Creating good random map generators is a difficult task in video games, and something we’ve seen done with relative success in all of the Diablo games. Diablo 1 had four dungeon types with quite different maps for each area, though all were just big squares with different arrangements of the pieces within them. (The Hellfire […]
Diablo 3 Podcast conversation about multiplayer co-op issues in Diablo 3. What are the benefits of playing with others? Should there be more party skill bonuses? Why are the Diablo 3 clan tools so meh? The curse of the double-Unity requirement. Show features Aahzmodius, Wolfpaq, and Flux. Click through for approximate segment starting times: 0:30 […]
- CaptainFalcon - Charge/Holy Fire Paladin Themed…
- [1.07 & 1.13d SC FAM FTS] ISO/FT
- A "Could Be Useful" Amulet - Item Find Thread August…
- The Daily -- All Daily, Every Daily
- Guardian Predator - Fire Hunter Druid
- The Impeachment Thread, vol.1 - "WMD"
- Anything Goes D2 Lounge!
- Off-Topic in the OTF
- Forum Mafia Statistics And Awards Thread
- Vote: Which are your favorite Legendary Gems?
- New Hamachi Network - For Fun D2 Multi-player Gaming
- Patriarch Dervish, Whirlwind Barbarian (Exceptional…
Wyatt Cheng Diablo 3 Interview via Live StreamPosted 16 May 2013 by
Yesterday, Wyatt Cheng spent an hour chatting on a fan via a Diablo 3 stream. It’s more of a casual chat than an interview, but there are some good nuggets of info, which I’m transcribing below. (Wyatt was followed by Travis Day for an even longer chat, which you can read here.)
The transcript below is *not* word for word, since it was a long conversation with a lot of chat and digression. I therefore tried to distill and abbreviate Wyatt’s comments into more direct, discrete points.
Can we see some sort of self found play mode?
Wyatt: We’ve given it a lot of thought. Right now I’m playing some self found Hardcore; alternating time on that and my higher level Paragon softcore characters. We like the idea of self-found, and of course players can just do it themselves, but ideally there would be some kind of recognition or insignia to denote self found characters.
An issue is trying to do it without cluttering up the UI. There’s a balance between giving expert players the options they want while not complicating things for other players. (Such as with Elective Mode and advanced tool tips.) Do we want another check box on the character creation screen?
I do think it’s a very cool way to play, but we don’t want to force it on people who don’t want to play that way. I know we’re non-committal about things a lot and some of the fans get frustrated, and I apologize for that. But our development process is super-iterative, and we’re always changing things around as we tested them out. And I don’t want to promise anything we might eventually test internally and then decide not to implement.
We’ve talked about making some kind of self-found insignia that would be on every character by default. So you’d create a character and they’d be self found until you traded or used the Auction House.
Gold sharing is another issue. Playing self found requires you to note how much gold you have to start with, so you don’t accidentally spend more than your character accumulates. We’d like to put in something that would let players show each character’s gold individually, but there’s no ETA on that.
Could we see some sort of self-found ladder?
Travis: Maybe. Again, sorry to be so non-committal, as they often are. It would be cool, and it’s come up internal discussions, but he can’t say more than that.
Nice that Wyatt replied with interest, but this desperately needed a follow up question. The entire point of a self found, Ironborn mode is that the characters are separate. It MUST work like Hardcore, to create an ecology of characters who can play together (but not trade) who are following the same rules. If you are in there growing chest hair by surviving with an Ironborn, you can’t be in a game with a character who has 10x your DPS and EHP from Auction House goods. It’s unfair, and would taint the entire experience since your “self found” drops wouldn’t really be self found, since someone else did all the work. So pity Wyatt didn’t mention that, and the streamer didn’t ask for clarification.
Click through for coverage of the hour Wyatt Cheng spent in the chat, with interesting info about upcoming itemization changes, overall skill balancing, lots of Auction House theorizing, and much more.
