Donate to Codex
Putting the 'role' back in role-playing games since 2002.
Donate to Codex
Good Old Games

RPG Codex Interview: Josh Sawyer at GDC Europe 2016

Click here and disable ads!

RPG Codex Interview: Josh Sawyer at GDC Europe 2016

Codex Interview - posted by Infinitron on Wed 28 December 2016, 10:03:16

Tags: Josh Sawyer; Obsidian Entertainment; Pillars of Eternity; Pillars of Eternity: The White March

Back in August, Obsidian Design Director and Codex anti-hero Josh Sawyer gave a talk at the annual Game Developers Conference in Cologne, Germany. The video of the talk, which was a Pillars of Eternity retrospective entitled Looking Back and Moving Forward with Pillars of Eternity, has yet to be publicly released, but the original slides and a summary are available. But that's not what this post is about. Former Codex contributor Bubbles was in the audience for that talk, and after it was done he met up with Josh for what was meant to be a brief interview. They ended up talking for nearly two and a half hours.

A couple of days after that Bubbles was off to Gamesom, and unfortunately he never got around to transcribing his recording of the interview before his unfortunate departure. We did however manage to secure Bubbles' permission to allow the Codex's #1 Josh Sawyer fan, Roguey, to transcribe the interview in his stead. We sent the recording to Roguey, expecting a long wait...and were provided with a near-perfect transcript less than 72 hours later. Probably should have done that sooner. So there you have it, four months late and just in time to be the last piece of Codex content for 2016. Merry Christmas, Happy New Year, and enjoy the read. It's a long one.


Read the full article: RPG Codex Interview: Josh Sawyer at GDC Europe 2016

[Interview by Bubbles]


Starting with Pillars of Eternity, a lot of people have asked about the inspiration from Planescape: Torment, what that was; you've mentioned the UI for instance?

All the descriptive text was part of it. That was a focus for Eric Fenstermaker. He was focusing more on using descriptive prose mixing with the dialogue. And I think also focusing more on the more philosophical themes: life and death, things like that, that weren't necessarily a big focus for Icewind Dale. There were more interesting thematic elements in Baldur's Gate and Baldur's Gate II, so I think the elements we wanted to take from Torment was mostly the narrative and descriptive text, a lot of the character reactivity, stats, and things like that. And then just more of an exploration of mature themes.

Of course that's a high goal to aim for. I remember reading a review from a professional website at launch that claimed there was a Torment-style twist in the game that transformed it into something like Planescape: Torment. I kept playing it, not knowing what it was, until I realized it was the soul reading.

Yeah, I wouldn't say that it was necessarily that. That's not as big as a twist as Torment was.

I've heard you mention the zoom correction by 7.5 degrees because people were very anxious to have that addressed again, and they were wondering if you were planning to change it.

About the angle?

Yes. People seem to really want something more top-down than what you have.

It's something that we talk about a little bit again at the beginning of development. For interiors, I guess we could do more of it, meaning raising it a bit more. The problem there usually is because our dungeon environments tend to be more cramped, that's where you want a higher angle. With exteriors, personally I don't think it's that big of a deal and I haven't seen that many problems with exteriors because the environments are more open and because we have more vertical things to show. It's not something we've been fiercely debating, let me put it that way, but I do know that every once in a while we look at the camera angles and sort of question like "Could this go up, could this go down?" It's something we discuss, but it's not something we've wrestled with or anything.

I don't know what specific elements of the angle people have a problem with, but for me I started replaying the game in anticipation of this grand lecture, and even though I have over 300 hours logged in, I kept misjudging the distance between people and walls in dungeons and it's hard to tell where the walls start when it's at the bottom of the screen.

Sometimes it's the art, sometimes it's occlusion, like how the occlusion system works just to make it more apparent where a character is in the environment. Our occlusion system shows where the characters are, but it doesn't ground them very well. I think there are ways we can improve both how our characters render in the environment to make it clearer where they are, and again, part of it is that we shouldn't design maps to be that cramped where you wind up needing a super-high angle just to see what the hell is going on. So level design is part of it as well. But if anything were to change, the interior camera angle is more likely to go up, but again, it's not something we've really talked about.

What about the degree of clarity in combat?

It's something that has a lot more to do with how the effects were rendered, and it's something that we went back and tried to fix later in development, and... it was hard. But we have fixed it. There are a couple of things: one is just being less crazy with effects overall, not being so crazy with that stuff. The second part of it is that we had to look at how the effects layer when they get stacked on top of each other. In the Infinity Engine games, those were all sprites and they were opaque and when they layered on top of each other it was fine. In our game, there was a multiplicative effect and you get things blown out, and once things are blown out it's like "Okay, I can't see what the hell's going on." We recognized that was a problem and we render those effects in a different way now and they're much easier to read. Obviously it's an ongoing process because half the problem is being judicious about how big those effects are, how long they last, but the rendering I think will help a huge amount.

That's hugely encouraging to hear. Another thing I noticed was that sometimes you can't tell what an effect is actually doing. If you have envenomed strike, it applies to AOE effects as well. Stuff like that is another source of confusion that might be difficult to alleviate.

Yeah, honestly, some of those logical gates that things go through aren't all that obvious. Once you see it in action and you go "Oh okay, now I see these things going together and they're not supposed to." I think you're right though, those things are harder to catch. I think our tools are a little bit clearer about how a lot of those effects work together and how the status effects are organized now, so on the authoring end we can catch that more easily, but again, sometimes you'll wind up with weird things where yep, that logically makes sense, but of course that's not the effect we intend.


Do you think the game was in some ways too complex in terms of rules?

I don't know. I don't think it's fundamentally that complex. I think that the graze/hit/miss/crit system is a little bit more complex than the basic hit mechanic, but there's also a lot fewer wacky things in classes, like we don't have equipment restrictions or things like that. In that sense, it's a lot simpler than D&D second edition. Especially when you get multiclasses and stuff in there and you have to figure out what's allowed.

We don't really have gear restrictions based on class other than soulbound weapons. A lot of people liked finding the Holy Avenger, and they like that only a Paladin can use it, but then obviously if you didn't make that one character then you can't use it at all. So the soulbound weapons were a way to sort of like "Okay, this hits a few classes" so it's likely your party has someone who can bind it. Even if you can't, you can still use it, or sell it, whatever.

I don't really think the system is that complex, but I do think our UI does not give great feedback about a lot of the things, and a lot of things give inconsistent feedback. Something we focused on a lot recently is cleaning up how status effects are organized, cleaning up how the UI displays that information to make it more clear. I do think our stacking rules could be simpler and cleaner. We do have to document better, and once it's documented, it gives you a baseline for comparison to say "Oh well the rules say that it should work this way, but it's not." Whereas, for example, when I played Baldur's Gate for the first time, I had already played second edition for years and years, so when I saw something I could say "That was definitely a bug" right away because we had the rules to back it up.

You said you wanted a more realistic, less saturated color palette, however, others wanted more light.

Yeah, some of our earlier levels are a little too drab. Our later stuff, especially some of our areas in the White March, had a lot more variety and intensity to them. They don't go to that super crazy colorful saturated land. I think it was a learning experience for us, and I think a lot of the expansion content got more to the place that we felt comfortable with.

Some people believe the White March is a blueprint for how the series is evolving.

Yeah, any lessons that we learned, sometimes we can retroactively apply them, but sometimes it's like that content is done and there's only so much you can do. As we develop expansions and patches, we are trying to address concerns in the base game or use the lessons we learned for new content. So for example, the Siege of Cragholdt, we were like "Okay we're developing new AI." I mean it's not great still, but it's better having all those enemies within Cragholdt use a much higher level of AI than most enemies in the base game. So anytime that we have new improvements that we are trying to apply in something like The White March where it's an expansion, even if people absolutely despise it, which thankfully it doesn't seem like they do, then we can say oh that went over pretty well, let's keep going in that direction.

Is there something you would say you've done better in the base game than The White March?

I think some of the quests in The White March I are bit more convoluted than we wanted them to be. Not convoluted in a good way, but confusing. I think also because we didn't have a level scaling system built in, the difficulty spike or drop is very disconcerting to a lot of people. That's not really a problem with the content developed for the White March, but sometimes it's hard to gauge. Oh, I do think that we went overboard with a lot of encounters. I think the difficulty was fine, especially in places like Galvino's Workshop and Durgan's Battery, but you wind up with way too many fights. Fights are cool, but if there's not a lot of variety between them, then they just get boring.

There's a lot of speculation on the forums about why there are so many mobs in the Skaen Temple. Some people believe they're there to punish people for taking the wrong entrance. Is that a correct assessment?

No, it wasn't intended to punish at all. It was the first dungeon we developed, and much like Dyrford Crossing, it didn't get a lot of attention later on because there were so many other things to develop. I do think it was an area that I had slated to have a reduction in a patch, and I don't think we actually got around to it. But no, we looked at it and we were like "Ah jeez, this is just way too dense," like enemies on every screen and usually the same. Some of those guys have distinctive behaviors, but not so much the whole dungeon. Yeah, we should have thinned it out, it was absolutely not intended to be punishing.

