In the second half of our interview with Supergiant Games Creative Director Greg Kasavin, we discuss their most recent game, Hades, and how early-access feedback impacts game development. You can read part one of our interview here.
You’ve discussed how you’ve worked to develop these complex, procedurally-generated branching stories within your game Pyre. What did your team decide to do for Hades?
One of the things we enjoyed working on as part of Pyre was writing a story with a big cast of characters with intricate relationships and so on. With Hades, we have another great big cast of characters.
To back up briefly, the thing about Bastion and Transistor, although they have a strong story to them, they have very small casts of characters. They can feel very lonely by design.
Although there’s like warmth there, and you feel this great bond with the characters who do exist, they’re these like sprawling worlds with very few inhabitants. With Pyre we wanted to make an inhabited world where you could make friends and go on this big road trip with them. With Hades, we were like, “we love the great big cast of colorful characters thing, let’s do that again, but this time there’ll be fully voiced.”
I think, as much as I loved working on Sahrian, if there is something more direct, it makes the character feel even more real if you just hear them talk in the language you do understand. And it’s more important to the specific flavor of Hades, and the directness of Hades to have them do that.
There’s still a ton of variability in Hades, but it’s around the game state. It’s that we have all these characters talk to you about all sorts of really specific different things, depending on the exact moment that you’re in.
In Hades, you’re the prince of the underworld of Greek myth, who’s trying to escape. Your father, the lord of the dead says, “You ain’t never getting out of here you ignoble brat. How dare you try to defy me? Be my guest if you think you could get out of here, go get out.” You’re like, “Oh yeah, I’ll show you.” So you’re trying to get out. Along the way, you have the help of the Olympians who appear—”Oh, we have this long-lost nephew. Come join us, we’ll help you get out of this bad family situation that you’re in.”
These different Olympians will say, if you’re low on health, “Oh my God, you’re in … What’s wrong man? What’s gotten into you?” Or if you’re in a particular area, they’ll say, “Oh man, you’ve made it all the way to the Asphodel Meadows, that’s great. You’re making good progress.” They respond very specifically to the game context. It’s procedural and [the lines] can mean different things, but it is the opposite of scripted. I come up with as many different possible contexts in which a character can talk to you and then we load them all up into the game and have the game hold the right one for the right situation.
It’s designed fundamentally to be deeply nonlinear and replayable, so that this content can happen in any order. And the great thing about it is that, even as the person writing it, I should have zero surprise around it. Right? I know exactly what all of it is. Yet, it surprises, just like Pyre did, it surprises me all the time because of the specific ways it can sequence together. There’s Aphrodite walking over there. Things like, you meet Aphrodite after you’ve met Athena in a run, because even that part is randomized, and Aphrodite would be like, “Oh, Athena has already gotten to you?”
The gods will respond to other gods who you’ve encountered, and that comes from the randomness of the game structure itself, and then the narrative falls into place to make it seem grounded in the world. While we’re making a rougelike hack and slash game, a style of game that many other developers have pursued, in the same way we felt like we could make something unique in the action RPG space with Bastion back in the day, we felt like we could add a distinctive narrative component and a sense of cohesion to this type of a genre.
How does having early access affect the narrative development? Because you now have live feedback coming in?
At the same time, how does early access affect the players? Do players play the game once in early access, and then put the game down because they believe they’ve completed it, even though it’s still in progress?
At the inception of the project, it was a really fun thought exercise. I think if you’re a fan of Supergiant games, the idea of us making an early access rougelike sounds almost antithetical to everything we’ve ever done because we’re known for these games with a traditional beginning, middle and end. We’ve prided ourselves on our games having like a sense of completeness to them. So, an early access game, what’s going on there? But we were like, “How do we do this?” We’re excited to try it, but we wanted it to have all the stuff that we like putting into our games. When it comes to the narrative, our approach is essentially to think of it almost like this initial early access launch would be like a pilot episode.
