Diablo 3 Interview: Lead Designer Kevin Martens
With anticipation for Diablo 3 reaching catastrophic levels, we got to chat with the game's lead gameplay designer, Blizzard's Kevin Martens, who told me that even the most hardcore Diablo fans "ain't seen nothing yet" and that Blizzard is preparing to "punch a mountain in space".
How are you approaching the build-up and the anticipation that fans must be feeling by now?
MARTENS: Well I think that people tend to get impatient, and also, especially hardcore fans who are, you know—those are the people that evangelize your game. Those are the people that stay strong and stay with you no matter how long it takes. But they also (understandably) get pretty frustrated as they're waiting. So I think our goal is to be pretty open with our information. In the past I think we've been very closed. We're trying to open up more and we're trying to be very direct.
You know, Jay Wilson, our game director, will often—if you've ever seen any of our BlizzCon presentations, he'll tell the full, flat-out, honest reason of, "Hey, here's one of our giant failures and what we've changed," you know, "Here's skill system #17 and here's why we're changing it this way, and here's what the risks are, and here's what went wrong last time and how we're trying to fix it." Since BlizzCon last year, where we showed them everything we were planning to do, we cut two of the systems we talked about, and we re-did the skills system three more times.
So we keep people from getting frustrated with the lack of information by being really upfront with the information. I think that when there's a lack of information the people will assume the worst, or will assume negatively, and that's not just a game fan thing, that's just a human psychology thing. So we've really striven to just be a lot more open with what's going on, you know, with the intent of really trying to avoid spoilers. So we want to show people more stuff. There's a lot of stuff we can't show them because it sort of spoils the first playthrough experience for them. So we can't show them everything but we can talk a lot about the gameplay systems and that's what we've done.
Is it a challenge to parse through the expectations that fans have built up to figure out what's realistic and what's too over-the-top to include?
Oh, absolutely. You mentioned too over-the-top to include—that's pretty rare. We're an over-the-top product. We're an over-the-top company. So you know, doing big, ridiculous challenges is one of the things that we do here, which is I think why the games tend to be milestones, and also why they take so long to some extent.
That aside, I think our goal as game designers (as opposed to just developers in general) is to not just look at what people are saying and parse through it, but to look at why they're saying it. What I alluded to earlier is that people will assume the worst from lack of information, so sometimes they're asking for things not knowing what else is going to be happening. So you know, people will be demanding creatures based on their Diablo 2 experience, plus some videos they saw, not understanding that what they're asking for doesn't actually apply, because they don't have the full picture.
So our job really is to figure out: what's their concern, and are we meeting their concern? And we try to—we look past what their suggestion is because people will go right to the solution. You know, they'll say, "I think you should do this." And we've got to figure out, well, why do they want that? Why are they suggesting we do this? What bothered them that got it to that point? And that's another natural thing. It happens internally as well. People jump right to the solution. We've got to figure out what the problem is so that we can find the best solution. Because we're the ones with all the information.
The game has gone through multiple different versions. What's changed through those revisions going into Diablo 3?
The real challenge at the beginning was you know, just like any sequel that's happening years later, was a technology one. There was leaked stuff of an early version of Diablo 3 from Blizzard North that looked a lot like the Diablo 2 engine, you know, a bunch of graphical upgrades, very cool areas, good designs, etc., and in fact (I'm not sure if we've mentioned this before, I think we have, though—if not, you have an exclusive here) we've actually taken resources from that engine. We've taken some of the monsters that were made in the old version and just did minor updates so that they fit in the current engine and ported them right over. Some of the monster designs straight from there have come over.
One of the—jeez, I don't wan't to spoil it [laughs]. One of my favorite monsters in Act 4 came from there. We'll talk about it next week, right? After the game ships. Talk to me then.
The technology challenge is we obviously wanted a much more modern engine, and we wanted more robust multiplayer, huge security passes, etc., to take care of some of the cheats problems with Diablo 2 with all the cheating and bots and spam and stuff like that. And of course in the meantime Battle.net was getting redone as well. So there's just been incredible technological challenges.