Your general approach to encounter design: do you see it as something that paces out the game to lighten up the dialogues or do you think it's something that should be constrained to a few quality encounters, that is, as few as there needs to be or as few as you could make very well?

I think combat accomplishes a few purposes, and depending on the sort of route we go with it, determines how we progress our encounter design. The best fights are the ones that stand alone, like very well-tailored encounters, whether they're boss fights or just a very well crafted simple encounter. I think that having smaller encounters along the way is good for both pacing of exploration and also the strategic sort of reserve of managing your party over time.

There's a lot of contentious discussion around resting, resources, things like that. You probably can't make everyone happy; there are some people that would like fights to all be oriented around encounters and those people would probably like for each fight to just be a a fight balanced around the encounter. That is a very different feeling from having a system that's intended to wear down your resources over time, where you have fights of small-size/medium-size/big-size, and the player's encouraged to meter out per-rest resources, using them sparingly when it's really important rather than go crazy, and it's more a strategic element there. I think if the game has a more strategic per-rest sort of orientation, then you need the smaller fights to help pace that stuff out and help the player make these sort of judicious calls about "Do I really want to use a per-rest now or should I hold off?"

Overall, I do think that having set-piece encounters is really what we'd like to focus on. As we went through and edited a lot of the old encounters, we said "get rid of these encounters" and then look at the encounters that were lame and try to make them more interesting. Because obviously if it gets repetitive, it's not going to be that fun.


What's your current view on how the per-rest system should be handled to make clear to every type of player that resources should be parceled out instead of being freely available?

We've looked at a lot of ways to handle it. I think focusing less on powers as inherently per-rest versus per-encounter will probably be a good thing. Some classes feel like they're entirely oriented around per-rest, like Wizards for example.

But they have pretty good per-encounter too.

Yeah, they have good per-encounter too. I guess one of the other things that comes up with characters that have a split sort of pool is that most players will not take per-rest abilities.

Yeah, if you're built around per-encounter, I think per-rest feels less valuable.

It sort of depends, because for example, we had the Barbarian Dragon Leap set up as per-rest and if you can use something twice per rest versus once per encounter, well twice per rest also potentially means twice per encounter, so you can go potentially very crazy with that. But people tend to veer away from that, so I think something that tries to make characters less oriented around purely being just a per-encounter character or just a per-rest character class inherently is probably the best way to go. I do think that losing the idea of the long term strategic wear-down would make the game feel less D&D-ish.

That's one of the reasons why we included mechanical bonuses in the inns, to make it feel you were encouraged to go to an inn. It feels appropriate in a D&D game to go rest in an inn, so let's make mechanical bonuses that are associated with that. And resting is a similar thing, it feels appropriate in this style of game to do this. Of course, on the other hand, you have something like Torchbearer or Darkest Dungeon which is very specifically designed to grind your characters down, so it's difficult in a game like this where there's open exploration, you can backtrack whenever you want, it's hard to find something that feels right and it's mechanically fulfilling, I guess that's the problem. That's an ongoing challenge, something that we're experimenting with. It is important to us and people have very strong opinions on it, so we're going to keep looking at it.

Some people were wondering specifically about, with regard to The White March and the Endless Paths, the way the level of progression and level scaling are implemented. Do you have a vision for a game where you can offer this kind of breadth of options starting from the beginning of chapter one without having to forfeit on level scaling?

We had to sort of hand scale everything, which is pretty awkward. In the future, I think that having a more systemic approach for it would be a lot healthier and easier. I do still strongly believe this should be something that people opt into or out of. I think a lot of people want to just play it at the level it's made for, and that's fine and good. I think giving people options to either not see the information is also fine. For example, some people will walk into Longwatch Falls or they walk into Cragholdt, which is funny because the steward will tell you you're too low-level and you're like "Okay, I know what you mean," and you go in there and get killed. If you opt into "show me what level this area is for," if you don't want to see that, you don't have to see it, and if you do and you don't want scaling, it's fine. And also if you do want scaling, it's a systemic scaling as opposed to the half-measure we did with The White March.

There's people on the Codex lamenting this perceived idea that there's not enough creativity that can be expressed with a system that's either level-scaled or very easy with nothing in-between, like a sequence where you have to use your wits or some specific skills, like say a low-level Survival check to bypass a high-level fight.

We do want to make skills and skill checks. People liked scripted interactions and one of the things we often talk about in terms of area design we're trying to reinforce more often is the ability to bypass or shortcut things either through using stealth or other skills. It's something we tried to reinforce and we're going to continue to try to reinforce it. We very frequently look back at Raedric's and say "this was well-done." We feel good about it, players seem to like it, it gives you a lot of options, both for routes for approach and the ability to use lots of skills, disguise yourself, and so on. That definitely, for us, is one of our best dungeons.

I'd like to mention the ogre at Dyrford. What I liked about it is that you don't immediately know how to placate him. Some people just click the skill check and expect that to be a good choice, but it wasn't the case here.

I do believe we should communicate through dialogue "this is the personality of this character, this person gets irritated if you do this sort of thing." It's like okay, you can ignore that and pick the thing that shows up that's like "I'm going to make a witty comment," and he's like "No, I don't like witty comments." So if you reward people for reading in a dialogue game that's cool, and I think we want to avoid anything where it feels like it's purely a guessing game where you're like "I have no idea, just let me pick something."

I noticed in chapter one there's much less of that than when you get to Defiance Bay and Dyrford. The Gilded Vale area wasn't impressive in that regard. What I saw in Dyrford was more impressive even though it was earlier-designed.

It was earlier designed, but I think dialogue got completely rewritten a few times. So that's probably why it wound up better.


I was surprised to see that you need to have Survival at level 9 to pass the check to keep the ogre from stealing pigs every time he's hungry. I liked that, because it's an area where you wouldn't expect to have Survival at level 9.

Very often that's the way that we look at skill checks. Although I do try to reinforce the idea that it's okay have low odd checks sometimes. The reason for that, and this is something we did in Fallout New Vegas, is it is okay in the early game to have hard checks to come back to, like for example Lucky the revolver, but it's also good in the late game to have low checks so if you just started investing in a skill it doesn't feel like "Well, I'm not even going to be able to take advantage of this until another three levels."

When you decide you're going to lower Resolve and you think about what Resolve impacts, once you go below level 10 you only have to worry about combat performance because there's no social check that says Resolve 6 required. Is that something you want to change or something you'd rather keep as it is?

We can look at our thresholds for attributes more, but I think that it's pretty rare that we puts conditional checks on things that are low, in part because it would probably flood the UI with checks constantly. So it's unlikely we would change that or we'd have to look at how the interface works. It's one of the reasons why dumb dialogue can be hard to do, because depending on the tone of the dumb dialogue, it necessitates several different levels of checks that can kind of clutter the dialogue window.

Do you see the possibility of having a skill check that's only available at low levels if you want to get yourself into trouble deliberately?

Yeah, I do think that's fine. That's one of the things we did in New Vegas a lot, that is have the failed skill checks. The rule is, they don't put you in a worse position than you're already in, but they're entertaining and they recognize that you have those low stats. Like if someone comes up to you and they're going to attack and kill you, if you say something and it doesn't work, well they're already going to kill you so you're not worse off for it. So having that stuff is nice, but it is a lot of work, so we have to be careful about how much of that stuff we do.

Will you implement more noncombat content in the next game?

I think a broader set of skills with more noncombat applications overall. I can do a whole talk on skills in a party-based game. It's a challenge because it's very easy to have skills that feel redundant, it's very easy to have skills that have a personal benefit but no party benefit or party benefit but other people can't contribute to it, so it's hard to come up with a spread of skills that feel like you can invest in them all the same way with the same pool of skill points. Obviously it doesn't need to be perfect, but you're not just doing something stupid like hey I have three lock picking characters. Great, that's a huge waste of skill points. If you have three characters that have lore, then everyone can read scrolls and they all benefit from that. I always have three characters in my Pillars parties who can read scrolls.

One thing I don't feel good about in our skill spread is that because there's a small number of skills, it's hard for those skills to make each character feel distinctive. For example, there's just lore, that's everything. I think having a larger skill set, as long as you can actually support it in the game, it allows you to feel like my character doesn't just know about lore, he knows about metaphysics or she knows about arcana. In D&D you can clearly delineate that I'm the expert in this thing, so it's a role playing thing. As long as you can make the game content use that in a way that feels satisfying and not frustrating like a waste of skill points, then that's cool. I'm always looking for ways to improve, because I recognize that people weren't too happy with the spread of skills that we had in Pillars.

What would your ideal combat experience be in contrast to what Pillars of Eternity has right now?

I think pathing can always be really improved. Pathing and also engagement and how engagement works. I think engagement can be useful, but it needs to be looked at again to make it clearer, to make it feel more useful, easier to understand, what the hell's going on with it.

You want to retain it?