We introduce the story, we introduce the cast of characters, and then with each major update that we roll out every month or two, we add to the story, we introduce characters, we introduce new situations, we build on existing characters and so on. The players experiencing the game in early access get to see the story unfold. And we’ve said that we intend to reserve the final narrative outcome for when early access is complete. Right now, the game does have an ending of sorts. It has a huge variety of different endings, given the replayable structure, but it’s not the complete narrative.
Even though that the game is early access, we have inworld explanations for the lack of completeness of certain aspects of the game. It’s always fun to think about how to justify aspects of the design in the context of our worlds, and I think we do it in a playful way that’s consistent with the tone of the game.
The feedback we’ve been getting has been really encouraging. To answer the other part of your question, I’ll answer it by way of comparison, and we can come back to Pyre. With Pyre, from a writing standpoint, I worked on that game for three years. And you play test it along the way, even with say dozens of people or hundreds of people, but even still, you don’t really know what you have until you put it out there. You work on it for three years, you put it out there, it’s this big cathartic moment as a developer, “What does everyone think?” People then burn through the game.
Pyre is a much longer game than Transistor, but even still, people burn through it in 48 hours or whatever, and then that’s it. All through development, you’re asking yourselves as a development team, “Oh, does this character work for people? Is this part okay? Should we be doing more here, more there?” With early access, with Hades, we introduce a new character, we find out right away if people like this character or what people like about this character, and that part has been really great from a writing standpoint actually because it lets me sharpen my own instincts around what aspects of the story are working really effectively, what are the characters who are really resonating with players.
We wouldn’t put in a character if we didn’t want them to resonate on some level, but of course some characters are by design meant to be more central than others. Are the central characters hitting the right notes for people? It’s helpful being able to move forward with the confidence of having had that feedback from a lot of players along the way.
To get a little bit more specific with Hades, the game has a sense of humor to it despite the kind of dark veneer, the underworld setting. It’s a game called Hades, you don’t assume a game like that would be funny necessarily. We try to play into those expectations and subvert them in a way by having some humor to it. But humor in games is tough. Not everyone has the same sense of humor.
With our early access launch, we have some humor in there, but it’s like, “Man, what are people going to think of this?” And it turned out the feedback we were getting around the humor was really positive. People liked the humorous aspects of the game, so it made us be able to move forward kind of with more confidence that this is something, this really is the tone of the game.
With that feedback, I think we have a really firm grasp of the tone and where the game should be silly and where it should be serious. We want a tone that is aligned with the play experience itself, and I think roguelikes, if you’ve played roguelikes, they have a very slapstick quality where one moment you’re on top of the world, you feel like you’re totally unstoppable. You’ve got the perfect build, you’re destroying everything in your path, but then one boneheaded mistake and you … It’s like stepping on a rake and having it smack you in the forehead. You feel like such a bozo you made a stupid mistake, and you could either be really mad at yourself, or you could kind of laugh at yourself, be like, “Oh my God, I can’t believe it.”
And we want the game’s tone to have that kind of quality of laughing with you, not at you. A character who’s self-effacing in the face of failure, but is very determined. So, when he fails, he’s like, “Oh my God. I can’t believe I did that,” but he’s going to go and try again, he’s not going to get super pissed off at himself. And hopefully, that aligns with the player’s own experience, helps players push through their inevitable setbacks and be willing to try again as part of the core experience of the game.
Roguelikes can be notorious for their difficulty. Is Hades leaning into that? Is getting to the end hard?
It’s a really good question. We set out to make Hades … We say that it’s a game where you don’t have to be a god to experience what is exciting and enjoyable, although if you happen to be a God, it’s got you covered. It can have a very significant and tightly crafted challenge to it. But, while we love the experience of rougelike games, but for sure they sometimes … Like, what’s so exciting and thrilling about them can sometimes be like hidden behind a really steep difficulty curve. You have to climb this mountain to be able to have this experience.
