We had the opportunity at BlizzCon 2013 to conduct an exclusive interview with Wyatt Cheng, a Senior Technical Game Designer for Diablo III. In the discussion, we chatted with him about the changes coming specifically in the
Mr. Cheng jumped right into the interview talking about how he enjoyed Diablo Somepage, and was recently looking at the site. He compares our
charts to their internal numbers, and considers that data from a developer's perspective. We're honored that he enjoys the resources on the site.
These are the topics covered in the interview; select a topic to jump directly to that part of the transcript:
Combat Changes and Lifesteal
Q: With patch 2.0 and then Reaper of Souls coming out, it sounds like almost all of the combat changes will be in 2.0. How many combat pacing changes should players expect in Reaper of Souls itself?
Probably a lot. Obviously, we've hinted at changes to Lifesteal. One of the things that we struggle with is: to what degree can you affect what people are doing right now, versus what changes we want to make to bring the game to what we believe is a better place? I do think that Lifesteal is fundamentally broken the way it is now. But we're not coming down and just cutting it at 60. Instead, as players level to 70, it slowly tapers off. So combat has the potential to feel even better at 70. But I don't want to disparage anything; 60 needs to feel good too, every level should feel good.
I probably deep-dove into Lifesteal a little more than I should have. There are other features like obviously Act V is a really big deal, which has not only whole new environments like the Blood Marsh and Westmarch, but new monsters for those areas. We also have the new character class, for people who want that, and there's a new feature called Adventure Mode, which we're really excited about. That's coming in Reaper of Souls as well.
Q: Right now, it seems like all the classes end up face-tanking, with a possible exception of Witch Doctors. Is that something you expect to see changing; for some of the classes that were originally ranged and are now face-tanking, do you think that'll be changing a lot in the expansion?
Face-tanking is a complicated problem. Do you want a deep nuanced answer, or the short answer?
Basically, are ranged classes going to be ranged in the expansion? Do you see that changing a lot...
Let me tell you one of the reasons I think people face-tank, and maybe that'll sum up why I think it's actually a pretty complicated problem. In general as a philosophy, we want a lot of build diversity, which means we want a lot of people playing in creative ways. Ah, where to start? Face-tanking is popular right now, because obviously if I have a skill that is a melee skill, as a Witch Doctor there needs to be a reason that I would use it over a ranged skill. More than likely, it does more damage, such as Firebats - Cloud of Bats
. You can't have Cloud of Bats and then have a comparable Fifty-Yard Bats skill and say, "there's no advantage to being close". Because then people would say, "then being close is just the wrong answer". So we give a little damage bonus for being willing to put yourself in danger.
The thing is, if putting yourself up in danger isn't actually
a danger, then it's just: would you rather stand close and do 30% more damage, or would you rather stand far away and do less? Well, as a player I'd rather stand close and do more! So, I say it's a complex nuanced problem, because we have to evaluate: what exactly is the bonus for being up close? How threatening is that? How much does survivability matter? I was hinting already that I think Lifesteal is fundamentally broken. One of the reasons is, doing more damage becomes more healing, which allows me to face-tank. What happens if Lifesteal wasn't the one silver bullet solution, where as soon as I have Lifesteal I just face-tank everything? I think face-tanking is enabled heavily by Lifesteal, because I can just crank my damage to just crank my survivability. So that's why it's sort of a complicated problem.
Will face-tanking still be an option
in Reaper of Souls? I hope so. Will it be the best answer for everyone? Hopefully not. Hopefully that's something where we can say: if you want to face-tank, face-tank. If your fantasy involves being a ranged caster, let's support that as well, and let's make sure there are advantages and builds that want to do that.
Q: Do you think something like a perma-stun Wizard is going to be viable, or something that's going to be scaled back a little bit?
Like the CM style Frost Nova
stunning? Yeah, I think we will probably want to scale back on that. But in order to do that, we should be providing really strong viable alternatives. That's why, when some people have asked, "Why is this the only build, why are Archon
and CM Wizard the only ones? Why don't we have more diversity?" Well, we don't want to just go in there and break the one build that people are doing, without providing alternatives. What we'd rather do is say, "oh you're playing with this toy, can I offer you these other five?" And that's where the expansion comes in.