Wyatt: I start my day with email and browsing. Check forums, fansites, reddit, etc. Then there’s internal work and communication. There are a lot of design discussions internally, and things are forever under development, but I try to set more concrete goals to finish things, even in just small steps. Like, “By the end of the day I’ll finish one Demon Hunter skill change. Or accomplish something concrete on itemization.” It doesn’t always happen, though.
Every Monday we get the whole team together for a meeting. Everyone together, face to face. We’ll introduce new employees if someone’s been added, and synch up on what we’re all doing on different aspects of the project.
Tuesdays are “no meeting Tuesdays.” Great to have an uninterrupted block of time to get work done without distractions.
Do you do programming yourself?
Wyatt: Not really. We have gameplay programmers who have created some powerful tools we use on the game. We also work with the tech art department a great deal. That takes a larger role in skill development for Blizzard than for most other game developers, since we really want to involve the art and graphics in how the skills work. Right from the start.
What are you most excited about with Diablo 3 right now?
This is a total tease, but… the item stuff. I’m excited about that. Travis is running point on that and we have conversations about it pretty regularly. We talk to people all through the company. People on the Diablo team or not. constantly running plans and ideas past other people to get feedback and taking advantage of all the awesome designers at Blizzard.
The multiplayer improvements have consumed a lot of my time in the past months.
So this week I got this idea from my lead gameplay programmer. What he does it logs in, picks his MP level, selects “any quest / any act” and just lets the game select for him and drop him in anywhere. It’s a fun way to meet new people.
The monster density is a big change too. It allows us to join any quest and any act. And sure, some areas need more work still, and we play all over now. And it’s fun to see the builds players are using. Sure, a lot of WW barbs and Archon Wizards, but I see Demon Hunters using Rapid Fire and other stuff also.
What class needs the most work now?
Wyatt: Realistically we’ve been looking at the classes across the board. I can’t promise anything now, but maybe to summarize… wht I often see from the community is concern about the Demon Hunter DPS. Lately it’s been fashionable to say “effective DPS” since the character sheet doesn’t tell the whole story.
Demon Hunters are ranged classes who have to kite more or… you figure if you’re a ranged class you want to be able to sit there and pewpewpew and do a lot of damage. The classic video is a Monk and a Demon Hunter wearing almost identical gear and the Monk is much faster. Underlying all this is some changes we should take a hard look at, and we started doing that long ago… Okay, sorry I’m rambling. It’s hard to do this because I’m reading the comments that are going by. So let me refocus my thoughts.
So the game shipped a certain way. We have some spreadsheets we use internally between the classes. We have some general philosophies. I’m going to use some hard numbers. If I activate a skill that costs no resource or rengerates a resource, that’s going to do ballpark about 150% weapon damage. If it’s AoE you’ll do a bit less. If it generates a lot of resource it’ll do less. If it’s not much resource it’ll do more.
If it’s say Cleave, or Bolo Shot that hit a lot of targets, it’s going to be lower. And if you spend a lot of resource, say Cluster Arrow or Hammer of the Ancients, you’ll do more damage. We have some multipliers we use internally to say, “hey, a point of Fury is worth about X damage.” We can deviate from the model, but it’s nice to have a mathematical model, it’s useful. If it can hit 5 targets but it’s hard to hit them, that should do more than a skill that’s easy to hit 5 targets with.
We had this model and did all this tuning, but when the game went live we saw things that were out. Rend needed a buff early on. Hydra was only Venom Hydra, so we buffed the others. Etc. You saw a lot of that in v1.03 and early on. Seven Sided Strike was much lower back in the day, like 1000% less damage.
So when we shipped a lot of our skills were very close to our internal mathematical model. Post ship, a lot of the skill changes have been more on the realities of actually playing. So we’ll see a skill isn’t worth using compared to others, so we buff the unused ones. The buffs aren’t based on the model, they’re based on the other options. So skills are balanced against each other than against the model.