We do want the barrier to entry on this game to be lower. We have a narrative to it. There are other reasons to enjoy Hades besides just the thrill of barely scraping by. That thrill is there too. And, being a game where you play as an immortal character, when you suffer these setbacks and get sent all the way back to the house of Hades (the starting point), you have the ability to get stronger in a permanent way. So not only are you as a player picking up, both as a player and as a character, that you’re picking up knowledge of what’s ahead.
Because for example, while there are random aspects, it’s the same guardian boss initially that’s going to be facing you at the end of Tartarus. You come back to her and you know who she’s going to be, and they have a conversation about it. It’s a rematch now, and you’ve learned a little bit more and you’re maybe a little bit stronger now too.
So we want for players both through their experience and through these other systems in the game to be able to overcome some of the earlier setbacks that they may have run into, and for it to ultimately be a game where, if you’re reasonably invested in this game, you care about it, and you want to get to the end, you’ll be able to reach a satisfying sense of narrative closure like you have been able to in our past games. However, it’s a game designed around replayability and it will keep going. There will be more story, there’ll be more to discover from there that, if you get better and better at the game, if you’re super skilled and stuff like that, there’s going to be plenty more stuff that you could see.
You mention more there’s stuff to see. If we come back to early access, is there content, like endings, that won’t be revealed until the game is out of early access?
There will be new stuff. Like I said, the ending of the game and other aspects of the narrative we deliberately are going to hold back. It’s not like we’re sitting on it and we haven’t put it in. We’re just going to make the endings after everything else, knowing what we want to do with it. But that will be in our version 1.0 — the last big update for our full launch. For right now there’s a mix, tons of unique one-off content, these bigger story moments that never repeat, and a lot of content that is designed to be repeatable. But before any of it repeats, you would have to have looped through the game like 20 or 30 times, so hours and hours of play before a small line will repeat, for example.
So, by the time the line repeats you probably won’t remember it anyway, because it’s not a line that should feel kind of repetitive if you were to hear it again. That’s how we designed the narrative of the game. It should feel like a unique forward-moving story no matter how many times you played the game over and over, because you’re an immortal god with no sense of the passage of time. You could just infinitely play the game as a denizen of the underworld and just keep having these unique experiences as you run into these different kinds of characters, stuff like that.
As the game develops, how far away do you think you are from 1.0 launch?
We expect to exit early access sometime in the second half of next year . Even though we have most of the full game structure right now, we still have quite a ways to go because we want to give ourselves lots and lots of time to iterate on it and improve every aspect of the game, and just to, hopefully, fulfill all the potential that early adopters have already said they see in it. We love working on this game and with each of our games, we want them to live up to our past work if nothing else and have the potential to be the favorite game that we’ve ever made for our fans out there. That takes us a certain amount of time and we just want to take all the time that this game needs to make sure we’ve done everything that we really want to do with it.
And then from that point forward, that version 1.0 launch is not necessarily the end, it’s kind of like at that point we’ll see how it goes, we’ll see what everybody says when we design the world of Hades. We designed it for early access from the ground up, and unlike games like Transistor that may have no obvious room for a continuation once it gets to the end, a game like Hades can be more of a world where there are many stories that we could tell here.
And that was by design? It was meant to be sequel-able?
Not even sequel-able, more like a living game where we can continue to add content. We don’t intend to have all 12 of the Olympians in the game. Canonically there’s like the 12 Olympians. A certain number of them will be there as part of the story, but that’s not to say the other ones couldn’t join the fray later.
That’s not meant to be a commitment of any sort, it’s just the sort of thing we care about. There’s so many gods, monsters and heroes of Greek myth that we could imagine in this universe. If you think about Greek heroes, they all die, so they all wind up here [in the underworld] too. We have these characters like Achilles and Theseus that are in the story already, but there are many more that we could imagine being there potentially.
