Trace Wysake
PAX West 2016 - James Petruzzi Interview
In which we talk about Chasm.
09.18.16 - 12:11 AM

While it's easy to call your game a Metroidvania, channeling those seminal games for which the genre owes its name is anything but a simple task. And even harder is standing out from an ever-growing crowd of solid tributes and lazy imitations, all of which seem so preoccupied to hit on even just a few key principles that they forget to ask where the genre should or could be heading in the future.

With both Nintendo and Konami out of the game for reasons that range from curious to outright harrowing, the advancement of the 2D exploration-based action-adventure ostensibly lies on the shoulders of indie devs like Bit Kid (formerly Discord Games), a challenge that they're more than willing to accept out of sheer admiration for the very games that inspired them to become game designers. But though the core principles have been long established, capturing the timeless magic of a Symphony of Night remains every bit as puzzling as the dingy recesses beneath Dracula's castle. Not to mention that je ne sais quoi we each attribute to these masterpieces' peerlessness is hardly unanimous, which makes pinning down their exact appeal all the more to process.

On the surface, Chasm stacks up favorably to the classics in terms of its ornately drawn, pixelated look and with its familiar RPG-infused gameplay, but make no mistake: this isn't your father's Metroidvania. Featuring a clever approach to procedural generation, the game promises that no playthrough will ever be exactly the same, where corridors will change more often than the phases of the moon and where secrets are as innumerable as grains of sand. For some, it's been a long time coming for Chasm, which was originally Kickstarted back in 2013 to the tune of almost $200,000, but the wait will soon be over as the game marches towards a 2017 launch window. I spoke with Chasm's director James Petruzzi about the game's development and other choices at PAX West to get a better sense of what the project has metamorphosed into as it enters its final stages before release.

RPGFan: Give me your elevator pitch for Chasm, but you can't use the words "Metroidvania" or "procedurally-generated."

James Petruzzi: Chasm is a 2D exploration-based action-RPG where you must traverse the world and find new abilities to reach previously inaccessible areas. I would say the one thing that differentiates our game from other Metroid and Castlevania games is that it features procedurally assembled pathways between areas. That means when you start a new game, the order that you do things will be the same, but the paths to get from point A to point B will always be a little bit different. Basically, we build every room by hand and our dungeon generator selects suitable rooms from a huge repository to occupy holes in world map, creating a brand new layout with each playthrough.

So how'd I do?

RPGFan: I think loopholes made that answer happen, but I'll accept it. Any how, what I really like about the game is that it's not a total grab. The world design feels very curated.

Petruzzi: Oh, most definitely. I knew if we were going to do this right, I should be able to hand you the controller and you would never know that the game features any procedural generation whatsoever.

RPGFan: That's a good goal! So how exactly do tentpole moments like boss encounters or item acquisition work then?

Petruzzi: They're all pre-planned and will always be in the same order. I knew from the very beginning if Chasm was going to feel like a crafted experience, you can't really change those things up because, like you said, those are the tentpoles holding the whole thing together. There are also persistent areas that have been hand-designed; for example, the tutorial segment, so I can teach players how the game works or introduce new ideas. Same goes for certain story moments, but the main idea is that you're never going to know what's in that next room whether you've seen it before or not. It's variation over complete randomness, which adds up over the course of the experience, even if it's a small change here, a small change there.

RPGFan: One thing I notice in a lot of Metroidvanias, new and old, is an aversion to precision platforming, but I don't get that impression from Chasm. I think the rationale is platforming on top of combat on top of item collecting on top of survival is a lot to expect of players, but I think you guys have struck the right balance.

Petruzzi: It's funny because I'm a huge platforming fan. Probably even more so than the combat, so starting out, I figured Chasm would be 50% platforming, 50% combat, but when we conducted our initial tests, the reception was overwhelming negative. People were saying, "this is way too technical to go from room to room" and "you've made backtracking annoying," and they were totally right, so we scaled the platforming way back. And the other thing is, people don't like precision platforming rooms because those obstacles often replace opportunities for earning experience points or item drops from enemies just because there just isn't enough room for everything without compromising flow or playability. I would say now the ratio is probably more like 25%, 75% in favor of combat, but I'm always trying to figure out meaningful ways to work platforming back in. Right now we're throwing around this idea of "treasure tombs" that are optional challenges with crazy platforming challenges. We're prototyping that now since we're nearing the end of development of the core game and want to include some fun extra content.