Yeah, I don't think it needs to be exactly what it was in Pillars, but I think that having a character who has the ability to sort of like pin people down, it's useful. So that general concept is still worthwhile, but I'm not married to the exact execution of it. I think a lot of the ways we have modal abilities work were often confusing. They're often buggy, which also doesn't help. So revising those so that it seems clearer when to use them, when to not.

I think our damage threshold system actually wound up being, I guess in some ways, a bit simple to understand, but in execution when you actually break down what you should be doing to an enemy at a given point in time, there's a lot of math that you kind of have to do on the fly and yeah, the game can do it for you, but not really. You still have to actually think about what's the most efficient way for me to do these things.

So looking at how those systems work, but on the content side of things I think it comes down to having system design that's more involved with area design and setting up encounters. Tim did a little system design at the beginning with Pillars, but I was the system designer for the rest, and I was also directing the game and a bunch of other stuff, so I didn't have a lot of time to sit down with the area designers and really work on crafting encounters. And I think that the way a system works is not just on the system, not just on content, they have to work together. So crafting encounters to be a more interesting composition, how the enemies are positioned, tactics that they use, that should be more of a conversation that goes back and forth between system and area, and I think having more set-piece encounters varied and with more depth to them.

I still think an overall reduction in encounters would be perfect. The game is long enough, there's plenty of exploration, and I don't think people get bored going two screens without having a fight. Just pacing things out a little bit better. I don't want it to feel empty. Sometimes as a designer, when you look at a screen, and you realize like "Wow, there's nothing there, oh I gotta put something there," but it's okay. Or maybe that thing that you put there is like a container, or it's something to harvest, or just a little thing. The important thing is that it changes; it can't just be a fight, a screen, and a fight.


I think I've noticed a rhythm for when things might be hidden somewhere because there's interesting scenery around.

Yeah, that's something Bobby Null encouraged, to his credit. He really tried to get people, when they place hidden items especially, to like don't just put it anywhere, look for something that's interesting and make the hidden container there because then that draws that reaction where you start to pick up like "Huh, that looks a little like it's off, and oh, I found it." Then you feel smart and get a little prize.

We also had our interview with Eric Fenstermaker a few months ago, and he had an interesting comment which we would also like for you to remark on as well. He said he would prefer to make somewhat shorter games focused on choices and consequences, replayability, and branching storylines. Do you agree with him?

I think overall, yeah. I think for Pillars, I don't think if we made a game that was a third as long as Pillars, that people would really dig that. A game, sure. If we made a Pillars sequel that long, I don't think people would be so into that, but I do think we could make a game that is shorter than Pillars, certainly, with more emphasis on replayability and branching.

Personally I think the game could stand to be a bit shorter.

I think it can be a bit shorter, but the question is should it be 25% shorter? 50%? If 50%, then you're done in the middle of act two, which might be too short for a lot of people.

You could adjust the length by removing trash mobs.

The thing is though, the focus of the narrative department is still going to be roughly the same. By removing trash mobs, you haven't done anything to reduce the amount of writing. When the total number of characters goes down, it does give the writers more time to focus on individual characters and reactivity. So in general I would say I think we can make shorter games. I think the challenge is: there are going to be people on your site who will expect our game to be as long as BG2, and that's a crazy length. Almost no games are that long, it's like 200 hours. It's really hard to make a game that big and also have a lot of choice and consequence and reactivity built into a lot of elements of it, because with more content overall, you have more characters, you have less time to spend on each branch of that character.

That being said, I think we can be much smarter about reactivity. I think we were much smarter with reactivity in Fallout New Vegas, and it's something I want to emphasize for all our games in the future. I think a lot of designers feel that when a story branches, that it has to have this strong hard branch where the whole story changes, and that's not really true. And players don't necessarily respond that well to that. The Witcher does a good job of this, I think we did a pretty good job in New Vegas where you have reactivity, but it's sort of limited in scope and scale and so a few hours after doing a quest you see someone in another location and you see the effects of what you did. And if you do that a bunch over the game, then yes, you have created this big network of cool changes that affect the game, but it's not the same as saying if in act 1 you make this choice then the whole story changes in this way, which is hard and difficult to author, and it doesn't really give you the replayability you think it does.

If people develop a consensus about this, would you be open to implementing more reactivity in the sequel?

Oh yeah, I think we always want to have more reactivity. We're always trying to get more reactivity, the problem usually becomes how we try to act on it, how we try to author it. Like I said, there are easy ways and difficult ways to author. And sometimes it's little things that go a long way. For example, there's a quest we have in the second part of The White March where you have to go deal with a Bleak Walker Paladin. And it's a perfect opportunity to just react to the Paladin's order, like if there's a Paladin in the party who reacts to the Paladin's order, places like that with obvious call-outs where a certain class can shine. It doesn't even change the quest that much, I mean it can go in a couple of different little directions, but it feels cool that if you're a Goldpact Paladin or if you're a Kind Wayfarer, there's just a different line for you to use and that's good. And they sass off to you. Like the Bleak Walker says different things to you based on that.

That's something in Icewind Dale II that I did that not many people saw, which is kind of fun. There's a death knight somewhere and you can confront him and ask him "Do you really need to do this?" and he says "It's my duty" and I can't remember if it's a Paladin or a Priest who can salute him and he'll salute you back. It doesn't change how the fight goes or anything like that but it feels cool like "oh yeah, I follow the guy out of duty, and I recognize this guy's bound to do this thing," so it's a little thing and a lot of little things add up to make the experience feel a lot different. You don't have to change the course of the entire river just to make the player feel like their class is being responded to.

Do you think it's realistic you can do more, like have a fight be skippable by a certain class?

Yeah, I think that's possible. Again, I think it's really about trying to encourage the designers to look for opportunities to use that stuff. And it's actually something we've talked about, like when you've built a quest, "What character builds are you trying to reward and make feel cool in this quest?" You're not going to feel the coolest in every single quest, and if you did, that doesn't even make any sense. You can say this is a quest, for example, where if you use stealth there's a very well-defined path to get through it, so if you have a stealth oriented character you can feel cool doing it. If you play a Wizard, you're going to get a chance to do arcane stuff and the quest is going to respond to you in a certain way and you'll feel cool doing it. So I think for us really, it's less about the will to do it and more about encouraging the designers to really sort of emphasize. Like, we don't want to have something where it feels like if you're not this class you can't do it at all. But if you're this class, it can be a lot easier or it's done in a different way and feels really nice.


Sometimes it's done well, but other times it's devoid of reactivity, like playing a Priest of Eothas from the start. It feels like a wasted opportunity.

I think with that it's down to being more attentive to circumstances where it feels like it begs for it. Like, we do a good big picture view where we can see where we react to all these people, but sometimes you miss an obvious thing, like you're in a town that has a temple of Eothas that was destroyed and so even if a Priest of Eothas winds up getting a few more lines than other Priests, it's okay, because like you said, if you play them then you feel like "Why is this not paying any heed to the fact that I played this character?"

I think the way it works currently, you don't actually notice as a player whether reactivity is in the game. Do you have any plans for improving that?

Ideally, and this never works out, you get a pretty good spread of reactivity for all the different races and classes. Part of that is due to how we track things. But I think, again, looking for more opportunities for encouraging designers. If you make a quest, think of all the different classes and races. Like where are all the places, and again, it doesn't have to be a lot, it should be a few times, especially when it feels the most appropriate. And if you can do that, the rest of it, sometimes it feels harder to work with certain character types reactivity-wise, but that's our challenge. If we use our tools we can see these classes or these races are not being responded to at all, and we can go back and try to address that. Sometimes it's hard.

It seems like with multi-classing it could spiral out of control.

It actually should be easier because you'll have more opportunities to react. I think the way we handled it in Icewind Dale II was if it reacted to a class it wouldn't matter how many classes you had. If you have a Paladin then great, you can respond to it, it doesn't matter if you have two other classes as well.

Have you looked into NPCs unlocking dialogue options for you like in Storm of Zehir?

One of the things we're looking at for skills is allowing for people to contribute more directly to skill checks, so if you have a character who has skills that could apply to dialogue, they can apply to those additively. However, Storm of Zehir is not built around the idea of a party leader, you just have a party of people, it's not really about one person going through the story. It's more difficult to have people unlock things for me; it just seems a little jarring. I'm the Watcher, I'm the focus of the conversation, but then Aloth is unlocking this line that I'm then speaking, or is Aloth speaking now? And that's where it feels a little bit weirder, so things where it feels like they're contributing in terms of skills, that feels a little more natural, like yeah, we're all intimidating this guy, or we're all really diplomatic. But if the idea is that your character is still the one speaking the line, you're still the focus of the conversation.

One of the few things Dragon Age II did well was give you the opportunity to let other party members speak up.

I think that when it's a specifically scripted example of a companion, that works very well. And we actually did that maybe only once? In White March part two there's a little girl you can talk to and she seems very frustrating and kind of a brat, and you can say "Sagani?" because Sagani's a mother and she has five kids, you can say "Sagani, can you just deal with this kid?" and she just gives the girl a stern talking to and the girl's like "Okay." But in that case it's specifically tailored to Sagani the character, it's not based on her stats like say "mom of five kids." Things like that work very, very well. I think where it feels like a little more of a mechanical generic line, that's where it gets weird.

