. We wanted to make sure that this wasn't going to offend the hardcore crowd, but we're introducing a scenario that nobody else has really seen before. That was really a challenge, and in early focus testing, we would run through the game and notice that people would go back to their old habits of playing it just like a regular shooter, and they would meet that with limited success because they don't really have a whole lot of cover, they weren't really thinking about using cover the way that we want you to use it, so a big goal of ours for the tutorial was to make sure that we were dollopping out the idea that you need to use cover, you need to use TD for navigation, just keep it right on the precipice of what we're thinking about in the game. Just when you forget about it, just when you come across a couple of enemies and you forget to use TD for cover, we hit you with a turret that's going to remind you that you can't get past this thing without throwing some cover up. There's a couple of tricks in there that hopefully won't stand out too much as you play through the game, but as you get into the second mission, as you get into the third mission, you'll start to remember to use this on your own, and it becomes kind of second nature as you play through the campaign.

DH: One of the things that you'll notice happens as you play the tutorial and as you go through and see the demo, is that, if we do it right, you shouldn't really notice that you're being taught. The first time that we use TD is to get over, under and through a couple of things, to gain access, to have exploration, and then when you go into combat, it'd be pretty easy for a player to go, "I'm under fire, I need to run to static mesh and hide behind it." If you noticed, the first time we introduced the TD in combat, in a way, it's kind of done for you. Pacificans throw out a tectonic and they create a loaf, and you are able to hide behind what they did to you. Deke did a really good job of putting out this little thing that you have to get on top of. The next time, you climb on top of that, and you roll around the corner, and the next thing that you have to do is save one of your cohorts, who's pinned down, and you do that by stopping the Pacifican fire coming in. Then you can go around the corner, and you face the famous turret — I think Deke and I talked about this for about three months. We went through maybe 20 different iterations of what that was going to be, whether or not you should kill the guy, whether or not the turret is unmanned, whether or not it's behind a door — how to finally be able to say, you've met an enemy, you've successfully defeated that enemy, but the key element was that you use TD to do it, and then you use TD to get out of the room. That was a really tricky thing to do.

WP: Moving on a bit to the weapon design. You've got some weapons that are fairly standard for the genre, you've also got TD variants of those weapons. How did you guys go about evolving the weapon design, and how did you settle on which weapon deforms the ground and which one doesn't?

DW: I think in the very, very early stages, we wanted to make sure above all else that we were making a shooter, and we wanted to make sure that we had all of the tools that most of the other shooters out there have and that shooter fans enjoy using, but we knew we had something special with terrain deformation, so we really wanted to bake that into what you kind of held as a truth that you held with the shooter formula. I think the best example of that is probably the Torpedo ST-4 Rocket Launcher, and what that allows you to do is basically stand behind a loaf or a tectonic mound and fire a torpedo that travels underground and detonates whenever you want it to. It's a variant on the standard rocket launcher that you have in other games, but it has a real unique terrain deformation/Fracture twist to it.

We kind of tried to think about that with all of the weapons that we created. We have the Vortex Grenade, which is kind of the darling of the game. You throw it, and all hell kind of breaks loose. We started thinking that it would be really cool if you could just put that wherever you wanted to, so that led to the Loadstone Rifle, where you basically shoot a puck into the world, and where it lands, it's going to create an obscene level of force that just pulls everything in.

Day 1 does a really great job at iteration and figuring out what's fun and how to really get at the core of what makes something unique but also playable. Controls were always important to us, the camera was always important to us, but the feel of the weapons, I think, whenever you get into our game, you can kind of tell that we put quite a bit of care into making sure that they feel right, even though they may not be something that you're totally familiar with, we wanted to attach it to enough familiarity to where you could pick it up and play with it.

WP: Just talking about the iterative process, what's something that didn't work early on in the game and got changed or modified to an element that appears in the final game? Is there something that you can use as an example of how you guys went through the development process?

DH: Obviously some of the characters and how they respond. The Sherman and [code name], that is, the Creeper — that was the early name for it, right? — so we went through a lot of iterations on scale, on speed, on the veracity of the attacks. You spend a lot of time dealing with the ground, and with characters like we think of bosses and mini-bosses, like the bola that's in there, we wanted to make sure that everything was tied to TD, so we spent an awful lot of time with characters, with bosses, thinking about how to make that work, and it was just getting it into the playground and play it. Play it over and over, and see what feels right and feels good. The one that we really played with, that we haven't really talked about too much, is the [code name]. We've put a vehicle in this game, and it's pretty cool. The Growler — right. It's gone through so many names and iterations, that I keep calling it by its old name. It is the Growler, and we focus on making that unique, and I think that the value of the tires, and the speed and what it does and whatnot, it was a really cool vehicle, and how we end up putting a weapon in that, and that's not something people have seen a lot of. We incorporate some TD weaponry into the vehicle, and that in itself is one of those things that we iterated on and iterated on, so that was pretty cool.

