Afterwords: Bastion

by Matt Miller on Sep 09, 2011 at 11:00 AM

Like a lot of the folks who played this summer’s Bastion from Supergiant Games, we came away extremely impressed. Now that the game has been out for several weeks, we wanted to ask creative director Greg Kasavin about the experience of making the game. He had lots of cool details to share. A portion of the interview appears in Game Informer issue 222. For all of Kasavin’s responses, check out the complete interview below. 

Where and when did the idea of Bastion’s narrator emerge? Why was it important to keep the narration a continuous part of the gameplay? Who ended up performing the narration, and how was that decided?

Like most of the ideas in Bastion, the idea for the narration was born in the living room of a house in San Jose where the studio is located. It wasn't something that was contemplated from the beginning of the project, but then, the project didn't start with some grand design document or detailed feature set. The narration came up as an idea during the course of the nine months the game spent in a prototyping phase, during which time it was mostly just Amir Rao and Gavin Simon, the studio co-founders, rapidly trying different things.

From the beginning of the project we felt we wanted to make a game that could have some emotional weight to it, but we also knew we didn't want to interrupt the play experience for any reason. At first these goals seemed contradictory, but as it turned out, using narration allowed us to deliver story and context and narrative depth at the player's pace. It was the perfect solution to what we wanted, plus it gave me some really interesting constraints to work with from a writing perspective: We would have a story filtered entirely through a single voice, who would be there to deepen the player's interactions onscreen by giving context to those actions.

We wouldn't have pursued this narration technique at all were it not for Logan Cunningham our voice actor. He's a childhood friend of Amir's and was rooming with Darren Korb our audio director at the time, so we had great access to this really talented actor who we were close with. And when you're a small team like we are, you need to make the most of every advantage like that. So we were able iterate rapidly with the writing and recording, trying lots of different things from a narrative standpoint. We never considered any other actor and Logan was just perfect for the part we had in mind, which happened to line up just right with the kind of story I had been mulling. In the end it turned out to be a large volume of content, somewhere around 3,000 lines of narration. Not only is this because the game itself is pretty big, at around eight or more hours for the first play-through, but because we wanted to account for as many little situational moments as we could think of in order to make players feel like their experience with the game was unique.

The music and visuals of Bastion combine to craft a world unlike many others we’ve seen in recent video games. What sources did you look or listen to for inspiration for the world of Bastion? What feelings or references do you hope the visuals and audio evoke in players?

We knew we wanted Bastion to be an atmospheric game because we feel that one of the most important aspects of games is their transportive quality, their ability to make you feel like you're in some completely different place. From the earliest days on the project we were talking about the tone we wanted to strike, and how we would express that through the game. One of the influences I brought up early on is the American author Cormac McCarthy, who's written novels like Blood Meridian, No Country for Old Men, and The Road. He's got this amazing lyrical writing style, minimal dialogue rich with subtext crossed with these beautiful natural scenes. His writing has an Old West flavor even if it's not set in that kind of environment. We wondered what that type of style would sound like in a game.

One important difference, though, was that we wanted to extract the sense of evil from out of his work. Plenty of games deal with evil. We wanted to have a more positive and hopeful tone, ultimately. We wanted the game to be suitable for players of all ages, and I was very interested in writing a story from that perspective – like a modern version of a traditional fairytale.

With some of these ideas in mind, Darren Korb our audio director started creating music for the game. He called the style he came up with "acoustic frontier trip-hop", as a short-hand for the fantastic-yet-familiar setting and tone we wanted. It was the first content to really make good on the tonal ideas and bring them to life. Darren's music just nailed the tone, and the music ended up being important all through development. Our close collaboration enabled some of the key moments in the game where the music really comes into play.

Our narration came soon after Darren found the musical style, including the narrator's voice and manner of speech. The art style soon followed suit, too, when Jen Zee our artist joined the project. While the art style of Bastion was still largely undefined when she joined, there was a sense for the tone and fiction, and she was able to take that and create this high-contrast visual style that we fell in love with. We wanted this world that was strikingly beautiful yet one in which you could tell something mysterious and wrong had happened, and Jen really captured that through her art. Also our collaboration allowed for the sort of world where all the different visual details have greater meaning if you want to look for it, so the game isn't just pretty for its own sake. I think this is really the key to having an interesting gameworld – there has to be something there beneath the surface.

Next up: What weapons just didn't make the cut?

From a technical and time-consumption perspective, what was the most challenging element of the game to implement?