I'm wondering if you're more interested in emphasizing the companions when it comes to things like reactivity or more about the character player's choices and fleshing out his identity through his actions?

I do think that when it comes to things that are really decisive in conversation, that should be more focused on the player. But I always think that there are fantastic opportunities to build on character themes and the world through the way that companions interact and interject. And I think that also we can do more with our companion banters just as you're walking around. Like ours were not super in-depth, so having them develop over time and feel like they really evolve with the characters over the course of the story is important and good. But when it comes to the conversations themselves, I think that in terms of dictating how conversations move forward, it typically should be focused more on the player character.

I thought it was a really good idea to give each of the companions a personal space in the Dragon Age games, and I don't know how hard it is to implement or even if there's any interest in it.

I don't think it's a question of difficulty, but I do think it's one of those things where we have to look at when and how they should come in. Doing more things like what we do with Sagani in that one little moment, I guess the thing that feels really good about that, is that they have their own spaces in character and it's not about the mechanics. It's about the fact that Sagani is a mom. And so the way they help you in conversations is tailored to them as a very specific person, not anything mechanically about them.


You generally don't want companions to act on their own?

In conversations, generally not. They certainly react to each other, but they don't do what they did in Baldur's Gate where they just go crazy on each other and initiate combat. But we do want them to feel like they develop over time and they have their own arc of development that's tied to their personal quest. I don't think that we ever did anything in the story where they would just leap into a conversation. And like I said, there were a few places where they would respond, but then NPCs generally wouldn't respond to what they said. And part of that is due to time, since if they responded to everything a NPC said it'd get very difficult, but having more reactivity to that, I don't think is a bad thing. I still don't think, in most cases, having companions just jump in and say things. As a rule, we probably want to avoid that. But there may be exceptional circumstances where a character would just burst out and say "I'm crazy."

So you're not ruling that out?

No, I think there's always room for characters. Like the thing with Sagani, there was nothing that ever said "Don't ever allow a companion to do something for the player," that's just not something that we did as a general rule. But it felt very appropriate there, so we did it. If we can expand how companions are involved in conversations, it's not mechanically hard, it's mostly just trying to find opportunities for it.

You want to maintain the player's sense that he's in control basically.


There are no outside forces forcing him to do anything in particular.

In general I would say that's true. And that's more that they would call on companions to help them than companions would just jump in front of them to do something.

That ties into something else people had a problem with, which is that Watchers didn't go insane properly enough.

We did eventually add in the deal with the sickness, which is very minor. I don't know, I think there's a lot of ways we could have handled the beginning of Pillars better mechanically and from a storytelling perspective. It was very difficult to get across the concept that, especially in an isometric perspective, that is this what your character's seeing or are other people in the world seeing this? Whereas if you play a game, for example, Undying, it's very clear that your character specifically is seeing things and having visions. It wasn't really communicated that well in Pillars, and mechanically it didn't really feel like it had any effect on you.

If you're continuing with the Watcher elements for Pillars 2, do you have ideas on how to improve that?

(laughs) Yeah, can't really talk about that.

There's one thing about Pillars 2 people asked about that's very hotly debated, which is the curve of what sort of enemies you might possibly face in the sequel to a game that has you destroying fragments of a god, dealing with Eothas, a Kraken, two archmages, a millennia-old entity, dragons, gods' assistants... that's a very high bar to raise if you go more epic. Is that your idea, to raise the bar?

I think we do have to a little bit, but we don't have to go up by the same distance that we did in the first game. I don't think we have to go way up into the stratosphere and be fighting like the gods themselves. You would expect to see more powerful creatures than you've seen before.

Something like epic level gnolls like in Mask of the Betrayer? It's a level 20 adventure.

I did some late-combat tuning for Mask of the Betrayer, and it's hard to balance around characters for that level. So we do have to be careful about how the power curve increases and about how crazy the enemies get, but we would want to go up, just not by the same proportion.

You already have something in mind?



So you have a plan now. I wanted to ask if you had a plan before Pillars, but...

No, I mean, we looked at Icewind Dale and Icewind Dale II, Baldur's Gate, Baldur's Gate II, so we had in mind that sense that where you go is going to go up even if the bottom end doesn't necessarily. In Icewind Dale II you started over at one, in Baldur's Gate II you started out somewhere around 7 and 8 and went up from there. There will be overlap I think, but I think everyone will expect the level of challenge to increase overall.

Do you believe there has to be some sort of power scaling for Pillars of Eternity 2 if you want to revisit enemies from the previous game?

Yeah, that's always a concern, but we're looking at making system changes. It's not like everything scaled perfectly in Pillars 1 anyway. Going back and looking at system overall can help, then looking at the content itself. Obviously you can run into things like "level 30 wolves, does this feel appropriate or should we have something else here?"

You have the advantage of not working with pen and paper, so you're not bound by those rules.

Yeah, we're not bound, but people will still sort of draw relative power balances, like "Okay, this baby dragon is this level and an elder bear is the same level, I don't like that, I don't think that natural creature should be on par with this supernatural creature." Some people will just look at the level as a way to gauge relative power level balance, it'll just feel weird, and even if it's not D&D, they'll say "I don't think a natural creature should be 20th level."

Some people have asked for a skippable tutorial.

You mean set apart from the game itself?

No, the idea might be like the Dungeon Be Gone mod for Baldur's Gate II. The idea is that if you've played before, you can just skip to the end of it.

So less about the tutorial itself and more about the content. I think we're always trying to make sure the beginning of the game doesn't feel like after you replay it, it drags, because people want to replay the game. So if that means it's skippable, or that it branches early enough that you can take off. I mean, that's why Fallout New Vegas is the way it is. You wake up at Doc Mitchell's, you go outside, and it's like there you go, you can go wherever you want. I think looking at ways to open up exploration for the player as early as possible is always good. In our case we're trying to communicate a very specific set of things, that whether we did it as a tutorial, we couldn't have just opened up immediately because if you never got to the machine, you never became a Watcher, how does that work? How does all the dialogue work after that point in time? I think some of it has to do with how we construct the story and other elements have to do with how we author the early game areas to allow you to access things more. But no, I realize that people, especially role playing game players, they're going to try things over and over again, so they should either be able to skip that stuff or if we can find a way to make the game open up like in Goodsprings, I think that's a good thing.

"A more direct connection between player actions and events," we've already discussed that I think. You're not in favor of having the storyline branch out entirely based on something you do in act one?

It just becomes really hard to author that. I do think that having more reactivity to what the player does more localized is achievable, and we want to do more with it. Especially when you get to something like changing a whole city, it can be a lot of work. Not that we shouldn't try to do it, but it's usually not that someone didn't think "Hey, shouldn't all this stuff be changed?" It's just making sure that we all get the time to do that stuff.

Would you like to do something like that again, like a Raedric-type scenario that actually changes a city? That's the type of reactivity people think of when they think of New Vegas or Alpha Protocol.

Even in New Vegas you don't necessarily change a whole lot. Things like Primm though, it's quite a bit of work to get Primm out, and it was so buggy too, like getting people to come back into the casino and stuff like that. I certainly always like to do stuff like that, but again it comes down to schedule and figuring out if we have the time to actually have all this stuff change.

I've recently played the Fallout 1.5 mod and I was surprised by how much reactivity it offered. It made me think of Pillars in a negative sense, because you don't have a whole lot of that.

Those are things we did do in New Vegas; the Legion can take over Helios One.

Is that something you're considering in the future?

Again, it's figuring out the logistics. Moving characters around can sometimes be difficult for us. For example, Bethesda's engine actually makes moving NPCs around pretty easy. If you mark them as persistent then great, they'll move on in the world forever. So in a lot of cases it's easy to move guys. For us, we have to pre-place most of our creatures and things. It's a little more difficult. I think having a world that feels more alive and reactive to time, those are goals of ours. I remember going back to classic games like the Ultimas, very robust. Ultima V, for example, is the game I have the most experience with, and they had a day and night schedules, and they had jobs, and they moved around. It was very cool, and so I think we want our worlds to feel more alive and have people be more active in them. And also, if we have systems to support that, it makes it easier for us to have reactivity where people move from place to place.


What do you think the writers should aspire to? Should they test something deeper about the human condition or should they focus on entertaining people no matter what kind of inspirations there are?

I think the writing in our games in general and also in Pillars should first and foremost entertain. People are coming to the games to be entertained. If the writing in the game does not entertain, but let's say it says something deeper about feelings and human condition, great, but a lot of people are never going to get to that because it's not enjoyable. Entertainment can mean a lot of different things to a lot of different people, so you have to be careful about how you use that, but you want to engage the player and make them interested in the conflicts the characters have as individuals. Also, I believe they should be tied into larger thematic things that you're exploring throughout the game throughout a series of games. Ultimately, I do think that if you can get people to think about them in terms of their own lives, things they might face in their own world, that's great because then they're actually empathizing with the people in the world in a way that feels more realistic to them, and I think they feel they have more investment in the choices they make.

