On paper, Chasm's pitch is compelling: a never-ending Metroid, where every experience is different. It's not totally random, as the game is pulling from a pool of hand-crafted parts to build the world, but the way it links them changes every time. Chasm's problems are myriad, extending far beyond its approach to level design, but even brushing them aside, it never manages to sell you on its most basic pitch. Rather than creating a sense of mystery, Chasm's randomness leaves a giant, frustrating question mark over your playthrough. Is this level actually boring, or did I get a bad roll of the dice?
One reason fans used to beg Nintendo for more Metroid was supply and demand; no one else was scratching that particular itch. That's no longer a problem in 2018, and arguably, too many developers are trying. But that's true for lots of genres these days, and so the problem facing developers is standing out in a crowded market.
The setup for Chasm is familiar. You're a young knight tasked with cleaning up a monster problem in the mines of a nearby town. Villagers have been kidnapped, and rescue requires diving into an underground labyrinth with... secrets. Each area has a different theme, and the requisite shuffling of enemy types, dangers, and bosses. As you gain new abilities—the ability to grab onto ledges, or slide through tight spaces—you'll want to return to old areas to uncover hidden treasures and passageways. You swing a variety of swords and maces, and occasionally cast a spell. Upgrades! New equipment! Loot drops! If you've played one of these games, there's little about Chasm will surprise you, in this regard. It's fine, you know? But there are lots of "fine" games like Chasm now, which means Chasm finds itself heavily relying on one unique asset.
What Chasm attempts isn't as ambitious as a game like No Man's Sky, which uses procedural generation to generate a near-infinite universe impossible for a small team to build, but the pitfalls are similar. The advantage of randomness is obvious: What the algorithm generates is unexpected. The downside of randomness is equally obvious: It might spit out something terrible. The upteenth boring planet was one of the (many) reasons I gave up on No Man's Sky after a dozen or so hours, but at the very least, giving it another try was easily solved; you could quickly shoot back into space and find another planet to explore.
The cost of re-rolling the algorithmic dice in No Man's Sky was small, in terms of the amount of time spent going from one planet to the next (a few minutes), the in-game resources required to turn the ship on (minor), and the sunk cost of having explored a boring planet (usually fairly obvious after you've arrived). Chasm, on the other hand, is placing much greater psychological burdens on the player for a re-roll.
It was an hour or so into Chasm when my yawns and pauses to check Twitter suggested a problem: boredom. The controls felt stiff, enemies weren't engaging to fight, and the dungeons showed zero personality. (One of the game's key moves, a backwards dash, felt so uncomfortable to use, I basically abandoned it.) But I've played plenty of games, even great ones, that start slow, and eventually find their footing. Maybe Chasm would, too? After nearly four hours with Chasm, with the game reporting I've explored 47% of the world, precious little has changed. The controls still feel stiff, enemies are nothing but fodder, and I can't tell you anything memorable about the places I've fought through, outside of one of them having lots of vines to climb on. That's a bad sign!
Typically, the conclusion from here is simple: Chasm is poorly designed. But Chasm's reliance on algorithmic design suggests an alternative: My version of Chasm is poorly designed, and if I were to simply start the game over, maybe I'd stumble upon one that was more interesting. Perhaps another roll of the dice would present a more varied set of platforming puzzles, weapon drops, and enemy layouts. Chasm doesn't make it clear what, exactly, is influenced by the dice roll, so you're left to wonder. In my case, it meant wrestling with my natural inclination to keep giving the game a chance, hoping it would get better along the way, before realizing this wasn't the case. When this realization dawned, however, I was hours into the game. My pursuit of a better Chasm would mean tossing away hours of time invested in this one. And what if the second one sucked, too? The third? The fourth?
I doubt Chasm was meant to be played this way, with players starting new games over and over, seeking the best variant. It was probably meant to instill conversation among fellow Chasm fans, as they shared unique discoveries in their playthrough, and give hardcore fans a reason to play the game multiple times. (The ability to share your world with other players by sharing a code is an inspired touch.) As someone who tends to only play games once, it all backfired; the regret over not having started over sooner weighed on me, and eventually, the sunk cost over that choice lead me to put the controller and move on from Chasm entirely.
Spelunky, one of my all-time favorites, operates similarly to Chasm. Every "run" in Spelunky is different, but you traverse the same world types every time, and accomplish the same primary goals to reach the finish line. But finishing Spelunky takes all of an hour and change, even if you're seeking the deepest secrets. Most Spelunky runs end (in failure) much sooner, making it hard to begrudge a particular layout for very long. It helps, of course, that Spelunky is a joy to play, moment-to-moment. Chasm does not not have this luxury.
It was impossible to play Chasm without thinking about Hollow Knight, as well. Due to my own proximity of finishing (and loving) Hollow Knight, it's been on my mind a lot lately. And though everything about Hollow Knight is meticulously hand-crafted, with seemingly no aspect of design left to randomness and chance, that's not entirely true. I've been meaning to write about Hollow Knight but haven't found the time, so I'll leave my central takeaway here: It's a game that puts absolute, absurd trust in the player to find their own path forward.
"Chasm's reliance on algorithmic design suggests an alternative: My version of Chasm is poorly designed, and if I were to simply start the game over, maybe I'd stumble upon one that was more interesting."
When we say the term "open world," what do we typically mean? A giant map with lots of quests, right? I'd argue Hollow Knight has an open world, too. When I shared my nightmarish encounter with the game's bug-infested Deepnest area, I quickly heard from a bunch of people who didn't stumble upon this virtual Hell until the end of the game. It happened to me right after I started playing! This notion of finding (or missing) things in vastly different orders became a common refrain among Hollow Knight players I spoke with. Though Hollow Knight has a progression, it lets the player advance in such disparate and divergent ways that it's easy to become convinced you're having the canonical, developer-preferred playthrough, when, in reality, they've set up the world in such a way that you can chart your own path.
That's clever, and gets at what Chasm tried to do from another, more successful, angle.
There's every reason to think a game like Chasm can work. No Man's Sky, for example, was rightfully dinged for how often its procedurally generated worlds were boring, but a few years of updates and many tweaks of the algorithm later, No Man's Sky is almost a completely different game. I'm not sure if Chasm will get that opportunity, but its failures aren't an indictment of what it tried to do. It means this attempt didn't work.
Follow Patrick on Twitter. If you have a tip or a story idea, drop him an email: firstname.lastname@example.org.