- 16 minutes ago
After setting a new bar for RPGs with Baldur's Gate 3, @LarianStudios is working on a new Divinity game, set to be their biggest and most ambitious project yet. At the 2025 Game Awards, we caught up with Swen Vincke, CEO of Larian Studios, Adam Smith, Writing Director, and Chrystal Ding, Lead Writer about the game's announcement, what players can expect, and how far through development they are.
Category
😹
FunTranscript
00:00At the 2025 Game Awards, we caught up with Larian, the studio behind the enormously successful
00:07Baldur's Gate 3, following the announcement of its new RPG, DFINITY. The flashy trailer was
00:13dark and cinematic, and Larian says the game is the studio's biggest and most ambitious project
00:19to date. What makes DFINITY so ambitious? Well, I mean, the level of agency that we're going to
00:26give you is very, very high, much higher than what happened in BG3, at least that's our ambition.
00:30And this is agency at pretty much all levels, right? So it's at the narrative, it's the systems level,
00:35so that makes us our most ambitious. It's the biggest game that we've ever made.
00:38Wow. Longer, deeper sequences of consequence. Okay. So things that, so one of the things that
00:44gives my heart joy when I play our games is when the things click into each other. The narrative,
00:51the systems themselves, they click into each other. And so when you have several situations that are
00:55going on and they all click into each other and they feel like a consequence of things,
00:58then it becomes really, really good. Can you explain the darker tone and atmosphere of the trailer?
01:02So we made it very dark because we want you to be the light that takes the darkness away,
01:08all right? Or be the one that pushes everybody into deeper darkness. And we want you to feel like a
01:13hero if you do the heroic things. And we want you to feel like an asshole if you do the
01:17asshole things
01:17or tyrant or whatever you want. But you get what I mean. Like we want to give you a wide
01:21spectrum of
01:22choice. Of course. And so things that were in DOS 2, for instance, were already pretty dark,
01:27but you didn't see them cinematically. Right. Now we're showing you what it looks like. Well,
01:30you can see it cinematically. This is what that means, all right? This is what it looks like. And
01:33it makes everything way more intimate and direct, which is great, which is what we want because it
01:38makes for a much more engaging experience. So yes, when you are going to see the darkness,
01:44you're going to understand that this is a game about contrast and where you can really be the beacon,
01:50the last hope of everyone or the one that really fucks everything up.
01:54I'm a huge 40k fan. So I'd heard rumblings of Total War 40k and I was really happy to see
02:01it.
02:01It's very different to us. We're not in the Grim Down as there is only war where there is light
02:05and
02:06hope. We don't want to make something that is all that tone. Can do that. But we want to confront
02:11you with horror and then say, what do you do with it and about it?
02:14Speaking of Original Sin 2, DFINITY won't be using the D&D ruleset. How's that going to play out?
02:20I mean, it's not a clone of DOS 2 and it's a new ruleset or own ruleset built on everything
02:28that we
02:28learned from all the previous games that we've done. So I think you're going to enjoy it. It's
02:32very easy to learn, very hard to master, but it allows you to do fantastic things, which is the most
02:37important bit. And it truly is going to give you and things like combat, it's going to really give
02:42you a really good time of figuring things out and how you can change things to one another and come
02:48up with all kinds of shenanigans. Terrific. And still turn-based? Yes.
02:52Okay. And similar party size to BG3? Yes. Okay.
02:57We are eager to learn more about the characters. Have you started casting the game yet?
03:01We're recording, so we're still casting, but we already have a bunch of actors that are in there.
03:07And can I ask if you're leaning more in the direction of some maybe returning faces that
03:11you've worked with in the past or some new voices? That's obviously for the cops.
03:16Right people for the rules. Very good. Very good. I'm sure you've had great experiences working with
03:22you. Yeah, for sure. So many of the folks who have gone on to like really celebrate the success with
03:26you of BG3. You're probably eager to work with them, I imagine, in some ways, but also you want to,
03:31you know, take characters in the direction that they're meant to go in that may be original and novel.
03:35But one of the things that was really interesting for me was realizing that once you cast somebody,
03:42you start to write for them. And that's actually really interesting because you get somebody in,
03:48you say, well, this person is not helping me to shape the character. They become part of the
03:51writing process, you know, the voice, you know, the, you know, the way their body language works.
03:56For me, it's very cool when somebody pops in unexpectedly and you go, I don't know where this
04:00person is. I don't know how they're going to conform. And they're giving you something that you've
04:03seen before. And you could work with these people for years, right? So Emma Gregory,
04:09who played with R&B 3. I will take something precious from you every hour that passes.
04:16The best I met her was at the rap party and she said it was like doing an invisible dance,
04:21something you've never met. And that video was great.