RPGFan: In terms of RPG elements, I know you guys are planning a crafting system and some character customization — tell me more about those features.

Petruzzi: Yeah, we're going with systems that you might expect of a Symphony of the Night-style game where you have leveling, you have experience, you have elixirs to permanently increase your health and magic, all that sort of stuff. Then there's interactions with NPCs like side-quests, but they aren't totally focused on you the player character either. They have back stories and problems they're dealing with and we want to incorporate those sort of details into Chasm because we're trying to build a world with some depth and personality. That's hard to pull off though, so even now we're still trying to make sure those moments will resonate with the player.

RPGFan: I literally just came over from demoing the new Dragon Quest game and it always impresses me how effortlessly those games handle flavor text and the characters' personality. The demo had me in stitches within five minutes.

Petruzzi: Exactly. It's so good and paints a picture for you the player almost immediately, and that's really the stuff we're trying to balance now with Chasm. We don't want the game to be totally serious, but we also don't want it to be gimmicky either. Ideally, we want to tap into that cute Zelda-style adventure where there can be darkness and light-heartedness to the story. Our goal is to take players through this emotional roller coaster throughout the game where you hit all these different highs and lows, but it's such a hard balancing act, though I think we're finally at that stage where we've nailed down the vision and now we just have to build out from there.

RPGFan: I can tell you've been itching to use Symphony of the Night as a reference point and have held back, so let's talk about that game for a bit. What's your favorite aspect of Symphony and how is it manifesting itself in Chasm?

Petruzzi: It's got to be the exploration and going back to what I was saying earlier about platforming — I love how Symphony executes those elements in conjunction with one another. Discovery is the key to these games and that's why we've invested so much thought into how procedural generation should work in Chasm because we want that discoverability to be maintained even after you've beaten the game once or twice or three times and beyond. We kept asking ourselves, "how can we make each playthrough feel fresh without disrupting the core experience?" I wouldn't for a second want to change the core experience of Symphony of the Night because it works on so many levels, but I also always wanted to go back and relive those early playthroughs because there was so much excitement with not knowing what was in the next room, which made the discovery all the more worthwhile.

RPGFan: That actually dovetails nicely into my next question: how do you design something worth discovering? Or how do you incentivize trudging through room after room to get to the other side of the castle for possibly nothing?

Petruzzi: I think a large part of that comes from this reptilian part of our brain, this strange ancient mechanism that compels us to collect things, but like you're saying, you may stumble upon a weapon here or there, but you can only acquire so many weapons before the shine comes off the apple, so then you have introduce armor and equipment, but then that can only get you so far, so then there's those elixirs I mentioned, and so on and so forth. So part of the answer lies in having a wide variety of things to unlock, but there has to a balance to it. It has to feel earned because if you're constantly throwing a bunch of stuff at the player, then there isn't much satisfaction in combing through pages and pages of inventory and then you've become numbed to the discovery, in a way. There's even more to it than that, but it's so hard to pin down what makes a game like Symphony of the Night work.

RPGFan: Well, and desirables don't necessarily have to be tangible things like weapons or armor. In my opinion, the strength of Symphony comes from small touches like that room in the library that has enemies entirely based on Wizard of Oz characters or how some items like those peanuts have hidden mechanics associated with them. Those things are desirable in a way that defies practicality, you know? Might we see some of these small touches in Chasm as well?

Petruzzi: Right, those are exactly the sorts of things that game designers, or at least me anyways, want to include even before we've actually conceptualized gameplay or story or anything else really. It's hard to resist, but you have to accept that those details are actually what you should focus on last. There are so many other things that have to be prioritized first because there has to be an actual game for those moments to really shine. Fortunately for us, we're at or very, very close to the fun stage of the project where we can include all those embellishments you're talking about, like "what can we include in the background of this room?" or "what should this character say when this happens?" or "what kind of mini-games would work here?" and personally that's really the stuff that excites me. I can almost guarantee you those things you just mentioned were added in at the very end of Symphony's development.

