The first thing you’re likely to hear about Chasm is that it took approximately six years to develop, but you really can’t make any inferences about its quality based on that. A hugely delayed release can speak to a developer’s passion and perfectionism just as readily as to their ineptitude. In this case, the wheel landed resolutely on the side of passion, but in a cruel twist of irony, that wasn’t enough to make a good game. The primary gimmick of Chasm – a procedurally generated Metroidvania that’s only generated once, at the beginning – is so inherently flawed that it doomed the product to mediocrity from its inception. The team at Bit Kid, Inc. has spruced up this problematic core admirably – enough to make it rather depressing that their love and time wasn’t spent on something more deserving.
The strangest thing about this situation is that it should be incredibly obvious to anyone familiar with the genre why this setup wouldn’t work: Metroidvanias live and die by their level design. Sure, procedural generation is conducive to their exploration component, but so many other things are sacrificed to accommodate it. The ability to carve out shortcuts and return paths using newly discovered equipment is absent, making gameplay a backtracking extravaganza. Furthermore, in order to facilitate stitching individual hand-made rooms together, most of them just end up as boring hallways with an enemy or two in the middle. And if you thought the difficulty curve of similar games was easily ruined with the acquisition of optional power-ups, imagine what it’s like when there’s no structure to that acquisition.
As I said, Bit Kid has done what they could to make the concept work. For one thing, save points appear to be hard-coded to exist equal distance apart from each other. They’re also definitely hard-coded to appear just before bosses – something even Dark Souls, in all its supposed “tough but fair” glory, couldn’t figure out. But this is like hoping a light rain will put out a forest fire. Any controls exerted to keep the random elements in check only serve to shrink the adventure down to a pattern to be iterated a dozen or more times. It’s longer than most indie titles, taking between 10 and 15 hours to beat, but after scouring the same corridors repeatedly, you’ll be ready for it to end well before then.
What variety there is comes from the enemies, of which there are over 80 types. This is clearly where some of those years of development were spent, and the game is definitely better for it. There is surprisingly little overlap between enemy behaviors. Even the rare palette swaps will come equipped with a different weapon. The word “pattern” still taints the combat, as until the final areas, most foes only have one attack at their disposal. But even then, learning and exploiting their patterns is pretty satisfying. The controls play into this in an odd way. They’re a little stiff, but it’s in that intentional, “every action must be deliberate” way that Castlevania itself pioneered. So while it may be instinctual to treat the “back dash” as a standard dodge, it’s really meant for specific times and places.
Contrasting the enemy diversity, the player’s moveset is severely limited. Offensively, your options are a primary attack and one of six purchasable spells, all of which start out so weak, situational, and magically draining that they’ll barely be touched. There are a handful of weapon types that slightly change how the main attack functions, but the inability to switch weapons without pausing weighs the selection heavily toward all-around choices like swords and maces. It’s a blatant missed opportunity, because a) spells have no such restriction, b) the two trigger buttons go unused, and c) it would be completely painless to implement. It’s also enormously disappointing that none of the exploration equipment have any battle application. In fact, some of these key items were such obvious afterthoughts that they’re only used in a single spot each.
For the first half of the game, difficulty is decidedly old-school – pushing to the border of unfair design but never crossing into it, before finally relenting as I punched through to a save point. Then two things happened. First, I unlocked a wall-jump that’s about as easy to use as the one in Super Metroid. But where that game treated it as an optional technique primarily used by speedrunners, Chasm devotes long strips of required map space to it, often demanding impossible precision. On the other hand, I also discovered how to upgrade spells, which I highly discourage for anyone satisfied with the existing challenge, as it totally inverts magic’s initial weaknesses into strengths. The final upgrades are so overpowered that I was able to dispatch every late-game miniboss before they could even launch an attack.
The spells are also emblematic of Chasm’s derivative nature. Upon seeing that one of the spells was a “magic axe,” I correctly guessed that it would be thrown in a parabolic arc, and that predictability is rampant throughout the campaign. Even the story, while serviceable, doesn’t have a single surprise in store for its audience. Aside from the stillborn procedural generation, this is just a by-the-numbers Castlevania imitation. I have a feeling this is why the long development cycle is such a sore point for some; it’s easy to look at how commonplace everything here is and wonder why it required six years to complete.
The answer, besides the extensive bestiary, is in the presentation. The sprite art and animation are superb, the music is enjoyable if a little repetitive, and the sound design is practically perfect. While all of these things are certainly nice to have, they’re not exactly top priorities. In the end, I’m just left wondering why Chasm exists. Even if the semi-random level design worked with this genre, what would it accomplish? Make each campaign slightly different and troll walkthrough writers? Metroidvanias are a pretty intense undertaking, so you won’t often hear one described as “disposable,” but that’s exactly what this one is.
This review is based on the PC version of the game. A copy was provided by Bit Kid.