DW: I'll talk a little bit about the Growler itself because there was a lot of back and forth on that. There were a lot of things that worked really well, but there's a pretty high risk factor whenever you put a vehicle into our environment. We have a lot of places that's pretty much anywhere on the track, where you would normally have a racing type game, right? Ours is totally dynamic, so you can haul ass down the road and all of a sudden, there can be a huge divot or a huge hill right in front of you. Our vehicle has to be able to maneuver around that, and over that, and through that, and make it still a believable vehicle with believable physics. I think that's something that we did pretty well.

One thing that probably didn't work as well as we'd like is the initial stages of the weaponry for the vehicles. We really wanted to have something that you could drive and you can shoot, kind of a combat vehicle. We found out pretty quickly that that was tough. It was really making the player keep up with a whole lot, so a couple of secessions that we made to that turned out to be for the best — we placed the turret on the vehicle as an auto-turret, and we turned what was originally the ramp gun, where you could shoot a ramp in front of you so you could be able to jump with the vehicle wherever you wanted to — and turned that into a torpedo weapon. It was just so much fun. All of a sudden, you have a torpedo that you can use in the on-foot sections, on the vehicle. So that was, in a nutshell, some of the things that we would go through in the iterations of this game.

WP: Let's briefly talk about multiplayer mode. We talked about some of the design challenges that you had with the single-player levels, where one player is running around and changing things. How different was that process when you were trying to design multiplayer maps, where you can have a whole team of people deforming the map at will and moving things up and down? How did you guys attack that, and did you realize during playtesting that early map designs were broken?

DW: Interesting story about multiplayer. We had just gotten the network code running we were just getting into the game, and we had two players playing against each other. One of the first things we noticed is that whenever you throw a dirt mound in front of you, you instantly give the person you're fighting a height advantage. So all of a sudden, we were looking at this game that we really wanted to have part of the core focus be on the multiplayer. We really wanted to not give the enemy a height advantage, and so that's the birth of giving a little bit of a speed variation whenever you're going up and down hills. Again, our engineers in Hunt Valley absolutely answered the call on that, and they said, let's implement a system that levels the playing field, so to speak.

But as far as the actual level design, there wasn't really a whole lot of difference between the single-player and multiplayer games. I mean, we had our constraints as far as how much memory a PS3 or an Xbox 360 have, but for the most part, philosophies were essentially the same, with the core difference being that we could control when the AI was going to take away your cover or give their own cover. You can't really control that whenever you're playing in a multiplayer environment, so there's some pretty crazy back and forth that you can get when you throw yourself a loaf to protect yourself, and instantly, it can go away, based on who you're playing.

DW: One of the things that also happened that was pretty cool, we'd watch people play, and that would almost help us invent a new type of game, like Team Kingmaker. We were watching somebody play the game, and their style was to walk up to enemy AI and subsonic underneath him and have all of the stuff drop down in the level and crush them just by proxy. They were doing this all the time and it was really successful, so a couple of discussions later, we thought, "How about if we had a King of the Hill type thing where you had to hold a control point?" Part of that was making sure that everybody was in that control point, so you had a finite choke point. Let's just put a vortex next to it, right? And all of a sudden, one person gets lucky and throws a vortex in there and you've got five kills and it's fantastic. But to make sure that people aren't just pounding on that, that choke point moves. You've got to jump all over the place to try and get to it. That was a great way to discover what Fracture could give. It was a unique gameplay element. Fracture was about discovery. Take a lot of the things that we had seen and learned and applied to this, and discover what's cool about it and discover what's new.

WP: Is there anything else that we haven't talked about, either in the single-player or multiplayer, that either of you wanted to add?

DW: I think one of the things that we haven't talked a whole lot about during the process is the key enemy encounters that you're going to have as you go through the campaign. One of the first mini-bosses that you run across is Sheridan's elite Hydras. So far, you've kind of been playing through the levels and battling guys who are soldiers, just like you. Then you hit the Hydras, which are bioengineered to bounce all over the place. They leap extremely high in the air, and they have a rocket launcher. So, it instantly changes your tactics, changes how you've been playing the game up to that point. That's kind of the philosophy that we want you to go through every time you're introduced to a new enemy character in the game.

Fracture is scheduled for release in North America on October 7, 2008 (Oct. 10 in Europe).


More articles about Fracture
blog comments powered by Disqus