RPGFan: Oh, without a doubt. I think I've heard or read somewhere Igarashi himself admitting as much.

Petruzzi: Doesn't surprise me at all, but yeah, those tiny details are always brewing in my head and you're totally right when you say those are the things that players want to discover even after they've beaten the game several times. They're very much a part of the fabric of what makes Symphony so special in my eyes.

RPGFan: There's this excellent article out there that attempts to explain what makes Symphony of the Night's level design so memorable and it suggests that the economy of regular and irregular-shaped rooms plays a huge part in this. In essence, the article asserts that a regular room could be a variety of shapes, but is mostly rectangular or horizontally oriented, whereas an irregular room is more asymmetrical and vertically inclined, like the giant staircase. I'd imagine procedural generation complicates this to an extent, but how integral is this concept in the world design of Chasm?

Petruzzi: Well, it's kind of funny because I've never really heard anybody talk about this, but it's something the team and I had to discover on our own through play-testing. We have a couple of names that we use for rooms like those horizontal ones with the enemies in a row, we call those hallways, and then you've got vertical rooms that are like towers that connect up to hallways, so that sounds pretty similar to what this article is saying.

Here's the problem: our early builds of the dungeon generator didn't assign weight or preference to a given room, so we would end up with a series of tower rooms one after the other or long stretches of hallways with nothing to break them up into meaningful gameplay chunks. Then the challenge was to figure out how to piece these vertical rooms and horizontal rooms together, so that's when we came up with the favoring system where the dungeon generator favors a set range of rooms that would best complement the preceding area and what comes afterwards. Once that happened, the game really started to feel like Symphony of the Night because it had a nice amount of variance that actually felt hand-designed.

RPGFan: Again, it's that carefully crafted, curated feel versus complete randomness like you see in other games.

Petruzzi: Yeah, when it's all tossed together like that, the gameplay suffers. What I would do is pour over Super Metroid maps and at first I was like, "this is all just spaghetti," but the thing with that game is, yes, it does look like spaghetti, but it's working on a whole other level of design where there's all these little tendencies everywhere that create something that's so much greater than the sum of its parts. It really is amazing.

One of the inherent flaws of Chasm that I will readily admit is that it's difficult with the dungeon generator to do loopbacks, meaning you can only go down this hallway one way until you come back around later in the game and traverse it through different means and that's something Super Metroid does so well. That's just impossible to generate with any success, so we have those sections in Chasm, but they're fixed in every playthrough. What that does is give us greater control over the level design since we're doing those parts by hand, and hopefully that on top of the procedurally generated elements it will help maintain the game's freshness in a satisfying way.

RPGFan: Honestly, you can't ask more than that from a Metroidvania, in my mind.

Petruzzi: Yeah, and that's why it's taking us so long to wrap this project up because, yeah, we could have just thrown some stuff together and you'd end up with some spaghetti dungeons with no direction, and ultimately they just wouldn't work as a whole. I don't know if you noticed this, but, in the demo, there's always a campfire on your way to the next tentpole moment, which is an implicit way of expressing to the player that they're heading the right way without telling them "go this way." That's what gives the game its pacing and we decided very early on that there's no way we could just toss rooms together and have ourselves a real Metroidvania. It just doesn't work. You'd never feel like you have a goal, you'd have zero sense of direction, and you would get bored quickly if there's non-existent pacing.

Pacing was another challenge we faced with this game. We have a number of sub-areas and each consists of about ten to fifteen rooms; those rooms have to be self-contained and they can have any number of exits to other places, which connects it all together. That was definitely the toughest thing about this project, and once we determined the rhyme and reason behind how things need to work, we had to scrap most of our progress. We already said in our Kickstarter that we'd have six areas and that's never changed during development, it's just been a question of how do these areas work individually and how are they interconnected to the other areas? Or how do you keep the difficulty curve steadily increasing at a balanced clip because otherwise the difficulty is going to be insane, but more importantly unwarranted. Once we broke down each area into measurable building blocks, we could then nail down where and what kind of enemies would be placed in each room. That's when the game really began to take shape and for the better.