I think that one of the best things we did with New Vegas was create interactions between factions that people really cared about, and a lot of that is due to the fact they emphasize and they associate those characters and those organizations with things in the real world. So I do think that's important and valuable because people really get worked up about it, like they really believe NCR is clearly the only faction that could possibly be good and others are like "No, Mr. House is." Not many people support the Legion, but you will find people that will defend the Legion, so I think having a connection to those higher themes is important because the entertainment is going to draw you in and get you interested in things, but then you need longer term sort of like second-level things to think about. As writers we need to keep those in mind as we're writing, so that we reinforce and touch on them.

With Pillars of Eternity, people seemed to care more about characters than the political themes.

I think for us, and this is something Eric and I talked about, I think it would have been more compelling if factions like the Dozens and the Crucible Knights, they continued to be present in the story past act two. And the fact that they weren't, and that they weren't really tied toward the ending, it makes that choice a little less meaningful in the middle of the game because it's, okay, well, it's sort of important in the context of Defiance Bay, but it doesn't really have the same impact of the Battle of Hoover Dam where it's like no, these guys are going to be in charge of this whole place and what they do is going to have a huge impact. So I do think that's a place where I think people would have been more interested in that if we had put more effort into making them more relevant for the rest of the game. I think that's our fault, I don't think that's anything on the players. Even in a fantasy setting, you have compelling political factions that people want to engage and are interested in, but yeah, we didn't carry that forward.

Do you think they should be more provocative?

I think it's necessary people give a shit about them. Like, there's an intellectual level you engage with things and an emotional level, and I think that you kind of want a mix of both, but if you don't have them engaging on an emotional level, then they're probably not going to engage at the intellectual level, so whether that means provocative or doing more extreme things, I do think that it's important to make a player feel that they really care about them, and it's hard to do that on a purely intellectual level.

Moving on, in Fallout New Vegas, why does Caesar want to ally with the White Glove Society?

Off the top of my head, I don't remember. I may have to ask Eric or John Gonzalez about the White Glove stuff.

Siege of Dragonspear has random encounters that are handcrafted. Is that something you would consider?

Oh yeah, those are cool. We've talked about random encounters and people definitely have different ideas of what that means. Some people think "this is filler," something that interrupts you in a random place and you just fight a bunch of guys, and it's unmemorable. But with a random encounter, randomness can just mean when it happens or exactly where it happens. It can still be handcrafted content for a little cool story with a memorable thing in it. That's what you want all your content to ideally be, but that can still be semi-random in where and when you encounter it. Something like that I think is cool and, yeah, I wouldn't be opposed to it.

That's something you can put in to make it more like Baldur's Gate. People said the same thing about hard counters, like "Oh, you have hard counters now, that's a change." Not hard counters specifically, I think you make a distinction between immunities and hard counters.

Yeah, it started with a desire to avoid boxing in people to feel dead-ended. Like "I can't progress, I don't have the magic combination of things to get through here." But it's sort of like the bar analogy: you don't have to go home, you just can't stay here. You don't have to use crushing, you just can't use piercing. It's not like they're immune to all the status effects, just a subset of status effects. Including some of those progressively, it does force the player, if they're really reliant on a certain type of attack or a certain affliction, to try something else. It's not radical or crazy or anything, but it took time to kind of see the best way for us to include them so that it didn't wind up causing craziness. We wanted to have it in there, we just didn't have time.


We have a list of old school features we'd like to see, and you can say whether or not they've crossed your mind. NPC schedules?

Yes, we've thought about that.

Interactive environments?

What does that mean?

In old games, for example, you could drain a moat by casting fireball, stuff like that.

You can get pretty crazy. I don't think we're going to do anything crazy, but we do want our environments to be more dynamic overall, so I'll just leave it at that. I wouldn't expect us to do anything really radical.

Unusual spells like walking on water or seeing through walls?

We certainly have thought about it, but it's hard. It's hard to put that stuff in and make it feel like it gets broad use. That doesn't mean we can't do it at all, but it's probably not something that would appear in the sense that it's a general ability that a Wizard would take. It'd probably be something like a consumable or a special type of interaction.

There's a question about the financial viability of these types of games. Is Pillars in a good place right now? There's been a discussion about how blobbers can never surpass ten thousand copies sold. What is your opinion on that: if a random indie developer makes a blobber, would you advise them to do it?

What do you mean when you say blobber exactly?

A recent blobber?


Well, you have the ongoing Grimoire, very interesting game. You have Paper Sorcerer, then you have Japanese ones like Stranger of Sword City which was very well-received, and we already have an interview with the creator lined up. In Japan, blobbers are huge; do you think they could be huge in western society?

I don't know. Huge?

Well, viable. Viable is huge in RPG terms.

I think so. I mean there are games that are super narrow in scope that can be viable. I think that's one thing that is good about development now. For all the things that RPG Codex folk, to a certain extent, don't like about how very broad audiences might change how we make games, today, because of all the different platforms we release on, you can have individual people and small groups of people make games that are viable even if they only sell a very small number of copies. I think it's good and cool, and PC is a great place to do it and even on mobile devices and tablets it's also possible as well. I know not everyone liked it, but I had a lot of fun with Battleheart. Battleheart obviously has the whac-a-mole system that a lot of people really hate, but it's confined enough that it doesn't get crazy. And that's a phone game.

The reason I asked is because of Might and Magic X. Because we talked to the director at Gamescom last year and he said that it wasn't viable to make more Might and Magic games at the moment. It sold a bit over a hundred thousands copies, and that wasn't enough for the expenses of a blobber project. Blobber fans are really concerned if that means companies don't see blobbers as commercially viable anymore, because it was tried and failed. Would Obsidian ever consider the idea of making a blobber?

I think we'd always consider it. We're making a tank MMO game, and that's like half the company. We're always open to making things. We made the Pathfinder Adventure Card game.

It's possible to put a smaller team on a smaller project?

Oh definitely, I mean the Pathfinder team was a dozen maybe? Don't quote me -- well, too late. But it wasn't a very large team. It was much smaller than the Pillars team.


Different methods of transport? People just want to travel by ship sometimes.

We've thought about that.

Can't say anything more at the moment?

No, but we've talked about it.

At PAX Prime you talked about more extensive mod tools. Do you know things you absolutely cannot offer modders?

That we can't?

Managing expectations, basically.

It would be really hard for us to offer level editing tools. Editing levels is just very complicated. Better file formats for editing and stuff like that? Pretty easy. Obviously I have very strong opinions on how to design and balance things, but I don't care how someone mods their game. I might think it's goofy or whatever, but I don't care, so having all those file formats as open as possible is good. Obviously there are always limitation to things, but that's one the places where I think we can make things more accessible.

What about adding new classes and races?

Classes would be difficult, but probably easier than races because races usually have the expectation of cosmetic changes; the models are different, the textures are different, that would probably be hard. Classes are probably a little easier, but off the top of my head I'd say don't count on it. Actually modifying, for example, this class in this way, that would probably be a lot easier.

Would it be possible to see, if I change the stats of the Fighter class, which enemies it actually affects?

Like if I modify Strength for Fighters, does it affect enemies in the world?


Yeah, that would be harder, because it's more code-level type stuff.

But enemies do use similar classes as player characters, right? On occasion at least?

It's a little weird. Characters that don't really have classes like kith races, they have abilities that are granted directly to them and it's not really an advancement-based thing. Every creature does have a class. For monsters, that class determines base defense and it determines base health/endurance and how it increases over time. Accuracy is generally for everyone, they all gain +3 per level and that's not defined anywhere, that's just a constant. And then when characters have abilities those are usually granted directly under the character template, it's not something that they sort of progress through. But I think making that stuff, if you can modify a creature, you can modify what abilities it has and stuff like that, but in terms of modifying... well actually, with starting health values and things like that, it's probably something else we can open up and make easier to modify. Because again, I don't care if you want to make everything bags of hit points or super-fragile or whatever, it's all fine.

Do you have the personal ambition to ever make a game with a modern engine as powerful as the one in Neverwinter Nights 2 again?

A personal ambition? No, but I do think that is very cool. I would say it's interesting because I actually met Ray Muzyka sometime in 2007, and we were briefly talking about the tools, and he said "Oh, you guys did so much cool stuff with the tools" and I was like "I kinda wished we had left a lot of it alone." Not entirely, but it is true that the Neverwinter Nights 2 toolset allows people to do a lot more specific and detailed things than the Neverwinter Nights 1 toolset did, and the Neverwinter Nights 1 toolset was so much easier for people to use. It was so much easier, for example, to make an exterior. It was just like a tiled interior, and yes, it didn't look as visually interesting, but it meant that it was very easy for someone to say "I want to make a level" and just jump in and do it.

That's one of the things Bethesda's toolset makes very easy. It's super easy to make areas, super easy to modify, super easy to track assets, and it's pretty darn powerful. Look at this way: there's no way in hell that our team could have made Fallout New Vegas without that tool. It was just impossible. And if you look at the mods, it's astounding what people can do with it. I personally think that is very cool. I hope we get to the point where we can actually develop tools like that. I wouldn't say it's a personally driving ambition, it's something that I hope we do.

