In-Depth: Far Cry 2's LeBel On Designing Compelling Multiplayer

Much of the attention placed on Ubisoft Montreal's upcoming Far Cry 2 has revolved around the game's open-ended single-player campaign, including its dynamic narrative system, but it's also shipping with an ambitious multiplayer level-editing component on both the PC and console SKUs. Aiming to streamline the team's own level design tools into something flexible, powerful, yet still accessible, has been a challenge to the team, and the goal has ended up driving many aspects of the main multiplayer modes' design. To gain insight into the development process behind that side of the game, Gamasutra sat down with Hardy LeBel (Halo, SOCOM 3: U.S. Navy SEALs), Far Cry 2's senior multiplayer designer. He discussed the intricacies of transposing the single-player game's feel onto a multiplayer environment, how it's harder to design a map when you have to worry about the sun setting dynamically, and why it's more fun to design a level than a gametype. What were your goals for the multiplayer component? Hardy LeBel: Our overall goals for multiplayer in Far Cry 2 were to take the elements that we saw as very successful from the single-player and fold them into the multiplayer action setting. But we wanted to do it in such a way where it was going to be fun and it was going to be engaging, but it was also going to be a very robust and flexible experience -- so that it would really work with player-generated content in the editor. In other words: Make something that was fun to play, make something that wasn't brain surgery, necessarily, so that the players could really experiment on their own using the editor and the tools that were available to them in there. What do you do with the editor that is unique? HL: What we're doing with the editor and what really stands out is the fact that it really is one of the most flexible level-editing tools and powerful editing tools that has ever been released to the public at large. Obviously, Quake has a very powerful set of tools, no question about it, but this is really a next-generation toolset with the mesh editor and the lush graphical content and the sheer density of the natural settings and the lighting options and the weather options. Really that's what moves the bar forward into next-generation territory. When you talk about the mesh editor are you talking about the terrain? HL: That's right, yeah. The terrain editor -- the underlying mesh that lets you control height. Is the tool basically the same on console as on PC? HL: Yes it is. The tool is actually sort of a subset of the full set of development capabilities that we actually use in the Dunia engine. That's what we were using to actually build the game -- the single-player and multiplayer side. And the development process was a very careful consideration of how to take those same incredibly powerful capabilities that we have and figure out how to make it accessible to a broader audience and not necessarily have to train them as a level designer at Ubisoft to be able to produce good results. On that note, are you also going to have traditional modding support on the PC? Have you thought about that at all? HL: We'll see. It's a little premature for us to make decisions about that. We're excited to see what people do with the tools that are out there right now but the future is wide open, I would say, if that makes sense. Have you got a lot of pre-fabs in there? Is there any geometry-level editing going on? What's the scope of what you can do? HL: Well, there's the basic underlying mesh editor and then there's an entire library of all of the building pre-fabs that we actually use for the single-player and multiplayer level of design. So, you're not actually editing the geometry of the physical objects, but you can take them, rotate them, combine them in any way that you want to, interpenetrate them or position them -- things like that. There's a nice suite of tools that lets you parse the library in a really quick and intuitive way. So, you're not actually editing the physical mesh of the objects themselves or part of the underlying 3D geometry, but you can take our entire toolbox of toys and play with it however you like. You were mentioning lighting options and different things like that. What sort of presentational stuff do you have going on there? HL: Well, the multiplayer game actually supports a full range of day/night settings -- a full 24-hour day/night cycle. It's a little bit truncated. In other words, the timescale is not real life. It's actually happening at a faster scale. You can also set the weather. What we're using is a dynamic weather system in the game that takes in atmospheric considerations and condensation factors and time of day and then generates realistic weather based on all those parameters that get fed into it. So, really, what you're doing when you're editing a map or playing in a multiplayer mesh, is you're seeding the dynamic weather system and saying, "I would like to see weather that would be based on these parameters." What other kinds of parameters? HL: Oh, like I said -- condensation and wind factor, and time of day is part of it. I'm curious about designing multiplayer with that in mind. I've talked to [Far Cry 2 creative director] Clint Hocking about how the single-player side is such a procedurally-driven, systemic design. How do you apply that to multiplayer? That's not traditionally the kind of stuff multiplayer is about. HL: No. That's true. That's absolutely true. For us on the multiplayer side, really, creating the multiplayer experience for Far Cry 2 has been an incredible growth process, a real learning process. To be honest, I'm a longtime multiplayer designer myself -- I worked on a bunch of different multiplayer offerings. A lot of the techniques and a lot of the types of things that I would do to make multiplayer levels work and be successful in previous titles just don't work because of the systemic process and, frankly, what the engine is capable of -- like the high dynamic range lighting and the weather factors and everything else and the sheer density of objects that you actually put into the environment in the game. A lot of stuff that I would do to make a multiplayer level in a competitive product successful, I couldn't do here because you just can't use a lot of the same kinds of cues and the same type of techniques. So, it really