So, bringing the story forward to today. In this process there are a lot of skills that can get overlooked. Impale for the Demon Hunter for instance. It’s not doing the job. It’s a single target skill that does big damage but spends a lot of Hatred. But it’s not as good as other options.
So we’ve accumulated a lot of lessons and we need to revisit skills across the board. Apply all that we’ve learned over the past year, give attention to all of the skills that have fallen by the side.
So when you compare potential builds to the most popular ones, it seems like a lot of the most powerful builds weren’t intended. I call them “exploit builds” since they seem like they’re using things that weren’t intended. Do you guys talk about getting rid of those or makng other builds that are as powerful?
Wyatt: High level philosophy. Diablo plays best when the game is moderately challenging.. Wait let me restate. Diablo isn’t great when it’s super easy. For some people it’s not great when it’s punishingly ahrd. Some people love that, but most don’t want it that difficult. Players like it when they can put on some music and getting in and relaxing and getting into the zone that’s between frustrating challenging and too easy. We want popele in that state.
Skills and mosnters play off each other. they’re two sides of the same coin. Player power vs. monster power. We need ways to provide moderate challenge at all times, and we want players to feel awesome, but that means the monsters have to be pretty tough at times.
Do you want to buff other builds to the level of the best, or reducing the power of the really big ones like WW and Crit Mass?
Wyatt: Don’t know yet. Take the zero cooldown Zombie Dogs build. I like that some legendary items are involved in making that. Say Mara’s Kaleidoscope. Not always in that build, but often. That’s something I think is underutilized in Diablo III now. We’d really like legendaries to not just be “stat sticks” but to open up new build possibilities.
There’s room for improvement. One definite way to improve is a legendary that opens up a new build.
Travis hinted at some. So if you could have like two Hydras up. That’s cool. A wizard could make a whole build around that. Also… four pieces of the Inna’s set makes Sweeping Wind very cheap. That’s a very pale shadow of what it should be. But it’s something we’re playing around with. That’s more like a “quality of life” improvement than a huge change, but we want to take that kind of idea and dial it up to 11.
We often talk about the classes we’d like to see in the expansion. It’s fun to speculate on that.
Wyatt: *laughing* Obviously I can’t talk about that.
But we were looking at which D2 class would make the best transition. The Druid since there’s no shapeshifter yet. You can’t talk about specifics, but can you give us some of your thinking. As you design classes for the future, can you talk about how the process goes.
Wyatt: I can’t say anything specific. From a dev standpoint we want to do something our artists are excited about. That’s a big selling point. One thing that makes Diablo special is a lot of the aesthetics of the game. It’s very satisfying to slay demons. If you line up the characters, they look good. Anything new has to fit that lineup and excite the artists.
I’d also look at playstyle. See… I can’t say much. But we’d definitely want to make a character that doesn’t overlap any of the current play styles very much. Like, if I can already get a similar gameplay experience out of the current classes, that’s not very exciting. We seek the overlap of good looks and play, good mechanical gameplay, differences from current classes, etc.
Endless dungeons? Or other content that’s separate from the other content in the game? Can you talk about your philosophy of the end game content?
Wyatt: My very first question about endless dungeons… what motivates people to want it? What is the root desire and motivation of players, so that we can best address it. There are a lot of different ways we can meet the wants and desires of players.
I get different answers. Some players say they really want to test their character, and an endless dungeon that’s getting harder and harder would do that. Other players want to play for hours on end without having to make a new game. Players have an intuitive sense that they want their gameplay to vary. In D2 people would do runs. They were very rewarding, but it was the same and lacked variety. The appeal of the endless dungeon for some is the game presenting different content and art and styles of play.
Just to wrap up on the endless dungeon… some players definitely want it. We’ve talked about doing… something. To fulfill those different player desires.