RPGFan: I think this will be the final question. So, I'm an abuser of this myself, but I feel as though the term "Metroidvania" has become increasingly trite recently since it's the trendy genre nowadays.

Petruzzi: It's like the Dark Souls of terms.

RPGFan: It's like the Dark Souls of terms! Yes, thank you! But it also seems like it would be hard to make a game where the expectations are right there in the genre's name. People hear it and expect Super Metroid or Symphony right off the bat, so how do you go about subverting those expectations? How do you push this genre forward if it's so often clinging to the past?

Petruzzi: I would say, on the flip side, there were games doing the same thing before those two games ever existed.

RPGFan: Oh yeah, we could rattle off games all day. There's The Goonies on NES, and Zillion for the Master System—

Petruzzi: And Todd's Adventure In Slime World totally had that exploration aspect too, so there were things that beat both Metroid and Castlevania to the punch.

RPGFan: Those two sort of calcified it into a proper genre, and certainly the genre that we know today.

Petruzzi: Yeah, and I personally don't think the word is all that great either, but what it does invoke is a very specific style of game, and until someone comes up with something better or with less syllables, people will find value in it. If I say something is a 2D exploration game, that could mean almost anything, and what is that actually going to mean to people? Whereas with Metroidvania, people understand right away that there's going to be power-ups, backtracking, RPG elements, and all that. It just has a very specific connotation that's conveniently expressed by merging those two series' names.

It's something I actually thought about a lot during the marketing phase of planning for the Kickstarter. Like, can I actually use this term? Both series are copyrighted, so are we allowed to use it to describe our game? But ultimately it's the best option for describing this one type of experience and people immediately understand it. That said, I'm hoping people kind of move on from using it, but I think it has a lot to do with whether or not more games will come out that are like this and where they'll take the genre. I mean, think back to when all first-person shooters were called "Doom-clones." Now they're just called FPSs because people took the genre in totally different directions, so now nobody calls them Doom-clones anymore. I think there's some growing pains involved, but we're getting there with Metroidvanias, or at least I hope so.

RPGFan: For the record, I think your team is doing a great job of taking the genre in a new direction while honoring what it takes to make this type of game in a post-Symphony of the Night world. This whole curated approach to procedural generation is refreshingly novel.

Petruzzi: Thank you. We're all big gamers, so we knew that we had to break new ground where we could in order to take this idea of variation somewhere because games, especially those in this genre, are too often static all the way through and are always exactly the same, but at the same time, you don't always want randomness with no structure. You'd never watch a movie without a core plot that doesn't go anywhere. You want something that takes you on that emotional journey. What I'm talking about has to be human-made at this point, and I can't foresee a future where a robot will ever make a game meaningful to a human without that human element — you need that human touch. I think we've found a balance in Chasm where we've got that non-static, dynamic quality to the world that, at the same time, has that structure it needs to be a compelling package from start to finish.

We at RPGFan would like to thank James Petruzzi for taking the time to answer our questions. We look forward to hearing more of him and the Bit Kid team in the future. To learn more about their current project, Chasm, please visit their website at http://bitkidgames.com.


Coming Up on Twitch

Star Ocean: The Second Story
Sunday, October 22 • 10am PDT/1pm EDT

Final Fantasy XII: The Zodiac Age
Mondays • 6pm PDT/9pm EDT

Vandal Hearts
Tuesdays • 12pm PDT/3pm EDT

Tues-Wed-Thu and Saturdays • 7pm PDT/10pm EDT

Featured Content
Guild Wars 2: Path of Fire Review
Guild Wars 2: Path of Fire
Retro Encounter 105
Retro Encounter 105
Fragments of Him Review
Fragments of Him
South Park: The Fractured but Whole Review
South Park: The Fractured but Whole
Etrian Odyssey V: Beyond the Myth Review
Etrian Odyssey V: Beyond the Myth
Pillars of Eternity: Complete Edition Review
Pillars of Eternity: Complete Edition