At first glance, Abyss Odyssey’s combat appears slow and sluggish. The platforming feels floaty and awkward, and the enemies lumbering; the bosses frustrating in their well-timed blocks, dodges and carefully considered attacks. It’s a bad first impression that might have been alleviated with a better tutorial.
But, once one comes to grips with the super cancel mechanic (the basic three part attack may, in this way, be combined with special skills; and thus long, complex combos performed), combat is transformed: the sluggish, ponderous pace of the combat becomes fluent: a transition between blocking, dodging, combos and mob control: Abyss Odyssey often walls the player in in cramped, trap-filled environments where the player may use these traps to their own advantage; and so, subtly, herd their adversaries into the traps—and, even if the environments are trap free, dividing multiples of mobs into singles is often necessary; or singles into mobs, instead, to devastating effect.
These environments are procedurally generated. As a result, the Abyss lacks the geographical quality of Zenozoik (the setting of the Zeno Clash series, also developed by ACE Team) and, worse still, rather than successfully creating a more heightened sensation of variety due to the controlled-randomness of procedural generation, the controlled part of the randomisation results in a fast-growing ennui of repetition.
Perhaps this is simply a psychological exaggeration engendered by the expectations of procedural generation, but when one wanders into repeated environmental templates—or similarly shaped level layouts—it’s harder to dismiss these incongruous consistencies than if they had been handcrafted and, thus, imbued with a more human quality; or a more tangible sense of sentient purpose. Secret areas, due to the procedural generation, are predictable in their placement: which robs them of some of the satisfaction of their discovery. But these secret areas are initially vitally important in a deep descent into the abyss: stockpiling weapons, healing items, or indispensible merchants dealing in all of the above; but once one acclimates to the mechanics, such material fancies swiftly become superfluous; even on Nightmare difficulty.
Not only are the individual environments procedurally generated, but so is the difficulty: this means that, if lucky, one might have a line of levels leading directly down to the warlock (the final boss; you’re in his dream: you’re a projection of his dream) or, if unlucky, nearly every level might be hard; though, usually, there is a path through the abyss which will provide the player with a rising challenge, as the player descends. And the luck and unluck is a reversed sensation as experience mounts.
The inhumanity of the procedural generation is instantly overcome in co-op. Although, at first, the awkwardness of the friendly fire (which can be turned off, but with the qualifier that one won’t qualify for the leaderboards and, if one defeats the warlock, his death will have no effect on the endgame goal: the warlock’s ultimate demise and thus his subsequent replacement with another dreamer) means that, especially in a cramped trap, the two co-operative players will often be as dangerous to one another as the enemies. But, as one begins to understand the movements of one’s partner, large mobs may be even more easily divided into more manageable numbers; and the lifeless quality of the environments is thus diluted by the spontaneity of another real, live human being jumping through the repeated templates; and bouncing enemies back and forth with one in a sadistic game of catch.
Unfortunately, the netcode is, at times, ugly: characters sinuously sink through the ground and spasmodically teleport instead of jump. Despite these aesthetic instabilities, the latency is presently, after patching, perfectly adequate for deep excursions into the abyss; interfering with, but not crushing, the precision required of the mechanics and the broader design of the combat—the AI is adept at blocking, timing attacks, and sometimes flanking. However, intermittently the epileptic fits and permeable levels render online co-op unplayable. But, the promise of the initial patches instils a degree of confidence. Even so, for the timorous, co-op may still be more easily enjoyed locally; even on PC: which is worthy of praise in and of itself—and that’s not to mention the local PvP mode, that the PC version also features.
The enemies also, on and off, inexplicably freeze or, as if petrified, lock themselves in an unending block; sometimes halfway across the screen in abject safety from the assault of the player. And the sound effects occasionally cut out. Both issues that, hopefully, will be patched in the future: already ACE Team have showed the potential for long term support, with the addition of a harder difficulty (which significantly improves the AI and makes combat more complex: prepare to be flanked relentlessly) and the option to turn off friendly fire; as well as the aforementioned patching of the netcode.
While the procedural generation has stripped the environments of much of ACE Team’s geographical sensibilities, the enemies remain just as rich and eccentric: large Father-Mother-like peacocks, floating Mortal Kombat monks, and ents are all animated with a Santiagan sense of style, and blessed with (sometimes drastically) unique sets of moves; which is important, because one can capture these monstrosities and then transform into them at will: effectively providing one with a third chance; a third continue.
A third continue because the second continue comes in the form of a soldier after one’s chosen character (there are three to choose from; two to unlock: all of who play vastly different to one another) dies. These soldiers are, essentially, more limited versions of one’s chosen character; and if one can make it safely to one of the shrines strewn about the abyss, then one’s chosen character will be resurrected: it’s only a dream, after all! Sort of. And the shrines are also used for the purpose of optional checkpoints and the rejigging of one’s special skills: three may be equipped (and upgraded with skill points three times), and one must juggle upgrading one’s skills and using skill points to fuel up to three super cancels in a row.
Procedural generation does not merely inhibit the aesthetic quality of Abyss Odyssey, but also affects the design quality of the levels. Beyond the randomised difficulty, some of the individual levels come out as completely bland and uninteresting templates which means that, if unlucky, one’s excursion into the depths of the abyss may be completely devoid of novelty or excitement as one trudges through a level labelled as hard but blessed with scarcely a trap or enemy in sight.
Nevertheless: post-mechanical epiphany, the tactile weightiness of the combat is immediately intoxicating; but how long will this intoxication last when the procedural generation can simultaneously dilute and amplify (how easy it is to be entranced in a pattern-less paroxysm of a fever dream) the sense of exploration and discovery; of variety? The cast of warlocks, and the speed with which the community can despatch them will presumably prove vital—and there are a few moments where the procedural generation shines unblemished: such as the unexpected discovery of a huge, terrifying, secret boss; or an encounter with the devil. More of this in the future, please: Abyss Odyssey needs more handcrafted content; not the quick-fix solutions to long term viability such as the proposed (even) harder difficulty—but apparently these will come, too. The question is: soon enough to sustain the community?