Damage Reduction on the Crusader
Q: Is the Crusader going to have the melee damage reduction?
That's actually something we're evaluating, that's a fair question; I don't know. We had the Crusader with the melee damage reduction, we played that way for a few months. We took it out a little while ago. It's hard because he's this mid-ranged class. He's got strong melee aspects, he's got strong ranged aspects. At the end of the day we gave the melee classes damage reduction because you take more damage from melee. Of course, this gets into the whole face-tanking thing, so I think we need to be evaluating: what does our combat feel like; what does it look like; how much danger are melee classes putting themselves in; how much danger do Crusaders have to put themselves in? And if we feel he needs it, he'll get it. I'm not against doing something in between: maybe the Crusader gets 15%, or 10%. Maybe he doesn't get any at all, because he's got a giant shield and a passive that lets him hold a two-hander, and that's kind of enough. Every class is different.
[Note: This question was later asked and answered, in the Diablo III Fansites Q&A With Developers
held the next day at BlizzCon. Senior game designer Andrew Chambers stated that the Crusader will not have a damage reduction bonus.]
Item Stats and Priorities
Q: As far as item diversity goes, right now the majority wants trifecta and quadfecta items. The AH obviously fed into that. Do you think with the removal of the AH and the Loot 2.0 changes, there is going to be more item diversity? Are there other new stats that are going to be added in the expansion that might change it?
Yes! We're adding a lot of stats in the expansion.
I think right now, much like the face-tanking problem -- which, it's so funny that you bring that up, because that one thing; I don't mean to rag on it too much, but from a player's point of view I totally get that face-tanking is just this single thing. And yet, I don't want to be rearranging the deck chairs on the Titanic. Yes, that's a problem, but the solution is much bigger than just trying to fiddle with face-tanking.
Back to items: yes, the AH contributed to trifecta and quadfecta, and helped accelerate us into a world where everyone is walking around with trifecta and quadfecta items. But I'd want to ask: why is the only stat people care about their damage? Why is it that I want that up green arrow on damage, and I don't care about anything else? That's what I want to be asking. Why don't I care about my health or my protection? In Reaper of Souls, actually with patch 2.0, we have three numbers now: Damage; Toughness, which is your effective health, it's your protection and your health put together; and Healing, which is currently an estimation of approximately how much Life you regenerate per second.
It's my goal that each of those three numbers be meaningful. Internally, we already have situations where people go: "Oh yeah, I gave up a little bit of damage, but my Healing and Toughness went way up." That should be something that I think everyone cares about. Obviously if you're a glass cannon Wizard, and you're saying, "I'm just never going to get hit, and I'm going to do lots of damage, and I'm going to stand in the back" then you might favor a lot of damage. But even that person I think is going to say, "well I'd be willing to give up a little
bit of damage, if I got a lot more Toughness and Healing." And someone who wants to be that tough armored front-line tank will say, "Yeah I'll give up 20% or 30% damage, to get significantly more Toughness and Healing." So we really want all three of those metrics to matter.
So, push all that aside: even if get to the world where all three of those matter, through our combat pacing changes, that's still not enough. We're also adding even more stats, that have nothing to do with that math. We're adding Cooldown Reduction, we're adding Resource Cost Reduction, we're adding Crushing Blow. We're adding all these stats which can't be quantified simply by Damage, Healing and Toughness, because we want people making these more interesting choices about how they gear. I think the summary would be: we're not trying to break trifecta, we're trying to offer you way more options.
Q: We focus a lot on the site on Diablo III class popularity
. Do you expect that the Barbarian is going to continue to be the most popular in the expansion? Or do you think with the Crusader being added, or with all the changes being made, that other classes will be a little bit more popular?
You know, it's hard to say. I would definitely love for other classes to be more popular. I think there are a lot of reasons why the Barbarian is currently the most popular class. I would love for people to try out the Crusader. I don't personally play just one class; I have a level 60 of every class, I think a lot of people do. So people want a lot of varied experiences. I think that's important, and I hope that people really get a chance to try out all the different classes.