And as for our file formats, that's something that whenever I see an opportunity, just make those files more open and easier to modify. Because, for example, things like Near Infinity and all those third party tools, we didn't use those when we made the Infinity Engine games, I had to make all my .2das in Notepad. I was a web designer so I was used to editing things in Notepad or maybe Excel if you're feeling fancy, but usually it was just opening it in Notepad. So if the data formats are accessible and easy to modify, usually even if you don't release tools, people will make tools, which is cool, but if they're human-readable then you can just open them in Notepad and edit them. If you're ambitious, you can do that. I love it when people mod the games, but I also know that with Neverwinter 2, it was an extremely huge undertaking to have that toolset. I don't want to promise anything like that because I just know it's expensive.


So if you compare the Neverwinter Nights 1, 2, and New Vegas toolsets, you'd think New Vegas was the best?

Well, here's the thing. I think our dialogue tools were probably better in Neverwinter 2 just due to the way we write conversations. But the ease in which you can put stuff together, and I also think it's a bit different because of the camera views, so it's not entirely fair to look at them the same way, I do really appreciate how easy it was in New Vegas to make stuff and modify stuff. And honestly, the scripting systems in Neverwinter Nights 1 and 2 are insane. You could do really crazy stuff, usually create a lot of bugs with that scripting system. The scripting system in the Bethesda engine is also very powerful and you can also do crazy stuff as well. But I do appreciate the ease-of-use stuff they had in Bethesda's editors.

I've worked with a lot of different toolsets and engines and stuff that we've developed internally, and making an engine, making a toolset is so incredibly time-consuming and so frustrating for so long. Keep in mind that when we released Neverwinter Nights 2, there were people that said "Give us the real toolset." "Like, what do you mean?" "You didn't make the game with this toolset, give us the real toolset." "No, dude, you're right. We didn't make the game with this toolset, we made the game with a toolset that was worse than this for two years, and you got the best version of it." And that blew people's minds, because it takes so much time and it's so hard and frustrating. So being able to go into Unity and kind of like drop things is nice. There's some things with the workflow that don't quite work for us, which is why we externalize things like conversation and stuff like that.

Do you think Pillars of Eternity could have happened without Unity?

I don't think it could have happened without middleware of some sort. Making everything from scratch takes so much time. It would have required some engine, I don't think we could have done it just by itself. And we already had our conversation tool we had developed for, well, originally, it was for Aliens and then it went on to be used in Dungeon Siege III and then it was fine for Pillars of Eternity, but we already had a tool that was for specific Obsidian conversation and quest needs. So we continue to update that and that's used separately. But yeah, without Unity itself... you could have said maybe you could have made it in Unreal or some other engine. But without a middleware solution, we did not... actually, you know what, it's possible we could have used Onyx, which is our internal engine, but Onyx's development was really more tailored to consoles, so it didn't seem appropriate for making a PC-centric game.

Have you thought about making an isometric PC RPG in Unreal?

Oh, we thought about it. We have a lot of people in the studio with Unreal experience. I worked a little bit with it on Alpha Protocol, but not that much because I wasn't on the project that long. Yeah, it's something we've thought about, but again not something we've looked into in great depth.

Unity was a better fit for you?

You would have to ask Adam Brennecke for the exact reasons why we picked it at that time. I think Unity, for whatever reason, seemed like it was easier to get up and running.

People have submitted a ton of questions about your dream historical RPG. What appeals to you about putting real history into a RPG setting?

I think it is cool to draw things from history, even if you're doing a fantastic spin on it, because you can draw a stronger connection to real history. You can, in some cases, demystify. So we have an idea of fantasy derived out of medieval or Renaissance things, but are so many stages removed from that, that now we kind of just understand history through fantasy. There's a blogger on Youtube who made a cool video talking about why it's cool to look at historical arms and armor when you're making a role playing game. And he mentioned how we're so many phases removed from the source material that we kinda forget how these things go together. For example, realism within a historical context for its own sake, I'm not that interested. I'm more interested in how it informs peoples' understanding of how things were and how they developed.

The Kingdom Come Deliverance guy just did a video on this. If you see what goes into a person putting on a suit of armor, you realize this is a huge ordeal, it requires help, the armor's expensive for reasons that are clear once you see how complex it really is, and it sort of informs like this is why there was a class of people that did this that were separated and they trained for it, they had endurance that was built around this. I think when you see historical games, you start to get an understanding of these things we understand through a fantastic lens, how they really fit into our past and how they evolved over time. I think you can also draw on things like religious and class distinctions. You get interesting themes to explore within the context of the game, and a lot of those things continue on to the current day. Like why are these things even relevant anymore, well, they had their basis all the way back here and continue moving forward.


And concepts could be different, like the concept of privacy.

The concept of madness, Foucault wrote a whole book about the history of madness. It's actually something we talked about with Brackenbury Sanitarium, which was like how fitting it is to have a place which is a separated location where people are mentally ill when it's not something that people would then in the equivalent time would have understood as such. Communities were generally much smaller in the olden times, and so if someone were having mental problems, people knew about it, everyone knew about it, whereas today the mentally ill, especially in large cities, are anonymous and sort of separated and there's a lot of negative things that come from that. But the idea of people being disturbed or having weird idiosyncrasies or quirks, the idea of that being insane as a certification or something to classify as, that's very different.

Anyway, I'm kind of getting off track but I do think that the history of madness does show that the whole conception of a concept like, what you said, privacy is very different. "Everyone has separate rooms, like what are you talking about? Everyone lives in this room, and when I say everyone I mean the livestock also live here, and they have a lot of body heat, it's cold. All of us are sleeping in this room and the chickens are over there and these guys are over here and we're all in a row and this is how we live." I think it can be very illuminating to try to get into the mindset of past ages and recreate that. And I think also it is crazy how just trying to recreate something the way that it was changes the perception of a fantasy setting putting a medieval veneer on a lot of modern things. And so if you try to get more into emulating it and simulating it, again you're getting more insight like "Oh wow, this really was different."

I did a Darklands sort-of thing for the Codex, and it's what got me into it. I mean I was sort of interested in history before Darklands, mostly from theater, because I did a lot of stuff with Shakespeare's histories, which were very fun, and then when I played Darklands I got interested in the Holy Roman Empire and the history of Germany, and then I went on to get a history degree. But it was through that sort of more-realistic, less-overtly-fantastic, more historically-grounded... it was still a fantasy game, but it had so many historically-grounded elements, like alchemy, that stuff is really cool.

I even went back recently and I was reading the Picatrix which is a source text on alchemy that was probably originally Greek and translated into Arabic and then translated into Latin in about the 11th century. It's really esoteric and weird, like when you see some of the references to authors, like alchemical authors, those people are then referenced in Darklands. Like the different formula they used, like Albertus Magnus, all these people were real alchemists... obviously, well did they actually make alchemy? They didn't necessarily do those things, but it's cool to see "Oh, these were real people," you can see the influence of Greek thinkers and philosophers and alchemists and Arabic thinkers and philosophers and how that worked together with alchemists working in the medieval period and moving on to the Renaissance.

So, seeing all that stuff together it's really neat to see how the information flowed through time. When I played Darklands, I was just fascinated by all that stuff. I went to college for it. Anytime someone plays a game that I worked on, and they take something back into their life that kind of changes how they look at things, that's always really cool. There was someone who wrote a postcard to us that said they played Honest Hearts, which a lot of people found to be a disappointing DLC in terms of gameplay, but this guy said he went to Zion and saw the national parks and became a park ranger. And that's incredible. I made it in part because I myself visited Zion and it was one of the most beautiful places I've been in my life and I wanted more people to know that this thing exists.

So your historical RPG is something of a passion project?

Yeah, it is definitely. When I got into the industry there were two types of games that I really wanted to make: I wanted to make a Fallout game... I actually wanted to make just a D&D game, I was interested in doing that. But I got to make it right away, I got to make Icewind Dale. And I really wanted to work on a Fallout game; I got a chance, and it was canceled, but I got to do New Vegas. And the last remaining one that I was interested in was making a historical RPG. And really, I guess you can look at when I talk about my three favorite games, it's Pool of Radiance (the original D&D game), Fallout 1, and then Darklands. So it's the only game that I really, well I don't want to say always, but for a very long time, I've wanted to make a historical RPG and it's something I'm still working toward.

I've talked to people about ideas, and one of the reasons I play Ars Magica is to play through a lot of the historical stuff and do research about character concepts and see how they work mechanically. Because you know, certain character concepts, historically... for example, mystic affective piety, like a holy mystic that goes into communion with god. Okay, that's a very cool historical thing, and how does that work in a game? And just thinking about it, and Ars Magica actually tries to have mystics and incorporate them into gameplay and seeing how that stuff works. I'm thinking about how it could be informative, but also a mechanically cool system to play with. And it's something I keep talking to Feargus about. The next project I work on, I really want it to be a historical RPG.

What's up with the alphabets you've posted on your...?

You mean the constructed languages?