On the gameplay side, we had to get that Diablo feeling back that, you know, kill a monster, treasure pops up, pick up treasure. That cycle had to be sort of perfect. And we started with the Barbarian, because first, we understood him, and second, you know that melee feel was sort of the primary one. So it took a while to get that feel perfect. The weight of the animation, the speed, the click, the sound, etc. So that was our first real challenge.
And after that, it's been trying to make a sequel that feels very much like Diablo, but is a different game. We don't want to just make more content for Diablo 2. In fact, Diablo 2 is still doing really well. The whole "Battle Chest" out there actually still sells pretty well, and always has. People can play all the Diablo 2 they want, and we're still supporting it, we patched it recently, etc. etc. So we need to make something different and new and it has to still feel like the game. It can't feel like a game that's set in the universe as sometimes a failed sequel will do.
So how do you feel like you both modernize and fix the problems of Diablo 2?
20 in-game years have passed between Diablo 2 and 3, and 12 years have passed in real life. Has anything that's happened in the last 12 years affected the direction of Diablo 3?
I don't think it's changed the direction as so much the method by which we get there. I look at early vision documents for the game before my time, what they wanted to do (since it wasn't "we" yet at that point) is the same thing that we want to do and have done now, but the ways that they tried to get there completely changed multiple times over the years.
So for example, one of the weaknesses of Diablo 2 (which is a great game, we're building on the shoulders of giants here), the story stuff is usually just a really long conversation. You know, kill a bunch of things and then follow up with a really long conversation, and then one awesome cinematic between each act. You know, that's how the story was told in Diablo 2. And we wanted to tell a better story, but we didn't want to slow down the game. We spent a lot of time, and a lot of years, really, finding ways to deliver story in the exact right bite-sized chunks where people don't feel like they need to skip it because it's too long, and also so it doesn't slow down the gameplay experience. Because blistering pace is one of those Diablo feel things. So that's been an example of a huge challenge.
The other major challenge I would point to would be the skill system stuff. It was so easy in Diablo 2 to make a build that would make it impossible for you to progress beyond a certain point. In normal difficulty, the standard spot where people would get stuck in D2 was the boss at the end of Act 2. People would have a build where they just, no matter how many potions they had, they just couldn't beat him. And you know if you went onto the websites and stuff and built a build according to some expert's version, well then you'd know how to do it. But that's not a good skill system, where you have to go to a website and build someone else's design for you after they've done all the math.
And there's all these crazy things, like if you're doing it—the best way to build a build in Diablo 2 is to hoard your points, but the UI and your natural inclinations are to spend the points when you get them. So there's all sorts of silliness. And that's why we spent so much time redoing the skill system. What we've got now still feels like Diablo, but works very different from Diablo 2. And it's really fun, and it's different, and it's new, and that's one of the good things about making a sequel.
Again, a long answer on the gameplay side. Briefly, the distance between the Diablo 2 story and the Diablo 3 story—picking 20 years was not an arbitrary number. It's recent enough that we can still have characters from the previous games be present in the world, even if it's just their corpse like Poltarh at the beginning of the game that people saw in the beta. But you can still feel their impact. We can still have characters we wanted back, like Cain could still be there, albeit very elderly at this point. But also it's far enough away that we wouldn't be constrained by the story of Diablo 2. The world could have changed in dramatic ways in 20 years. So that's why it's 20 years later.
Would you say that you're trying to make the game more intuitive, so players can jump in and play the right way rather than floundering at certain parts? Is that a trend in the industry over the past 12 years that's influenced you?
I think it has happened in the industry, but not often enough, really. There's still way too complex systems out there. I think that there's too many games where more features or more options equals depth. That is exactly not the Blizzard approach. For us, it's fewer, deeper systems with a very nice wrap-up of complexity. That's how we make depth. And you'll see that in any Blizzard game. You know, like, imagine you're a first level World of Warcraft player and there's just a couple things you can do, all the way up to your level 85 complex badass PVP builds and stuff.