has been a huge learning process down to the point where, at one point we were playing on a map and nobody could see anything because it was too late in the day. It was too damn dark. We realized that we literally had to change the orientation of the map 90 degrees so that the setting sun was still peeking over the hills and casting enough light so that you'd be able to find your way through the environment, otherwise the shadows were just too dark. (laughs) That's so funny because usually when you think of multiplayer maps, everything is so baked in there. You've got skyboxes and static lighting, and everything is so defined. HL: Well, that is true. Yes, that is very true. Interestingly enough, and I'm sorry to get all designer-y on you here, but what's interesting is that a lot of the best titles and real next-gen titles are based on taking and combining certain random elements together. So, more and more, game design is more about saying: OK, rather than you experiencing the exact same sequence every single time, it is about taking and combining random elements in different ways so that each time the player plays the experience is fresh and interesting and it has more vibrancy and is more alive. On the multiplayer in Far Cry 2, faced with those kinds of challenges -- the dynamic weather systems and all those other elements that we were getting from the single player side, I felt pretty comfortable trying to shape an experience that was going to be fun and interesting for the player. I was trying more to create a framework for them, something that they could enjoy, more than I was trying to create a very specific sequence, if that makes sense. Yeah. HL: It's not meant to be as super hardcore as something like Quake. It's really meant to be more like: here's this suite or this broad palette of fun things that you can play with. In some ways it's the anti-Quake. Someone shoots you and you're slowing down, and you're pulling a bullet out of your arm, and fire is spreading across the grass -- it's almost like a distilled version of what the single player is. Quake is more of a glossy experience in gameplay terms -- very precise, every microsecond counts. HL: Well, it is. And, for us, on the multiplayer side, there was a real challenge, because the single-player side is meant to really immerse you and make you feel like you're there. You know, you're in Africa. But I've said this to Clint -- Clint Hocking, the creative director on Far Cry 2. After the game starts, and there are the sound effects and the lighting, I've had moments where I really feel like I'm in the jungle, and I'm kind of pushing my way through the bushes. But, the truth of the matter is, as we started to play with this very, very realistic setting, and very, very realistic weapons, and stuff like that, it's not that much fun to just get killed. You have to dial that back a bit for multiplayer. HL: Yeah, absolutely. Finding a balance that would have its own unique spirit and be fun to play, but still borrow some of the elements that were really working for the single-player and put them to use for ourselves, was a really big part of the challenge. You mentioned the self-healing. You know, the single-player guys have a much more complicated set of rules about how you can heal, and when you're allowed to heal, and stuff like that, which suits the single-player game. For us, by contrast, in multi-player, we really wanted to make it a bold, dynamic choice. You can heal yourself completely, at any time, but to do that, you actually have to do one of those elaborate self-heal animations, where your camera view goes off of your opponent, and you're quite vulnerable. You're literally sort of sitting there helpless. It's more of a pure gameplay mechanic, straight-up risk versus reward, as opposed to a more immersion-driven event. HL: Correct, absolutely. And so, it varies from the single-player experience, but, what we found is this really fascinating thing, where, you're in there, and you're fighting, and then you're really strongly compelled to take cover, and heal yourself. And so, making that balance between their realism, and their immersion, and what we wanted to create, as a fun playground, was an interesting challenge. Do you have a lot of sort of unusual objective-based modes in this game? HL: We don't have a lot of unusual objective-based modes. We have two. One is Capture the Diamond, which is our wrinkle on Capture the Flag. We're imagining that it's conflict diamonds. It's such a theme, throughout the entire game. The leader of the UFLL, and the leader of the APR, are the faction commanders in the multiplayer game. It's the same actors, and the same characters, coming over your headset, and giving you orders. There's another unique game type, which is Uprising, control-and-conquer gametype. There are three strategic points on the map, and you have a commander on each side, and the commander is the only person who can capture the point. One of the reasons that we stayed away from really complicated or esoteric game types goes back to the editor. If it was rocket science, to kind of make game types, or really make them work on the levels, we thought that that might be overly complicated for people who are just getting into the editor, trying to make their own stuff. So, our gametypes are very object-associated. Obviously, the flags and control points are items you can place in the editor. We feel that once players experience those and start to mess around with them, they'll discover their own combinations, varying the game rules and the map design and the level placement to make their own kind of fun. Did you guys look at Halo 3's Forge editor or gametype options? I suspect Forge is more limited than what you're doing. HL: It is, in a way, more limited than what we're doing. I know that, on the editor side, they were trying to pay attention to all the kind of competing products that were out there. But our impulse, our goals, were really driven by taking the Dunia tool set and the editor in terms of the professional development tools. Finding a way to make that transparent and accessible to as many people as possible, and kind of calibrating that, was the goal. The way that theirs breaks down, is that their map editing tools are extremely limited, but their gametype editor is very extensive. Your balance seems like the exact opposite of that. Is that accurate? HL: Correct. Yeah. That's exactly right. Did you feel that it was important to be conservative with the modes so that you weren't introducing too many potential variables? HL: Yeah, that's kind of a way of restating what I was saying before. We didn't want to make it too complicated to author unique game modes. And to be honest, and I know this is going to sound heretical considering my background [on games such as Halo], but speaking solely for myself here, I'm not as big a fan of modifying the game rules. What I tend to find is that that fragments the overall player community. People are like, "Dude, we don't use that gun," or "We use special rules," or "We play some small in-house variant of the game." It's like the handball house rules on a school playground. HL: Yeah, exactly. Handball rules on a playground. For me, what I prefer is to understand the rules broadly across the board, and to try to have people be comfortable engaging with those rules and playing with them. And honestly, it's more fun to author a physical space, like a level, than it is to author game rules. It's just more fun to sit down and make hills and draw rivers and draw bridges than it is to kind of place spawn points and hook up all sorts of complicated game rules. Even though you can have a big impact on the final product in that way, it's just not as much fun. [laughs] So, I don't mean for that as a knock on the ability to author game rules or customize them. But, my own personal preference is to definitely go in the opposite direction. I was a fan of the original PC Far Cry, and it's been interesting seeing the Crytek guys do a successor [Crysis] that goes in one direction, and then you guys do a successor in a different direction. They're somewaht divergent design attitudes, both are which are taking from the original focus on open-ended environments. HL: Yeah. It's true. The divergence and the thinking process of the single-player guys -- how they've approached the product that they're trying to make -- is interesting. I find it fascinating to sit there and listen to them talk, to goad them a little bit into talking about the topics that they're passionate about. For me, looping it back to multiplayer, it's been an incredible learning experience. Working on multiplayer maps in Halo, working on multiplayer maps in SOCOM, and then coming to the kinds of challenges that we're faced with this engine, with this technology and system -- honestly, it's been a whole new learning experience to try to make next-gen multiplayer just faced with some of the graphical challenges. The story I told you about the shadows -- fascinating, right? You don't have that problem on a Halo map. [laughs] What happens when the sun's setting in Halo? It doesn't. It doesn't happen, it doesn't get too dark. But in this case, it does. The dynamic fire is another interesting, highly tactical, element. As we were designing the maps, we were trying to figure out how to use fire -- one of the signature pieces of technology in graphics in Far Cry 2 -- and how to integrate it so that it was meaningful and interesting, because it is such a dynamic element. You may or may not see big fires on the map, and making it so that it factors into the gameplay in interesting ways is really cool. As an anecdote, we were playing against the Frag Dolls [Ubisoft-sponsored gaming team] at one point. They had come in to playtest the game and give us some feedback. We responded as a team and we went running around the corner of a building, and the map that we were playing on -- one whole side was on fire. The whole thing was on fire. We were kind of standing there thinking, "Oh, crap. Okay, we're going to have to go the other way!" Learning how to use that as a tactical element -- to block off portions of the map, to set fire breaks that will help you defensively or prevent vehicles from getting through -- it's just been a whole new layer of my own understanding of taking those dynamic elements from a lot of single player development and folding them back into a multiplayer setting. I heard a story from the development of the single-player where some tester lit the grass on fire, and hours later, the game ended because the end boss died in the fire, miles away. The fire eventually got to his compound, or wherever he was, and killed whatever NPC you've got to kill at the end of the game. They couldn't figure out what happened for ages. Obviously they made it so that can't happen anymore, but that points to an interesting and uncommon style of game design -- sort of reminiscent of Fallout, where you could just go kill the last boss immediately if you knew what to do. HL: A little bit, yes. I think it was the success of Grand Theft Auto and those games that showed you the incredible power of a procedural approach, exposing the player to those kinds of procedural systems and letting them do whatever the hell they wanted to in those sandbox-style games. But, that being said, something like Call of Duty is really not procedural, at least from my perspective. It's is in the Valve on-rails vein. HL: Yeah. It's an amazing rollercoaster ride where the ups and downs of the pressure and the drama are so perfectly timed and scripted that at the end of the single-player campaign it's like, "Woo! That was a fun roller coaster!" It's just so super fun! There's that complexity of creating the sandbox experience versus the more intricately-scripted single-player thing, or even finding a balance between the two. It's a big design debate. HL: Yeah. It really is a big design debate and I think to a certain extent it has to do with your exposure to one school or the other. One of our level designers on the multiplayer side had a lot of experience as a single-player level designer and was absolutely frustrated by the randomness. He was saying, "Ahh! This is driving me crazy!" He was used to having much more finite control, and over time had to learn to deal with that randomness and embrace it and shape player behavior as opposed to dictate player behavior. Finally he was onboard and became a master at it, no question. It's definitely a learning experience and for some people it's almost like a religion: "I hate it! I'm against it!" So, I think, from a pure development perspective, it comes down to your exposure to it and how well you can develop an understanding of the techniques on the other side and really use them to your own advantage.