We already had the constructed languages in a very basic sense, vocabulary and stuff like that, but we didn't have orthography for those. And if there are going to be things sort of actually written out in the game world, it seems like it's appropriate to actually have orthography that works with those languages. It's the next level of constructed languages, you have obviously the English sort of way of spelling things, but then you also have this culture who uses this alphabet so they spell things differently.


Do you expect to have stuff players have to decipher on their own?

Decipher on their own, or they might just see it. Maybe it's a clue or something like that, or if it's something where it's a large structure they might have something inscribed on it. This is a pet peeve of mine, but when I just see a bunch of rune-like, bless the artists, but they're just going to do a bunch of crazy runes and it doesn't mean anything. Whereas, if we do something like that, it actually means something, well then, figure out what it actually says in the constructed language and then write it out in the orthography that culture would use and that way it fits. So even if you wind up just clicking on it and it says "In Aedyran this says ladadadada" then great, but if you're an Ultima fan, you can actually look at the runes and you can actually read it yourself. You don't have to rely on the tooltip.

Would you say you're still expanding the Pillars brand in general?

Yeah. The best assets a video game company can really have outside of its people, which they can't own, are intellectual properties. It's certainly something that we want to do more with in the future, make more games with, but also people have asked about a tabletop game. Yeah, maybe we can do a tabletop game. Personally, I think it's a little bit weird for a couple of reasons. One, we designed the mechanics to work great on a computer. Whether or not they work great on a computer is one thing, but they certainly don't work well on tabletop because you have percentage operations, you have all sorts of things that don't really work that well when you're sitting around a table. Given that it's a class-based game, it's very D&D-like, I kinda just figure "Why don't you just play D&D in Pillars of Eternity?" Or just play Pathfinder in Pillars of Eternity. I guess from my perspective, I could also see some sort of tabletop adaptation where it is using another rule system, but I think if I were to personally spend any time designing a Pillars of Eternity tabletop ruleset, I would just design a whole ruleset for tabletop for it. That's something we regularly talk about, and I think people ask me about it almost on a weekly basis. The card game I think was received well, which is good, and of course we want to make more computer games.

Someone wants to know what your favorite history books are.

I made a couple of Youtube videos. One was about the importance of real world knowledge, and there's another one just about history stuff, and I mention a number of books. Favorite specific history books of mine are The Memory Palace of Matteo Ricci, Jonathan Spence wrote that, and it's about Matteo Ricci who was a missionary to China, he became fluent in Chinese and at that time they were amazed by it, that was something westerners had not really done. He was not very successful as a missionary, even though he became very well integrated into Chinese society. It's a fascinating look at how he was split between the two worlds.

You sympathize with that?

I don't know how much I can sympathize, I'm not a Jesuit or anything like that. I think I sympathize with the general idea of people living between two sort of cultures and having difficulty reconciling those differences. For him, it was extraordinary because he was doing things that almost no one had done.

Another book I really like is The Cheese and the Worms by Carlo Ginzburg. He also wrote another great book called The Night Battles. These both deal with folk beliefs in I think 15th and 16th century Italy. The Cheese and the Worms is interesting because it's about this miller who had really wild personal religious beliefs, strange cosmological beliefs; he thought the world was formed out of cheese and the worms ate through the cheese and became angels, and like all these other things. He's just a miller, and he has these very in-depth thoughts that he probably got by talking to people traveling just because the area he lived in was very well-traveled. He was brought up on charges I think three times and eventually executed for heresy. But what I think is interesting about it, is it shows that even simple people, even peasants, could have actually very deep sort-of ideas and thoughts about the nature of reality and the nature of world, god, the afterlife as opposed to this idea that they're all ignorant and the Church kinda defines everything, and that's all there is. So The Cheese and the Worms is very fascinating.

The Night Battles by the same author is also very interesting because it's about an Agrarian cult, the benandanti, and they are the welfarers, because that's what it means. They were born with a call, and that means they were designated to become these mystic warriors. During certain times of the year, they would enter into dream battles with witches and it was believed that if they won, then their crops would be saved and if they failed, it'd spoil the crops. They were Christian, but when Inquisitors found out about what they were doing, they were like "No, this sounds kinda like you're witches" and they actually convinced the benandanti that they were witches and they were like "Oh wow, you got us." So that's a very interesting thing.

It's more like a light history or pop history, but there's a book called The Perfect Heresy by Stephen O'Shea and it's about the crusade against the Cathars in what is now southern France. It's interesting on many levels in part because France would not look the way it does now if it were not for that crusade. It basically involved conquering all the heretics down there. It's very interesting because most heresies that existed in Europe were fairly localized and not very well entrenched, and the Church was pretty aggressive about extinguishing it very quickly. The Cathar heresy flourished and Cathars lived alongside Catholics in what is now southern France for a very long time without any opposition from the Church. So when it actually came time to deal with that movement, it required a crusade. They just couldn't get rid of it, it was so entrenched at all levels of society. It was pretty fascinating to read through.

Here's one more I read through recently, I think it's called By Sword and Fire by Sean McGlynn and the subtitle is Cruelty and Atrocity in Medieval Warfare. It kind of looks at the aspects of brutality in warfare and how often they were less about being insanely cruel and more about being practical. Like, sieges were very horrible and some of the things that happened during sieges were unbelievably awful, but they had a very specific strategic purpose. Also, a lot of the things that seem really brutal and bloody were done for very specific expedient purposes. Like say, for example, the Battle of Agincourt, at the end of the battle, Henry V ordered the execution of a huge number of captured French soldiers, which is pretty gnarly, until you look at the details and realize all these captured soldiers, who outnumbered them by the way, still had their armor on, were not far away from their weapons, and there was a group of French cavalry that had regrouped and was going to attempt to storm their camp. And so Henry said "Sorry, kill 'em, there's no way we're going to let these guys come in here and retake all their buddies who are still ready to fight." It's horrible, but I guess it's interesting to look through and see how this was the norm, almost everyone engaged in it, usually successful people. For example, King John of England: generally felt as a bad and cruel king. Richard I, he's the greatest, he's the paragon of chivalry, except that he did all these brutal things, but he was victorious. And so we kinda forget about all those horrible, cruel things he did.


You care deeply about history. Do you ever feel the impulse to add that to a fantasy franchise?

In what sense?

In the sense of bringing some elements of real history to make it more interesting.

Yeah, I mean there are certain little touches and hints of things that I put into Pillars for example. It has a very fantastic bent, but the story of St. Waidwen is based on Hans Böhm, the Drummer of Niklashausen, and the peasant farmer who saw a vision in the field that led him to turn society upside down, and in the case of St. Waidwen, become super crazy. So there's little bits of that that I take from history, but really most of these details I want to put into an actual historical game.

So you're saving them up.

Yeah, I guess the thing is because I know the context of a lot of these things is so important. For example, we have saints in Pillars of Eternity, but the concept of what saints are and what they are not, it's a very Catholic thing.


Yes, I've read a bunch of them in college, one of my focuses was hagiographies.

It's very specific to the mindset of the time and what people wanted from their stories.

And then also later, they took on a different shape as... there's a great book by Ken Woodward called Making Saints and it's about how the process of people becoming saints in the public consciousness has changed over the centuries and really it moved from something that was very much a local cult, because they used the term cult of the saint, and then later the Church is like "Hold on, there's a lot of people becoming saints we can't quite keep track of, we're not really sure if they're real or if they were pagans or a dog" in the case of Saint Guinefort which was a very interesting case in France where the story was, a snake had come into the Lord's baby's room, the dog killed the snake, and the Lord ran in and saw blood and the baby crying and killed the dog, and then the servant came in and realized the dog saved the baby, and they buried it and oh, the Church tells us that martyrs are saints. And so St. Guinefort is a dog, whoops, the Church didn't like that.

So the more the Church became involved, then the hagiographies became a means of confirming or documenting the true and verified miraculous properties of this person. Because the Church developed more and more criteria, like okay, it can't just be some guy who's really cool from your area that you celebrate on a certain feast day, like they have to perform miracles, they have to have done this and done that and so over time hagiographies had to check certain boxes for the Church. And they used to go to the, I'm going to get the term wrong, but basically Devil's Advocate which was a real position in the Church, because the Devil's Advocate would say "Okay, I'm going to tear this apart. You say this guy's a saint, I say no, this can be disproved, this isn't true" and so on. And after it got past the Devil's Advocate, it'd go to the Pope for official canonization, put 'em in a book, put 'em on a calendar, they're ready to go. I think a lot of that stuff is really fascinating, especially seeing how over time it changed from something that was a populist thing where all you had to do was say, and eventually the Church is like "Yeah, yeah, these guys have been around for a long time, of course we recognize them as a saint" to being something like "No, no, not until the Pope actually says, you have to go through this very regimented process."

All these story elements would be very well suited for a RPG. Why has nobody ever done this?

I don't know. One of the reasons I became fascinated with saints is from playing Darklands. It tells the little stories of the saints and I got my Oxford Dictionary of Saints which I still have, and you can read all their little stories. Some of them are angels, some of them never existed, it's very interesting to see the different types. Like Joan of Arc was kind of killed by the Church, not directly, but the Church was basically involved in condemning her. So she's a saint, but she's not actually a saint of martyrdom because the Church killed her. She's a saint of virginity.