Philosophically, I don't necessarily think we need to be in a world where every class is 20%, either. Like, all the classes being balanced does not mean an equal 20% representation. I think what I really want is: if you like a playstyle or a particular fantasy, that there's a class you can choose to meet that fantasy. I think a great example is: what if I love playing a DoT class? Who do I play right now if I want to be a DoT class; it's like there's no one. Maybe people will play a Barbarian because his Rend
is so good; it's like, what, really? But people are like, "But I wanted to be a Witch Doctor as a DoT class." I think we should fix that, I think we should make Haunt
and Locust Swarm
really good. There's a new passive, I'm not sure if it'll ship because it might be totally overpowered - but there's a passive on the Witch Doctor coming that makes his Haunt and Locust Swarm last forever; I think technically it's 5 minutes or something like that, but it's really really long. And that lets you play as a DoT class. So to answer your original question: do we expect to see more class parity? I think yes, but that doesn't mean we're going for everything to be even.
Console vs. PC Stats
Q: I don't know if you have the stats available, but is the console version class popularity similar to the PC version?
I don't know the stats off the top of my head. Even if I did, I don't know if I'd be allowed to say.
Q: Right now, it seems like Witch Doctors have this dichotomy where they are the least popular level 60 class for softcore, but they're very popular with hardcore. With the expansion, are there going to be other classes with other abilities to lend themselves to being similar?
You mean like hardcore versus softcore? Yeah, hmm... you know that gets back to kind of what I was saying before. We're not looking for equal parity across all the classes. What I really care about, is if someone comes to me and says, "I play a Witch Doctor to me and I love it!" It is more important to me that the people who choose a class and play it, love it, than it is for their to be parity. Do we want 5 bland classes, 6 now with the Crusader; or do we want 6 classes that people are super-passionate about? We want that second one. If it turns out that 25% of people super-love the Monk, and 25% super-love the Demon Hunter, it's more important that people are in love with the class of their choice.
Some people really love the Witch Doctor in hardcore. They say, "his pets make me feel safe", or "I love having Spirit Walk
", or "I love having Spirit Vessel
". I don't want to go and just give Spirit Vessel to everybody. I think if there's something that makes a class very, very cool for one particular reason, that's awesome. We should do that more; there should be classes that are awesome for a very specific reasons.
Proc Scalars on Skills
Q: Very early on, players were complaining about Lifesteal and saying it wasn't worthwhile. You predicted, running the numbers, that Life on Hit was going to scale back and Lifesteal was going to win there...
Oh, yeah, that was a long time ago! That was our reddit AMA
on like the second month we were out. That's a really long time ago, I'm surprised someone remembered that.
So obviously that prediction came to fruition. Are there other things like that, that you saw or predicted, that ended up being completely different than what you expected?
Yes. Are there things that turned out different than I expected... The importance of proc scalars, I think, is something that we underestimated. There are some skills that are huge in popularity because they have a way higher proc scalar, it's kind of out of the norm. The most obvious example is Fists of Thunder - Thunderclap
, which has a way higher proc scalar than pretty much any other skill in the game. And then everyone uses it. And there are other skills which have particularly low proc scalars, and people don't even know which ones those are, I'm guessing, because no one uses them! So I think that we underestimated how important correctly tuning proc scalars was going to be.
Going forward into Reaper of Souls, this is sort of a debate that we've had internally. Should we have skills that are good because of their proc scalar; is a high proc scalar one of the reasons I should be using a skill? Or, should we normalize the proc scalars across the board? We've opted for the second. We're saying: look, we have a ton of cool skills, but if we have a skill that has a proc scalar that is way too high, that one
skill is going to unlock all these other broken combinations. Yeah, it's cool that that combo exists, but it's not
cool that I can't use any other skill. We'd rather say: if you want to be a
Wizard, you shouldn't have to use one or two specific skills with high proc scalars. If you want to be a Critical Mass Wizard, you should be able to use Critical Mass and combine it with any other skill that has the geometry, or the elemental damage type, or the casting mechanics that you want.
And with that, we were out of time. Our thanks to Mr. Cheng for his time and insightful discussion, and to the Diablo III community team for organizing the interview.