You are free to join me in screaming at the lack follow up questions. I don’t know anything about this streamer. Does he just never talk nuts and bolts of game design or mechanics? He seems disinterested or incapable of engaging with Wyatt on any specific points or theories, and the missed opportunities are driving me crazy.
So how did increasing monster density satisfy any of those player desires Wyatt mentioned? More monsters in Act 1 and 2 doesn’t test your character with greater challenge, and it doesn’t let you play for hours without creating a new game. As I’ve argued on the podcast several times, the main benefits of an endless (or at least deeper than 2 or 3 floors) dungeon is in changing up the gameplay. New level layouts and designs, new combinations of monsters, occasional higher densities of bosses, champion packs of 6 or 8 individuals, timed dungeons, etc. All sorts of stuff could easily be pumped into a bonus dungeon that would greatly vary the play experience in ways that can’t be done just by adding a few more trash mobs to Acts 1 and 2.
On with the transcript:
Are you putting in features lately to make players rely on the Auction House less? A player asked about a phone ap for the AH, and part of the reason you guys said you weren’t planning that was that it would increase reliance on the Auction House.
Wyatt: When we talk about the AH… it serves a need in Diablo. In a game with random loot, there’s definitely a difficulty for players trying to obtain a particular item or item with particular stats. That’s what you have to trade for, and in Diablo 2 that trading experience was pretty poor, with spammy trade channels or having to use third party sites. We wanted to provide a better experience in Diablo 3, and that was the main goal of the Auction House.
That said, there’s a balance to strike, and I don’t think it’s ideal when players have all thirteen item slots filled by pieces they purchased in the AH. Those v1.07 crafting recipes were a step in that direction. So after that patch, at least a few of the items, maybe all 5, or at least two or three, were things players could feel lke they’d made themselves.
There’s a debate to have over how many of your character’s items should be from the Auction House. Some players who are going self-found obviously think that number is zero. Other players who use the AH constantly feel it’s 13. I feel like that number was too high at launch. Just my gut feeling, I think that maybe about 3 of a character’s items should be from the Auction House, while I’ve found or made the others myself.
Ladder resets. I’d like to see a ladder added with seasons. but I can see why it’s not that simple.
Wyatt: There are 2 aspects of the ladder that are worth talking about.
1) The competitive aspect. The race to the top. 2) The other is the economy reset. Sort of this fresh economy where everyone starts off clean. That’s really fun. It’s awesome to level up new characters. I like to see my character growing every more powerful, but the frequency of upgrades is much less.
So when I play a new character up to level 60, I’m finding an upgrade to some item every 15 minutes. New gloves or ring or whatever. As yo continue to level, then your upgrades come more slowly. Maybe every 45 minutes. And then longer as your gear goes up. When players use the AH they can hurl themselves forward in the gear. And then basically never find an upgrade.
So a ladder reset is fun in that way, since you’re then on a new clean character and finding new items all the time. And it’s fun to do it globally as a community, so everyone is in the same pool.
So is that something you like and we might see it?
Wyatt: Um… Me personally, and this is something others on the team disagree with, but I’m more interested in the clean economy, rather than the race to the top. I like that players get really into it and stock up on caffeine, but it’s not my thing.
I’d like to insert a side comment on white items. I don’t like white items in the game as they are now. I know it’s a popular request from players to simply hide their display, and perhaps hide blue items as well. I can understand that request, but it’s not something we want to do. We’d rather get to the root of the problem.
White items either need to not be there, or they need to be useful and interesting in some way. People are asking about runewords. Those could make white items awesome. I don’t want to pick up all of the white items, but I’d like to know… this is something D2 did pretty well, but it made you excited to find one very specific white item. So you’d leave 90% on the ground, but once in a while you’d find one and be very excited for it.
