The Design of StarCraft II

Dustin Browder's game development career stretches back more than 15 years, to when he worked at Activision on games like MechWarrior 2: Mercenaries, then moved to Westwood Pacific (eventually merged into Electronic Arts Los Angeles) to work for six years on numerous real-time strategy games, mainly in the Command & Conquer series.

In 2005, he joined Blizzard as the lead designer of the feverishly-anticipated StarCraft II -- and now he's back at Activision, in a sense, with last year's Vivendi/Activision merger.

Gamasutra recently sat down with Browder to discuss the inevitable difficulties that come with updating a 10-year-old game while trying to steadfastly avoid feature creep -- and why even after more than half a decade making RTS games, he still wasn't fully prepared for a lead design role at Blizzard.

With StarCraft II, you've said you're trying to avoid making a game that is significantly more complex than StarCraft, even as many individual elements change. How do you make the call as to what stays and what goes?

DB: It's just really tough. We have to make these calls on a daily basis. There were for a long time, and there's still a little bit of it, big debates on the team as to what is enough and what is not enough. What is too much?

We just walked the line, and we look at stuff that we sort of feel is core to the experience: "This is a defining element of the Zerg. We have to have the creep. It's defining." Or, "Siege tanks are a defining element for the Terrans." Then we look at elements where we'll say, "This was fun. We love these things. The fans have obviously been using them. Vultures and spider mines are huge -- but they're not defining elements. You could live without them. I can imagine the Terran army without these things."

So it's a combination of conceptual elements and mechanics. Where we think we can do better, we'll try do better. Stuff that we feel conceptually is not necessary, we'll remove. We've always felt that solo play and Battle.net were the areas where we could really do some stuff that was crazy, that was new and really interesting. And multiplayer really needed to harken to the game's legacy, while at the same time creating enough strategies so you don't think, "Well, I've played this game for ten years, guys. What are you giving me here?"

We want to have enough that it's still fresh. I think we're walking the line pretty well right now.

Even between the single-player and the multiplayer, there have been design differences that mean there are discrepancies in terms of what units are available.

DB: Totally.

Do you worry about that being unintuitive, especially for people who are new to StarCraft?

DB: Not really. We did for a little bit, but then we looked back at our previous games and realized that our solo campaigns have never prepared anybody for an online experience at all. That never worked, right? We always sort of touted it that way -- "It's going to prepare you" -- but it never really did.

Looking back at that, we feel that never really works anyway. This lets us make a much more compelling solo play experience. It can run free and be its own gameplay experience with all kinds of units and all kinds of upgrades, which wouldn't have been possible if we'd restricted ourselves to only the multiplayer set, because the multiplayer wants to become small.

It wants to be reasonable. It wants to be enough that you can keep everything in your head -- so not only do I know what I can do next, but I know what you can do next, and I can play the game in my head a little bit before we actually engage.

You want that sense of there being a very limited number of opening moves in chess, and then a larger set of second moves, and so on.

DB: Yeah. And how many moves can I think ahead of you? That's determining whether I can win or not. So the multiplayer experience needs to be really tight, while the solo play experience doesn't have to be that tight in that sense.

But if we restricted ourselves to the multiplayer units [in single-player], we would ultimately lose a lot of gameplay. We've got a lot of tools that we're going to use -- our challenge mode, like our tutorials, like our improved score screen and improved replay screens -- all to try to help players make that transition from solo play to multiplayer so they can acquire the right skills, instead of leaning on something that never really worked with to begin with.

For the multiplayer, I think you were the one who said you were being so stringent that whenever you add a new unit, you must remove an old one.

DB: Yup. Keeping it tight.

How frequent are the internal battles over whether a unit gets cut?

DB: Constantly. Constantly. And I have guys who tell me I should cut everything and start all over. And I have guys who say, "Don't you dare touch anything. It's perfect." It's definitely difficult. There are definitely the extremes on either end.

But there is a wide group of people in the middle who feel like, "Change some, but not too much." And then they'll argue about which ones are okay and which ones are not okay.

Ultimately, we have an advantage of having a very robust engine, a very strong data editing tool that allows us to try things out. It's very easy for the design team to put things into the game -- "Hey, I know the medic was your favorite unit. We're going to try the game without them. Play it, and see what you think."

Once you get over the sort of the initial "Oh my god!" emotion reaction and you start actually playing the experience, then it's very easy to prove or disprove to anyone whether your choices are good.

You come from a background of RTS design prior to your work at Blizzard. I tend to think of Blizzard as a fairly insular company. How much now do you look at what other developers are doing in this space?

DB: Oh, we look at them all the time. We're hardcore PC gamers. We play lots and lots of games made by lots and lots of developers. We try to look and analyze what is working for them in their game, or what is not working for them in their game. Is there anything about what they're doing that we can learn from, either what not to do or what to do? We're constantly looking at what other people are doing and trying to decide what's right for us and what makes sense for us.

At the end of the day, often, our games have their own style and personality, so we end up often speaking to our games and not just importing things willy-nilly from other games. But a lot of things that we've done have been inspired by things we've seen in other titles that we liked, but that we thought we could do better or that we thought we should do differently for our game in particular.

Are there any specific RTS games you've played recently that you've liked?