We make hardcore games. All Blizzard games are hardcore games. But we make them accessible. And that's our goal for this as well. And I think that Diablo 2 was accessible probably for the first ten hours of the game or so, but it was easy to accidentally make mistakes. And we—not just intuitive, we want to go beyond that with the Diablo 3 skill system and make it so that people can feel free to experiment and find something that matches their personal play style.
You and I could both play a Barbarian with the exact same skills and the same runes and not take advantage of any of the huge numbers of build differences, and it would still feel differently because we would still use those skills in different orders or at different speeds and in different combinations. And we wanted to make people free to find their perfect match with every class. And that's why the system is the way it is.
Do you think it's still possible to surprise fans after all this time?
Oh yeah. Yeah, you ain't seen nothing yet. Without ruining too many surprises, people have played the beta, and some of them quite exhaustively, but most of the systems in the game, a lot of the things, just haven't been unlocked at all yet. The beta was maxed out at what, level 11? Level 14? Somewhere in there? There's tons of stuff people haven't seen at all. You know, like, the rune system's not in there for example. And that's one of the huge, massive, awesome things that we've done. You know, PVP is launching in a few months and that's something new and different. And of course, there's all the content and story stuff and some very epic moments and different sequences of gameplay and so on. There's a huge portion of the game people don't know about almost at all.
There have been a few Diablo-style games in the interim, for example Torchlight—why do think there haven't been more that emulated that style? Is there anything you've learned from those games?
Well I think as far as why there haven't been more is, I think there've been a few big ones that have and for such a disarmingly simple and direct game, it's actually quite complex to build, if you're trying to do all the things that we're doing. And even us, with all this expertise from all these years, it took us a long time to get Diablo 3 right and off the ground. So I would suspect that there's a lot of companies out there that have tried and then quietly stopped working on projects like this over the years. And there's been some great ones out there.
As for what we've learned, I'm actually drawing a blank on that right now. There's certainly games we respect, you know, Torchlight shut down the team for a while. So did Borderlands, which while not as obviously like Diablo, has a lot of similarities on the item side. We love that game.
It is fun to see people take our itemization approach and apply it to other games, like shooters. I thought Borderlands did a great job on that.
What exactly are the reasons behind delaying the PVP?
We have limited resources, and we still had a lot of work to do on the core co-op game, and that's your primary method of it. So it's really just like, well, many of the people working on PVP have been helping out on this, and we could ship PVP maybe in less than the few extra months we have, but then the rest of the game has to be delayed overall. So it was really just that choice, that okay, let's—everybody quickly concentrate on the main game, including all the PVP people, and all the people that we hired to look into the console stuff, all of them were helping out as well, and many people from other teams. So it was a big sort of group effort at the end to iron all this out.
We would rather get this part of the game into their hands and then everybody can focus on PVP who's needed and they won't be short-staffed and be doing it in bits and spurts anymore. It was just one of those. We can get it out in a very reasonable amount of time I think. A few more months, or—well, we'll see exactly, but we'll get it out soon-ish. And in the meantime there's still lots of Diablo 3 with so much co-op and single player to play that that should keep people busy.
What are you guys anticipating as far as the server load on launch day and what are you doing to prepare for that?
I'm not an expert on that, but I know that there is a live team in all regions who are waiting and preparing and ready. We did our big server load test during the beta period and learned a lot from that as well. For our part of this office we've got people who are going to be here many hours before launch. We've set up a war room. We've got a big screen showing everything going live all over the world. It's going to be a little party [laughs]. There's people who are bringing cupcakes, because why not?
So right here at the office and at all the support centers they've been running scenarios and things for a long time.
It sounds like a scene from Armageddon.
[Laughing] Yeah I think we're going to blow up some sort of rock out in space. Sure. Punch a mountain in space.