It is interesting to see how those conceptions change over time, and now when the Church recognizes that, well okay this person probably didn't exist, or there's something else that's wrong, they can't de-canonize them, but they just kind of no longer put them on the calendar. You can celebrate St. George I guess, but maybe he didn't really exist at all. Or if he did, he didn't exist in any form that we know of him.

I'd like to see this kind of stuff in a RPG.

Tell Feargus. Feargus was not keen on the idea of a historical RPG for a long time, but he's warming up to it.


I think there's a market for it.

Yeah, and I do think it's something... like for example, some people talk about Lionheart, which was a game that Black Isle published, but it wasn't really that historical. It's something that I feel like if you want to make a historical game you kinda have to really embrace that. I think Feargus's fear, which I know because I told him how unfounded it is, is that "Oh, I'm going to have to know a whole lot about history." And I'm like no, you don't have to know a whole lot about Forgotten Realms when you play a Forgotten Realms game. Like that's kind of part of the experience, you're learning the lore of the world, Earth, but a few hundred years ago. There's a whole bunch of source books we can pull from. We have Jorge Salgado, he has a Master's in history. So we definitely have a lot of people interested in history stuff. I love all that stuff, and I'm always thinking of ways in which it could be a cool part of an interesting game narrative and also mechanics, things like that.

If Dungeons and Dragons was the big influence on the first RPGs, what do you think is going to be the big influence now?

At least from my perception, I feel like other genres are influencing RPGs more. So obviously you're seeing more elements from shooters coming into RPGs. I do think now we're referencing other crpgs than we're referencing tabletop stuff. To a certain extent that's healthy because some mechanics and styles of play are very well suited for a tabletop experience that aren't suited for a crpg experience necessarily.

Like I said in the talk, games like Torchbearer, I would be astounded if the guys who made Darkest Dungeon were at least not somewhat familiar with Torchbearer. I think games like Torchbearer or The Burning Wheel in general, I think it would be cool to see games based on that. Although making content for that... so, Burning Wheel has something like a hundred and something skills, it has skills for beekeeping, it's super-expansive. So as a player it's kinda like trying to find the crazy way you can use those outlandish skills, and it's very cool for a tabletop environment, but it'd be really hard to make content that supported all that stuff. But that'd be an interesting challenge. I would like to see more games like Ars Magica.

Would you like to make an Ars Magica game?

I'd rather make my own thing. I do enjoy playing Ars Magica, but it does have a very specific focus on being a hermetic mage in the 13th century, which is cool, but I kinda want to go a little bit later and broader. And I was very disappointed that the Kickstarter campaign didn't go very far. I would have liked to have seen someone make it. It has a lot of cool flexibility in the system, and I do really think that... like D&D is kind of like weird and abstract and you're just this roving homeless band of murderhobos, that's the best description of it. The thing I really like about Ars Magica is that it's centered on a covenant and it's much more suited to real sort-of medieval styles of like, people don't travel that far usually. You don't want to travel across Europe, you don't want to travel across the country you live in probably, it's dangerous and hard. But having things built around this place that changes over time is very cool, it makes you invested, both mechanically and emotionally. I think that can be very compelling. That aspect of Ars Magica is something I love that I would like to have in a historical game as well, like this is my place and this is how I want to protect it and build it, see it grow and change over time, I think that would be very compelling.

For a tabletop game that's already fairly niche it's hard to find the traction that's something like Pillars of Eternity. Like, there were a bunch of different things we could have tried doing, we could have tried to make something super outlandish, but we weren't even sure we could get a Kickstarter period to be successful. So that's why we did something that was traditional, Forgotten Realms-style fantasy, very D&D-like class based. I really don't like class-based games personally, I like character development that's a little more fluid and open.

Like Tyranny?

Like what?


Oh... yeah. Not necessarily learn-by-doing, but like the classless skill-based approach. Learn-by-doing for me is really tricky to balance around. Although it is what Darklands used, though it is difficult to balance around in Darklands. But if we're going to be successful, the best chance we have to be successful is make something that feels like an Infinity Engine successor.

I would still love to make a turn-based game. By the way, if I make a historical game, it's going to be turn-based. I know that Bobby Null and I would both love to work on more like a Gold Box-style game, which I think maybe, I don't have any special insight into this, like it's not like I'm hanging out with Brian Fargo all the time, but I think maybe Bard's Tale. I don't know if they want to do something like maybe a first person that goes into a top-down, that would be cool. That's not quite Bard's Tale, but that style of the Gold Box games with first-person exploration and then it goes out to a turn-based iso view.

Are you interested in doing something deterministic at some point?

I think that's very interesting. To the extreme, you have something like chess which is completely deterministic, and so it's ultra tactical.



Well, I think of tacticool as having the appearance of something tactical, but not actually being very tactical. I do think a deterministic combat system could be very interesting, but you really would have to develop the tactics to such an extent that it felt as rewarding as something like chess. It doesn't seem like it should be a tall order, but it is a tall order. Chess has had a thousand plus two thousand years to sort of develop and refine itself over time, and it went through some periods, like medieval chess is almost unrecognizable. Like en passant came a long time to come in, castling took a long time to come in. Something like that would be interesting. It's something I have thought about in terms of a historical game because you have to think more in terms of like what you think about someone who's actually a skilled combatant. Like no dude, he's just going to destroy this person. So it feels like it should be a more deterministic thing. It's something I think about too when I look at games like X-Com.

But X-Com cheats of course.

I know, but when you're wrestling with RNG, and you have a pretty solid basis of mechanics... I would be very interested seeing an X-Com style game where it's deterministic, where it's more like yeah your sniper can shoot from this distance, but you're only going to do this damage. If you shoot here, you can do even more damage guaranteed, but it's positional. So that's the decision you're making. I think that could be a really neat thing.

Even in terms of a more melee-oriented game, I think it could be interesting to sort of play around with that. If not deterministic, then much more sort of reliable, and it's more about the specific tactics than just chance. But it's also hard to dispute that, especially with a lot of unique items in Pillars of Eternity, when we moved more toward extreme results on a small percentage, people tend to respond a lot better to that. Like there's a three percent chance per hit that this guy's going to get petrified, and it's like don't do anything differently, just hit the guy and 3% of the time they're petrified or hit by a lightning bolt, and people like the jackpot feeling, and it's hard to deny that.

Will you be involved in the new game Tim Cain and Leonard Boyarsky are working on in any capacity?

Who knows?

We'll know in time. Have you ever been interested in other genres than role playing games?

One genre I'm interested in is, this will sound goofy but, team-based shooters. Like objective-based shooters. I play Overwatch quite a bit. I'm not that good, but I played a lot. It's very appealing. I played a lot of Wolfenstein: Enemy Territory, I played a lot of Battlefield 1942 and Battlefield 2, and so I do like that team-oriented objective-focused multiplayer.

I think the focus with, and Overwatch kinda proved it... you saw some other objective-focused team based games come out after Wolfenstein Enemy Territory, but they were all, and this sounds bad, a little too complex for people to sort of grab onto. And I think Overwatch is very straightforward. Objective: stand on this thing. You can't get a whole lot simpler than that. And each character is really distinctive, they each do three or four things, it doesn't take long to learn. For example, I was getting really tired of getting killed by Bastion, I'm like okay, what do I have to do, and it's like oh, Roadhog. How do I play this guy? And I went into the tutorial level for about three minutes and I got it, I went and played and I knocked Bastion out of turret mode and blew him up. So being able to quickly and easily play all those classes is great. The objectives are really straight-forward and simple. I think that's a big draw. And of course the characters are very distinctive, the animations are cool, I mean Blizzard always does a good job on executing that stuff.

If you were personally offered the option of working on something like Overwatch, where you can make something purely based on player skill, would you be interested?

I would be interested in that. I don't think I would ever want to permanently move away from role playing games, they're a big part of why I'm interested in doing this in the first place. But yeah, I would like to make a game like that. I enjoyed playing Armored Warfare and giving feedback to the Armored Warfare team. Because when I played Battlefield 1942, I was always jumping in the panzer. I had fun memories of that. I mean, I don't work in any meaningful capacity on their team, but I give design director feedback. It's fun being involved in that. I'm not going to give up working on RPGs to do it, but if they're like "Hey Josh, we're in-between projects, want to do this where we're taking Pillars of Eternity characters into a competitive game?" "All right, let's go!"

Finally, we have something a bit more lighthearted. You know, Tyranny uses the great hashtag #evilwon. Which hashtag would Pillars have?

I don't know... #wehavespacepigs?

Wonderful. Thank you very kindly.

There are 165 comments on RPG Codex Interview: Josh Sawyer at GDC Europe 2016

Site hosted by Sorcerer's Place Link us!
Codex definition, a book manuscript.
eXTReMe Tracker
rpgcodex.net RSS Feed
This page was created in 0.065061807632446 seconds