The story won’t feel incomplete without them, but the point is that we could have more stories. That’s the potential that we see in the game in the long term. But, that’s getting a bit ahead of myself—we have to see what the response to the game is at launch. So far it’s been great getting positive feedback and everything, but we have plenty more to do. Our full focus is nailing each of our major updates in the process of getting to that version 1.0, and then from there, it’s a game that hopefully we can justify continuing to work on.
As we come up on time, just to close us out, you’ve now written all these various games. You’ve created lots of characters. Were there are any characters that were your favorite to write?
It’s always fun to ask this question.
No, it is. I love the question, but as a father of two children, you are straight up asking me to choose a favorite. It sounds disingenuous or something, but these characters are, they’re my baby. I love our characters very deeply. I remember all of them vividly from Bastion to Transistor to Pyre, now to Hades. We had to nurture them into existence. And then once the games launch, they’re kind of like, now that they’re their own thing, whether people like them or don’t like them, they’re kind of like, I no longer control them, they’re out in the wild now. But I love them dearly. But oh man.
Having said that, there’s some characters in Pyre that were very near and dear to my heart. There’s the character Sandra who’s the only optional character in Pyre. She’s the character in the Beyonder Crystal who you can really get to know over the course of the game. But, while your relationship with her can go very deep, it’s purely optional. It was fun to think about a character who had been trapped in a crystal ball for 837 years and what that would do to a person. I was really glad to see Sandra stand out to many Pyre players. She was a character, that during the course of development, there were times when we didn’t know if Sandra was going to make it because she was often looked at as, “Well, we need the necessary characters. Do we really need this Sandra character?” I’m like, “We need the Sandra character, please let me try it.” I love her a lot.
There’s a character called Volfred Sandlewood in Pyre also who is central to the story. I love trying to write complex characters, and to me he’s a complicated guy. You start off feeling one way about him and I think end up feeling a different way about him, and I really enjoyed trying to thread the needle on that. He had to come off as very persuasive because he’s a leader. But he’s a leader who, when you first meet him, you’re not sure if maybe he’s up to something. You’re suspicious of him and you don’t get off on the right foot, but you come around hopefully to respecting him and understanding that he wasn’t trying to pull one over on you, he was sincere.
It was trying to make a character who you, as a player, could be genuinely persuaded, and it was a really interesting writing challenge. But man, I could go down the list.
I love our antagonist characters. Especially our antagonist characters; Zulf in Bastion, Royce Bracket in Transistor, Oralech in Pyre, Lord Hades himself in Hades. I say antagonist, because to me it’s more interesting than a villain. It’s a distinction. An antagonist is someone who opposes you, not necessarily someone who is just trying to create evil. They’re just someone who wants something that is opposite from what you want. Are they a terrible person? Maybe, but not necessarily. Our lives are filled with antagonistic forces and people who don’t want what we want, but they’re not necessarily bad people. Sometimes it’s just the circumstances that put us at odds. And our games deep down, despite they’re wildly different settings and themes, they’re about character’s trying to understand each other in the world around them. And I think our stories always have empathy at the heart of them, understanding one another, even when someone may have done something really terrible, at least understanding what led them to do that.
So I enjoy writing antagonists and going through the thought process of where did this person go wrong? Where did this go astray? What caused this person to hate someone or something like that? What led to that to really get in their heads that way? It’s really fun and invigorating from a writing standpoint to try to understand the character that way, and I think my love for them comes from that.
I feel like I honestly know the characters in my stories just better than I know most people. In spite of all my blabber-mouthing to you right now, I am a very … I live almost like a hermetic lifestyle. I don’t have much of a social life. I am very introverted in social situations, so I know these characters better than I know most humans. And I connect with the world through writing, and then I connect with other people through their own response to our games. That part is very gratifying, when people enjoy our worlds and characters and stories, I feel like I connect with them that way more than is often possible just having a normal conversation with them.
That’s part of what I love about the process and the result.