04:23Are there any technical updates to your engine that you are particularly excited about?
04:28Yeah. So, um, so one of the things that happened actually is when we started,
04:32when we ended the project that we were working on for D&D, we actually already knew we were going
04:37to work on Divinity. We had the engine that was waiting for us for that. So when we said,
04:41okay, our hearts are not in it, we have to do something else. We actually didn't have anything
04:45ready. All right. We have plans of doing things and making them ready, but we didn't have anything.
04:49So we had to scramble. But luckily we already had been building our engine for the future.
04:53We accelerated that development. So we actually just transitioned to it now. So it's new tech.
04:57And it's going to be in there. It's built obviously on the existing tech that we had
05:01before. But there's a lot of things that we can do now that we couldn't do before. So that's going
05:04to be really cool. That's actually the end gameplay. I think that's going to be the bit that excites
05:09most people other than the visual fidelity, which obviously has come up and the audio fidelity also,
05:14by the way. You mentioned that recording is already underway. Do you have a sense of how far
05:19through the project you are? So we think we're going to be recording all of development. We're not going to
05:24stop recording. So we built in Guildford, actually, we built a huge studio with six recording studios
05:29in there. The idea is that these are going to be recording nonstop on this game. It gives you an
05:32idea on the ambition. It's big because we want you to have lots of choice, lots and lots and lots
05:36and
05:36lots of choice. And that then leads somewhere, right? It doesn't just not lead anywhere. Whether we'll
05:41succeed, we'll find out or say it together. But that is the plan. So it's not because we're recording
05:48that we're already almost done. This is a project where recording is almost at the beginning. Coming
05:52off a game as critically and commercially successful as Borders Gate 3, does moving on to a new project
05:58feel like more pressure or less? Yeah, it's more pressure. So there's definitely the weight of the
06:04expectations weighs high. And so we're trying not to think about it because we have to make our own thing.
06:11But even when we don't think of the outside expectations, just the expectations come out for
06:15ourselves. We need to sell that high. And so that's what we're working towards. So it was an
06:21interesting journey that the team already has done on this because when we started on this, it was
06:25like, oh, we know how to do this. All right. By now, we've been humbled by the actual development
06:29again. So we don't know anything. We have to relearn everything. But we do have the experience of the
06:33past with us. So every game has its own language that you need to learn, right? So nothing crumbles
06:38your ego faster than trying to make a video game. A friend of mine who wrote a very long form
06:43of
06:43feature racing said, like, games don't want to be made. They're like, every one of them is like a
06:47miracle because they resist. This is 100% true. They really don't want to be made. Every single
06:53morning you're happy if it actually even starts up. So it's very messy. We're in production. So that
06:58means that we already passed pre-production. So that means that we already step further. Production is
07:03hell. And in a game that iterates so much, or with a developer that iterates so much. So what Adam
07:08said is very true. A game resists us, but we shall vanquish it.
07:14Going back to the characters. Would you say they're not complete until they're cast?
07:18Believe until we should, honestly. You're always learning about them, right? So I think
07:23because the way the game works and because the player is the most important character,
07:27or the players are the most important characters, so anything else is reacting to them. Which means
07:32that you don't know what the character needs until you know what the player needs. So even as you come
07:36toward the end, one of the great things about working on Baldur's Gate 3 was you would have
07:40these moments where you say, oh shit, they'd do this. If I say this or if I do this, they
07:45would
07:45have to do this. And you hadn't planned it. You hadn't thought about it. But once you know them,
07:49you know how they react. So you're kind of improvising along with them. So we have characters
07:53that I'm already happy with. I don't know who the hell they are yet, because they're not on the end
07:56of the journey yet. And we'll know who they are then. Right now they're a sketch, and we're starting to
08:01piece them together. Don't forget the visuals also. The visuals change also through development.
08:07They have a big impact also. The identity of characters is a continuous journey. Plus,
08:13if we go to Early Access, it also depends on if players like them or not. Do we have to
08:18like
08:18Shadowheart, for instance, went through a journey of transformation. So did Astarion actually. So
08:22they will. So there was a lot of tinkering and toying with the characters as we progressed,
08:27then we got to know them better. Are you concerned that players may be expecting
08:31Borders Gate 4? Yeah, I know. I mean, so we had the same thing with, when we announced BG3,
08:36everybody said, oh, it's DOS 2. Yeah. Right. Yeah. So that was, that was for us a huge surprise.