DB: I certainly found [Relic's Warhammer 40,000:] Dawn of War II to be interesting, and I learned a lot about RTSes by playing that game because they've done some interesting things.

Did you bring any of your experience at Westwood -- I guess it's EA LA -- to Blizzard?

DB: At the end of the day, no. Not a damn thing. [laughs]

At the end of the day, I learned nothing. I thought I was more prepared than I was when I came here, because I'd made RTS games at that point for five or six years. I thought, "Okay, I've got some chops here," but Blizzard is really its own beast.

Blizzard has its own style, and more importantly, Blizzard has a lot of institutional knowledge, which a lot of other studios don't have about making these games and making them great. At the end of the day, what I brought was a basic level of creativity, an understanding of how these games are built, and... Yeah, that's all I brought. [laughs] And hopefully, a positive team working attitude, right?

But I had to relearn kind of all the basics all over again, because the games that I'd made before were fundamentally not designed be an e-sport. They were not designed that way.

Command & Conquer, and so on?

DB: Yeah. It's not a focus of those games. Command & Conquer 3 tried to do that, and they're going in a different direction. While I was there, certainly we wanted to make good multiplayer games, but e-sports in Korea, to the level that it is at Blizzard, was not part of our focus. Certainly, the kind of context that we have in the Korean community at Blizzard goes way beyond anything we had at EA at that time.

There was just a lot of relearning to do, coming into Blizzard, to truly understand what StarCraft was all about.

On a coincidental note, it's interesting that EA recently said that with Command & Conquer 4, they would be doing a lot of the online authentication processes with single-player that you guys soon after also announced -- really trying to link the single-player and multiplayer experiences. There definitely has been an outcry against both Blizzard and EA in response. Do you feel you needed to reach a certain point historically before you could just say, "Alright, internet is pervasive enough that we can just do this now"?

DB: I don't know what the thinking is at EA, but I know for us, we've seen the success of World of Warcraft. We know people have internet connections. We know a lot of them do. And every PC you bought for however many years now comes ready to go. And how common is broadband?

When we shipped the original StarCraft, you were looking at a game where a lot of people were on dial-up. Not everyone is up on dial-up. We built a game that was functionally designed to work on your PC, and if you have this piece of equipment and you know how to make it work, you can then play this other part of the game.

Now, looking at what the PC is today, that's not how the PC ships. The PC ships today with the internet. It comes with everything that you need to make that work. That is the machine we're building for.

That is the platform, so it just makes a lot of sense to us, since that is the way it's been now for many, many years. We've seen online-only games become a huge, huge success that it's something we can actually use.

We can actually leverage this now into our design process and actually do something cool with it, we hope, and have a fully integrated experience that shows you the news -- "Oh, there's the campaign right there," "Hey, my friends are on," "Hey, what are my friends doing?" You can just feel like you're all part of the experience.

Certainly, looking at what other companies have done, and looking at Xbox Live, which is just a blast to play on, you see another example there of someone who's fairly successfully integrated the whole experience in a really positive way. We hope to accomplish that as well.

From another angle, what about pessimism about the PC platform generally?

DB: Whatever. PC games have been drying for, how long now? Shouldn't it be dead by now? I mean, it's been on its sickbed for ten years. Give me a break. Obviously, it's doing fine. I think that if developers make great games, people will find them, wherever they are.

We get this every couple of years from the movie industry. They're like, "Aw, nobody's going to the movies." And I always think to myself, "Yeah, but you know what? You haven't put out anything I want to see?" And then they ship, I don't know, The Dark Knight, and they're like, "Naw, we're having a great year! Who knows why!?" Well, you made that great movie so we all came. Make great movies, and we'll come.

So, if people make great games, then they'll show up. But if people don't make great games, people will wander off. But they'll always come back if there are great games. It's hard for me to say, though, being inside Blizzard. Obviously, we're a PC company, and obviously we're very comfortable with that, and we feel we've had a certain amount of success with it.

Looking at what we do, I think it's fine. I think it's a very viable platform. I think it's a very fun platform to play on. I think there are things about the PC platform that make it superior to other platforms. It's all about what kind of games people play.

At this point, Blizzard hasn't made a non-WarCraft game since 2001. It's now spent longer just releasing WarCraft games than it did making all the other games for its three main franchises up until that point. Does that seem odd at all, especially from the perspective of someone not working on a WarCraft game?

DB: For me, I don't think it has anything to do with anything. It's just all about scheduling and resources and what happened. World of Warcraft was not something this company expected to be this successful by far. They were hoping for a couple hundred thousand subscribers. That was what had been done in the past -- "If we could equal those numbers, we'd be fine."

So, I don't think there's any sort of conscious focus on a particular franchise in that respect. The Diablo III team has been working really hard to get their stuff to a point where they could present it, and they finally did. It was last year when they announced and said, "Hey, we've got something to show."

We wish we were shipping by now as opposed to still working on the game. So, I think it's just that these games are taking longer to make than we'd like. And the World of Warcraft team is doing expansions. They've got a strong team. They've got an established toolset. These guys know what they're doing. I think it shows in every expansion. They just get better and better and better every year.

Certainly, we're hoping, as we're developing this new engine [for StarCraft II], that once we get to that point, we'll be able to maintain a higher rate of speed than we have in the past, but I think up until now it's just the way it's worked out.