In-Depth: Why Halo's Multiplayer Almost Didn't Make It

One could argue that the multiplayer component in Bungie's original Halo: Combat Evolved single-handedly made Microsoft's Xbox a staple in college dorms nationwide, giving the company's fledgling console a foothold. But when Microsoft acquired Bungie, at the time a veteran PC and Mac developer, and repositioned Halo as a flagship Xbox title, its multiplayer mode was almost axed in the interest of expediency and cost-saving -- a move that may have altered Microsoft's early status in the hardware race. Gamasutra recently caught up with multiplayer designer Hardy LeBel, half of the two-man team that brought the mode back from the brink. After Halo 2, LeBel went on to head up Zipper Interactive's SOCOM: U.S. Navy SEALs franchise -- itself a multiplayer staple on Sony's PlayStation 2 -- and now serves as multiplayer lead on Far Cry 2. In this retrospective interview, he recalls his Bungie days and comments on the various design approaches available to multiplayer developers. Were you on Halo when it was still targeting PC? Hardy LeBel: I was at Bungie when Halo was a PC project, but the origin of the title was out of the Chicago office, and I was the creative director for the San Jose office. When we got bought by Microsoft and relocated, it was clear that for us to be able to hit the launch window for Xbox, we all had to work on Halo, so we all kind of got pulled in. What was your project before that? HL: I was the lead designer on Oni, which was a kind of a third-person action title. When we got bought by Microsoft, Alex Seropian and Jason Jones, who were the two principals of Bungie, came to me and [former Bungie engineer and animator] Michael Evans, and said, "Multiplayer is cut from Halo because we're trying to make it really work on the console and we just don't have the resources." But we threw a fit and were like, "No way! You can't cut it! It's just too cool!" They said, "We were hoping you'd say that -- because you two guys have to resurrect it." Wow, talk about changing the course of history. (laughter) That component of the game is a big part of what made it so ubiquitous. HL: Well, my goal with the design for Halo was to make something that was a shooter that played like an action game. In other words, I honest to God wanted to make something that would have felt like it could have been made by Nintendo. It was just -- oh yeah, you get in there and everything feels good. It's smooth, it's really accessible, the sound effects are really accessible. Even the naming conventions -- I'd like to point out that I didn't call it "deathmatch" because I felt like [the term] "deathmatch" was too hardcore, perhaps. I called it "Slayer" instead, because I wanted it to be more broadly accessible as a naming convention, like [the Halo gametype] "Oddball." Bungie had a tradition of that -- in the Myth series and so forth. There was a history of off-kilter stuff in the multiplayer gametypes. HL: Yep. Absolutely, there was that tradition. So the two of us basically dove in to resurrect multiplayer. Michael made a set of tools for me so that I could create multiplayer maps that plug into the Halo toolset. It wasn't the full set of Halo tools and we didn't actually have a level artist, so a lot of the levels in Halo were me learning how to use 3D Studio Max. (laughter) Really? HL: Yeah! I was like, "Oh, I wonder what happens if I use this extrude tool. Whoa! That opens up all new possibilities!" (laughter) Anyway, it was a very, very small team. So I worked on Halo, I was a contributing level designer on Halo 2, and then I was the creative director for the SOCOM [U.S. Navy SEALs] series of games. I left Microsoft and I went over to Zipper Interactive. That's interesting, because the SOCOM games were almost the counterpart to Halo in terms of being the big flagship multiplayer games for the PlayStation 2. HL: It's true, yeah. Absolutely, absolutely. In each one of those games, to kind of bring the conversation back around, in Halo and in the SOCOM series of games, encounter design and level design is very much about creating a basis for the encounters and then populating it with a bunch of different elements that are sort of randomly generated. Every time you play Halo, if you were to play the same level or the same encounter three separate times, you're going to get different guys every time. And you're also going to get different guys depending on the difficulty setting. You can try that. Next-generation development is more about saying, "OK, we have this bed, or this group, of elements," and you're really trying to conduct them more than you're necessarily trying to explicitly say, "I want this guy to run this way and this guy to run that way." I've spent a lot of time and experience trying to shape the random elements together to make something fun and interesting. In terms of including all the crazy modes, Halo still does that more than most games -- although I imagine you guys were influenced by Unreal Tournament in that regard as well. HL: Well, sure. With Halo, I can authoritatively say that we learned at the feet of Unreal how much value you can glean and get out of the ability to customize your own rules. However, in terms of the overall fun play experience, I personally never got as much huge value out of weird, kind of custom rules so much as a really screamingly good map with kind of good fundamental weapons and some complexity to the combat model. You know, that it was sort of deep enough for me to be able to explore and enjoy. For me, that's so much that you can really sort of get into and plumb. Counter-Strike is the absolute opposite end of the spectrum, right? Right. HL: Counter-Strike is all about map customization, really not about game play customization. Love that. Just love it. Then some games are sort of neither -- you've got Team Fortress 2, which is more, "We've whittled this down to a few really solid things that we think work. We're pretty confident you'll like this mode on this map, because you're going to be playing a lot." That seems to be working for them. HL: I think, it is. Yes. It can work. It is interesting how modal that is. In other words, if you find TF2 to be to your taste and you like it, then you'll like it anyway. But, if not... Right, you're kind of locked out. HL: Exactly. You're kind of frozen out. So, I think, for that particular product, that strategy makes sense, because there's a proven history, and there's an audience out there of people who are sort of looking for that team based, class based game play model. But, yeah, with not wiggle room, you either love it or you don't love it. That's a little bit tough. In single-player as well, that's Valve's design aesthetic, is to go for that highly authored experience -- but they're also doing Left 4 Dead, which has the dynamic spawning and all that. HL: Oh, yeah. Absolutely. I should say, I worship the ground that Valve walks on. For me, their authored experiences are better than anybody in terms of creating and maintaining that amazing sense of location and feeling. But, yes, a lot of games -- Halo, SOCOM, Far Cry -- are looking more at introducing those random elements so that it's [about] the fun set-ups. And really, so much of what comes out of this systemic approach in the single-player side in Far Cry 2, because it is so systemic, everyone's story ends up being incredibly unique and incredibly personal. You and I may have run into separate checkpoints, separate encounters, separate gunfights, and our personal story about what the weapons did or what the weather was or anything else, becomes personal.