One of the litmus tests we’re doing internally: “Can I picture this item being useful for someone?” Say, a quiver with strength. That’s really dumb. No one is going to use that. So we’re trying to address items you can’t imagine anyone using. Versus items that maybe aren’t for your character, but could be useful for someone else. Say a belt with Thorns. You might not want it, but some day when we improve thorns, then there’s someone who would really want it.
We want white items to be kind of the same way. You might not want it, but you’ll know there’s a purpose and scheme for it in the economy on the whole.
Can you talk about how you’d incentivize players to use gear the find themselves?
Wyatt: We’ve seen people suggesting say, a way to enhance an item you found yourself to be 25% better. Or to add a socket ot an item. I can’t comment on anything specific, but it’s something we do discuss.
Many players assume we’ll see item upgrades, perhaps via the return of the Mystic, in the Expansion, which is perhaps why Wyatt is so vague about it. (And he and other devs have ducked the question previously.) Since it’s far off and will be part of their big Expansion news when they do break it.[/blue]
The host had no questions left, so Wyatt started taking some from the stream chat:
Players ask why they can’t get items with only good modifiers. Why can’t all items just have the six perfect stats. I try to turn that question around and ask why are there just six perfect stats? We want to put more play style and options into the game.
Wyatt: Let me comment on One With Everything. Monk passive. I think the most likely — and I’m definitely not committing to this as a final plan — the most likely solution is to try and get a new set of resistance packages into the game in the future. That’s pretty far down the line probably. It’s part of the itemization changes.
Those will make the individual resistances roll higher than All Resistance can roll. People think it’s weird that resistance to everything is a bigger number than individual resistances, but let me explain how we got here. History lesson.
When single resist was way higher during testing, you were incentivized to keep a bunch of varied resistance gear in your inventory and swap it in for specific circumstances. So when you saw a Molten or Plagued or something, you’d change to your item with really big resistance to one modifier. So to change that we made All Resistance with the highest value, and that was kind of better than gear swapping, but I think ultimately we just traded one problem for another.
When the higher single res values roll out with the itemization changes, you will no longer see single resistances on the same item with all res. And OWE will still be a big benefit, but it won’t be as huge as it can be now. So to use some numbers, say that All Res goes up to 80 on an item, in the future single res will go up to 100 or 115. But you’ll never get both on the same item.
Wyatt: Open world mode. No quest, no cinematics, no other interruptions. We’ve given that serious consideration, and figured the pros and cons and how it would affect other aspects of the game. I’d love it if some areas had different monster distribution. D2 did that well. But the open world thing might make that not work for technical reasons.
Two billion gold cap in the Auction House.
Wyatt: It would be nice if the gold cap were raised. I can’t say beyond that. It’s not really a design issue. Everything related to the AH affects more than just design.
Wyatt: I’m definitely not bored with the game. Just got my Monk a new upgraded Echoing Fury with just 5% chance to fear, and that made me happy. I was playing with a friend last night, a guy from the WoW team. He plays an Archon Wizard. He was bringing the pewpew and I was tanking and going with Cyclone Strike and it was great.
Wyatt: Will you ever be able to waypoint across the different acts. That would be very cool. Can’t say more than that.
Wyatt left off at about 60 minutes into the stream, and promised that Travis Day would step in next, and that he might pop back on later in the day.
If you guys are wondering, for months and months we’ve been making requests through the Bliz PR channels to take part in interviews with the devs. Either for print, or as a guest on the podcast. That’s yet to happen, though it might yet. Bliz PR never says no, they just give us something like, “Oh that would be great. We’ll look into that and see if the scheduling can work…”
And yeah, obviously that’s an excuse, as the devs seem to have plenty of time to AMA on Reddit, or reply to random posts in the B.net forums, or hang out chatting on live streams. And yeah, it’s frustrating to listen to something like this when it’s so rambling and hosted by a guy who doesn’t seem to have many questions or much interest in D3 dev issues or core game debates.
But there was some info, and it’s good to hear Wyatt’s thoughts, and it’s better than nothing, eh?