Getting the game running on the Xbox 360 the way we wanted it to was one of the bigger technical challenges. Bastion started its life as a PC game though we always wanted it to be on both console and PC. When we first got it up and running on the 360, the game moved at about two frames per second when it wasn't crashing altogether. But thanks especially to some technical wizardry from Andrew Wang, our chief technology officer who joined us from Infinity Ward, we got the game up to our intended 60 frames per second in a relatively short amount of time. Also in general there are some really specific and unusual technical requirements for making an Xbox 360 game, which Andrew and Gavin had to solve for.

I'm not an engineer personally, though, and in my case probably the most challenging and time-consuming aspect of the project was getting all the writing and narration feeling just right – both from a content and an implementation standpoint. I was tuning thousands of lines of audio to fractions of a second to get the timing just right in-game, since the cadence of the narrator's speech was nearly as important as the speech itself. The timing needed to sound perfectly natural, as well as to account for a wide variety of different playstyles. I also shouldn't discount the effort that went into creating and tuning all the levels and game content. Amir was tuning all the weapons, creatures, and other aspects of the game pretty much from day one through to the end of the project. He and I were constantly tweaking encounters and level layouts to get them feeling just right.

A lot of the sense of progression in the game comes from acquiring and upgrading weapons, and each of them feels relatively distinct. How did you determine the number of weapons you wanted to include? Were there any weapons or abilities you cut from the final game?

During Bastion's prototyping phase, all of the weapon ideas came together at least in a first-pass form. There was a lot of stuff during that time that got prototyped and ultimately cut just because it didn't feel distinct enough from other ideas or otherwise didn't feel right. We spent a lot of time trying to get the weapons to feel great. In a game like ours where the basic unit of gameplay is a combat interaction, if you don't have great-feeling weapons and combat then you've got nothing. So, our goal was to create as many distinctly different-feeling and satisfying weapons as possible, and to make each of them viable through the course of the game. In a lot of action RPGs, you're constantly discarding weapons in favor of bigger or better ones, whereas we wanted players to get attached to their different weapons in this game. We ended up creating some of our game levels expressly to showcase some of the weapons we had, which might sound a bit backwards but tells you just how important this aspect was to us from a gameplay standpoint.

As an example of something we cut, we had a bullwhip weapon that got pretty far along before we realized it just wasn't working out well and didn't meet our criteria. It fit the theme of the game just fine, but it was shaping up to be sort of a medium-range, crowd-control type of weapon... and we already had the Scrap Musket, essentially a shotgun, which was filling the same role and doing it better. So the only rightful place for the whip turned out to be the cutting room floor. We even prototyped stuff like jetpacks, which neither fit the theme or the level design. We had remote-detonated mines, which were pretty cool but unnecessarily complicated. Some of our design philosophy around the weapons must come from us having worked on real-time strategy games where there's a natural tendency to want to design units that neatly fill specific roles. Having the maximum number of discrete units can make for an elegant game.

While Bastion offers a mostly linear story, there are several elements of the game that offer choice, from small things like how you upgrade your weapons to bigger concepts like the conclusion of the game. Why is choice important in a game like this?

Providing players with meaningful choices around their play experience not only makes for richer gameplay but can make the entire play experience more expressive and rewarding. One of our goals with Bastion was to present the player with an ever-growing series of meaningful gameplay choices. We wanted players to feel a frequent sense of surprise at the expansion of the game's scope. I think this can be a very motivating feeling in a game, as opposed to getting that sense of "OK, I've seen everything there is to see here, so I don't need to play any longer."

We wanted to strike a good balance in this game between, on one hand, having a strong narrative structure where you feel like you're pushing forward toward some sort of meaningful outcome, and on the other hand having a sense of freedom within the world of the game to leave the beaten path. For me personally this comes from trying to reconcile the best aspects of Western role-playing games, known for their open-ended structure, and Japanese role-playing games, known for their crafted (but generally very linear) storylines. I think you can have it both ways.

Our gameplay choices around building up the Bastion and choosing or upgrading weapons, or engaging with the Shrine system that affects game difficulty, are all designed to custom-tailor the play experience to feel very specific to the individual player. Rather than have a bunch of different character classes, we focused around one character so that we could have a meaningful story and a hopefully-memorable protagonist, but we give players a lot of choice in how to specialize that character. We also focused on rewarding experimentation. We didn't want players to stick with the first effective set of tactics they discovered, we wanted them to feel like they could keep trying new things.

To your point, we also introduce some narrative choices hopefully at a time when they're not expected. These are expressive choices that relate to players' emotional connection to the story, if they have one.

Next up: Will we ever see a sequel?

Bastion is the first game from Supergiant Games. How did the team come together? Did you know you wanted to make Bastion even before the team was fully formed?