08:43You said, yes, no. Because I'm sure you don't want to make the same game. No, no, we don't want
08:48to make
08:48the same game though. Yeah. So we want to make a better game, which has new things in there that
08:53lets you do
08:53things that you haven't done before. And I think we have those things in the game. So we just need
08:58to make them work all together, which is, might take some time. Well, Sven said earlier, there was
09:03a moment we realized our hearts weren't in it. Our hearts are in this. We're excited. We wouldn't
09:08be excited if we were making the same game again. So it needs to make us, we need, we need
09:11to,
09:12because we play it every day, right? And you need to, you need to have those moments where you go,
09:18oh shit. Yeah. I haven't seen this before. I haven't done that before. One of the worst things
09:23about trying to make this goddamn game is the amount of times we do something. We go, this is
09:26really cool. But also we did do it in Baldur's Gate 3. And it's not a specific thing, but you
09:31recognize the patterns, right? And, uh, and we were very, very harsh on ourselves when we were making
09:37BE3. Same for every Larian game. You go, I don't want to repeat a pattern. You want to keep people
09:42something new so they don't get familiar and they don't say, ah, I did this quest already. You know, like,
09:46uh, so this one is the one where I do A, then I do B and then I do C.
09:49And now we're like, ah,
09:51shit, this is really cool. But also, yeah, we did do that. We did that already, you know? So
09:54we're constantly trying to find things. And this is where the, the agency comes in again,
09:58because if you make the connections between, uh, characters, narratives, situations, gameplay,
10:05more complex and how they ripple out, then they don't feel complex to the player because
10:10they're just naturally experiencing them for us though. Oh my God. Uh, like, you know,
10:14you're, you're looking at this matrix and saying like, how does it all fit together?
10:17But what you want is for the player to just experience it and go, yeah, that was just a story
10:20that I got told. With so much player choice in your games, how much does flexibility in your workflow
10:26and your ability to adapt act as your secret source? It is. I mean, because the, but we build our,
10:33this is also why our new engine is so important to us because it allows us to do things more
10:37flexible
10:37also. So everything we do is to allow our efficiency to become higher. And then we use
10:43that efficiency to be able to try more. Uh, so if you, for instance, um, the recording that's happening
10:49in our, uh, recording studios in, um, in, in Guilford right now, uh, we automated a bunch,
10:55a whole bunch of the pipeline. So at this point it's almost as if the moment it's recorded is in
10:58game,
10:58not a hundred percent yet, but we're getting there, which is internal allows us to do more,
11:03right? Because now we can record more, we can try stuff, we can say, and, and, and do like that.
11:07So it's a, it's an important part for us. And the same thing goes for our systems. Also the,
11:12there's a new system is called FOF. It doesn't matter, uh, but it's, it's a subsystem that allows
11:17our designers to more rapidly experiment with rules, uh, which is good because it allows them to do
11:21flexibility and then we have better analytics engines. So when our play testers are playing,
11:25we say, ah, that's not working, that's working. So now we can change the rules faster, right? So,
11:29and then we've seen the story is not working or permutation or work. It's fine. We'll just re-record it.
11:32We got the studio directors are there anyway. So, uh, so you got all that, but it's, um,
11:36it's taken years and years to, to get to this point, right? So every single game we've improved
11:41the stack a little bit. And then this, we have tech that plays the game automatically. So there's
11:45automated testing going on so that things are being, because it's, we already had on BG3, right?
11:49So large. So, and any, so when a developer, we're almost there, not a hundred percent,
11:54when a developer does a change now, the game or first place itself so that the change doesn't break
11:59for the entire company. Uh, and only then do we get builds in our hands, but that happened on,
12:04you don't want to know how much, uh, servers are active doing this thing. Uh, but it, it helps us,
12:10uh, tremendously in, in, in more rapid iteration. Uh, I was talking to somebody who, um,
12:17had worked on a game and they had had script lock before they had a level built. Oh yes.
12:23And I was like, how do you do that? Because like for us, we have to be walking around the
12:26game world
12:27to know the writing is good because it has to be, we have to see it in the game. Yeah.
12:31And obviously
12:31it's an imperfect version of the game, but the closer it is to the thing the player will be looking
12:35at the better we know it because the thing is like, you know, we, we're writing for a very specific
12:40medium and a very complicated genre within that medium. So on paper, it can be a really good
12:46story you put in the game. It's often really, isn't a very good. How often that you'd find yourself
12:50writing like a dialogue and you read it through on the, on, in the file and you're like, that seems
12:55fine.
12:55Put it in the game. Oh, that's garbage. I didn't do any of that. Yeah. No, well you, you say
13:01sure,
13:01but until you see it because it's an, it's an, it's an, that, that bit is an unbelievable part
13:05of the process. You feel it in your gut. Like it's not, it's not even your brain just go,
13:08that, I don't, I don't buy that. Yeah. Yeah. And flow also, right. So,
13:12because the pacing of things is so important. For more on DFINITY
13:16and the biggest upcoming PC games, subscribe to PC Gamer.
Comments