I met the cofounders of Supergiant Games, Amir Rao and Gavin Simon, while we were working together at Electronic Arts on the Command & Conquer franchise, specifically on C&C3 and Red Alert 3. We all quit one after another in August of 2009, and one month later, Amir and Gavin had dropped everything to move into a house in San Jose where they started working on Bastion. It was just the two of them at first, and the team gradually grew to its full size of seven. While we were thinking about a lot of different ideas in our time between jobs, Bastion was always meant to be the studio's first project. Even the title of the game dates back to the studio's earliest days. It started out as a simple idea – an action RPG where you build the world around you. We wanted to explore some of the positive feelings around returning to town in that style of game, and to make a highly accessible pick-up-and-play style of game that also had some narrative depth to it.

The team is rather close-knit. Darren our audio director and Logan our voice actor have known Amir since middle school. Andrew and Amir also go way back to their days as interns in the game industry. The long-standing trust between members of the team was really important to our development process.

You’ve got a significant background in writing about video games before working on creating them. How did that background inform the creation of Bastion, if at all?

I worked for years as an editor on the media side of the game industry, most notably at GameSpot, but writing game reviews of course is different from writing fiction. I've written fiction in my spare time and always had an interest in it, and I've also studied story structure in preparation for being able to work on game stories one day. Besides which, I was an English major in college and all that. I think my background of playing lots of games, reading a lot, and writing a lot is crucially important to my ability to write for games. But even though I've wanted to make games since I was a little kid, specifically to create worlds and stories and characters, no one was about to just hand me that job in spite of my background as I was getting into game development. They had no good reason to believe I could do the job.

At Electronic Arts I was able to contribute some writing, creating characters and writing unit responses for some of our real-time strategy games, plus I was making headway on story for a project that got canceled. All that was enough to convince some of my colleagues that I had an interest in this stuff, if not a knack for it. And sure enough, it took joining a studio created by my friends for me to finally have my shot at writing a game.

Creating the world of Bastion, the characters and creatures and places and the story and its themes, are the kinds of stuff I've always wanted to do, so this was really a dream project for me in that regard. I also owe a lot of credit to Amir, who served as my editor on all the writing in the game. Between the two of us I think we pushed the writing pretty hard – we took it seriously and tried to make it good by making it a priority, a necessary step in having a decent story in a game.

Would the team at Supergiant Games ever be interested in creating a sequel to Bastion? Are there particular things you’d like to do differently if you ever re-approached the franchise?

One of our design mantras on Bastion was to make a complete game. No half-baked features, no fat. We held back nothing for this game, and approached it knowing it may well be the one and only game we'd ever get to make as a studio, so we had to give it our all. It was very important for us to have the game finish strong, to leave players feeling fulfilled and satisfied at the end. So in that respect, we never really imagined it as the beginning of a franchise or anything like that. It would be a self-contained story and world. We just wanted to make a great, stand-alone game, and we figured we could do it again and again with future projects if we had the chance – we have a lot of ideas.

That being said, and I realize this sounds a bit contradictory, we're certainly not opposed to doing something else in the world of this game. We all really love the world we've created for this game, and especially in my capacity as the writer and creative director, it was very important to me to create the sort of world that felt detailed enough to potentially support any number of different stories.

Without spoiling the game for those who haven’t yet played it, do you have a sense of what percentage of players chose one path or the other at Bastion’s conclusion? To put it another way, what percentage of players unlocked the final secret achievement? Were you surprised by the choices players made?

As of right now, we don't have exact data yet about how the endgame panned out for different players. Anecdotally we have a strong impression that players liked going through the game more than once both for gameplay and story reasons, which is great, because we really wanted to include that New Game Plus mode in there. We've been getting really great feedback about the game ending, including some players telling us that they've sat there paralyzed by it for like 10 minutes or even 30 minutes... that's been really great to hear. We didn't know exactly how people would take the ending of the game, but it's the ending we had intended from relatively early on, and we also knew we were going to set aside an ample amount of time during development in order to execute on it as well as possible.

The choices around the ending are meant to be expressive. Many games offer moral choices but I don't think we go through life making moral choices, or at least that's not how they feel. People make choices based on logic and faith... we don't weigh the good or the evil of our decisions, we just do what seems appropriate and will bring about the outcome we want. So it was important to me to have an ending in which every outcome was fully justified, and that put the essential question of "how will you leave the world" in the player's hands. In any story, for the story to be meaningful, there has to be a dramatic change that occurs. It was very important to me that players feel like they could have a hand in that change, rather than for the change to be strictly dictated to them by the story.