Ten minutes into Epic Mickey's tutorial level, Mickey got stuck between a statue and a wall. No amount of fiddling with the controls would extricate him, so I rebooted. Disappointing.
Later, Mickey stood facing a boulder. This boulder was nearly identical to one I erased with thinner minutes earlier, but no amount of thinner would make this boulder disappear. Confusing.
Moments later, Mickey reached the edge of a platform with a missing section. I aimed my Wiimote and squeezed the B button, but the paint sprayed wide of its target. I moved Mickey closer, aimed again, and missed. I adjusted the camera and pointed directly at the gap. Another miss. Frustrating.
Disney's Epic Mickey is a confounding game. I know because I played it from start to finish. Twice. Why would I do that to myself? It's complicated, but I'll try to explain. The short version: Epic Mickey is a disappointment - and maddening at times - but its vision makes it hard to dismiss.
Part of me despises the game. I'm frustrated by its basic ineptitude at presenting a navigable 3D space. In my boss battle with Captain Hook, the animatronic pirate killed me once; my own mistakes killed me twice; but the infernal camera did me in over a dozen times.
The reviewers were right. Epic Mickey suffers from major camera issues, and Mr. Spector's recent defensiveness on the subject is discouraging. Third-person camera may be "the hardest problem in video game development," but a creative artist makes the hard stuff look easy. Or he conceals it. Or he goes with Plan B.
When I must fight the game's sluggish camera and imprecise aiming system to make Mickey do what the game requires, that's a big problem. Naming it so does not mean I "misunderstand the game." In fact, my understanding and appreciation for how the game occasionally transcends its problems are precisely what kept me playing past its less than stellar first hours.
The game confounds in other ways. Its paint-by-numbers approach to presenting each level denies the player an opportunity to develop strategies through exploration. A new area is introduced, a video tour reveals most of the key items, and Gus tells you exactly what to do. I realize Epic Mickey was designed to welcome newcomers, and that's a praiseworthy goal; but too often it spoon-feeds instructions to the player, diminishing what might have been playfully discovered. Even worse, its quests treat the player like an office intern, running errands, retrieving books, and fetching ice cream for people you barely know.
And that's where Epic Mickey's moral choice system falls down. EM's theme-based worlds are full of NPCs, but they're cut-and-paste versions of characters you meet in the first hour. This is by design, and the game explains why these forgotten characters all look alike; but the result is a universe that feels emptier than its designers intended. Spector's "playstyle matters" mantra only applies if I genuinely care about the world and the outcomes I provoke.
Oswald is a fascinating character, and Mickey's unintended impact on his life makes for good storytelling, but he is the only developed character in the game, aside from Mickey. Unfortunately, the game wants to involve me in the snapshot-view lives of many other characters, most of whom exist only to gate my progress. Emotionally detached from a series of similarly-desolated environments, my moral choices carry little weight. If I save Wasteland and redeem Oswald in the end, regardless of my actions and choices, did my playstyle really matter after all?
The game contains other baffling design choices. Enter an empty house (presented in 2D) and discover nothing to do there. Move to the door, and the game asks, "Are you ready to leave place X?" Yes! What else is there for me to do here? The time wasted clicking through these unnecessary prompts far outweighs the time I might have lost accidentally exiting a few buildings.
Speaking of wasted time, was it really necessary for me to backtrack ten times through the same platforming section separating Mean Street and Ostown? What design goals did such required backtracking serve?
Paint the town to repair damage or restore life, return moments later, and those changes have disappeared. Memory limitations may restrict permanent changes to the environment, but the result is a nagging inconsistency throughout the game. Sometimes Mickey is asked to alter the environment and his actions stick; other times he performs the same tasks, and they evaporate.
The game's internal logic doesn't hold. If a tool produces an outcome, the game world must be designed to enable those outcomes consistently. Epic Mickey wants very much to be a Zelda-like adventure, but its towers and castles lack the internal coherence of Zelda's dungeons. Paint and thinner are marvelously apt "weapons" in Epic Mickey's universe, but too often their efficacy feels arbitrary. They work when and where the game says they work.
So here I am 800 words into a negative assessment of Epic Mickey, and I haven't explained why I also like the game, despite its many flaws. This part is harder to account for because Epic Mickey's charms are so subjective.
I'm drawn to the basic premise of Epic Mickey: a devastated world of abandoned and forgotten cartoons, with the world's most famous animated character to blame. The game does a remarkable job of getting this right tonally. Much has been made of EM's art style, but screenshots fail to convey how unrelentingly dark the game really is.
I found myself yearning for an Okami moment - Mickey's brush magically restores the land to its pristine beauty - but such a moment never comes. The game world feels like an enormous dingy warehouse with makeshift sets constructed to resemble a sketchy memory of Disneyland. For a game chock-full of Disney-ness, Epic Mickey's resolute desolation - we don't even see the sun until the very end - is a wonderful, if disconcerting, surprise.
The game also conveys an unexpectedly bittersweet tone, due mainly to the presence of Oswald. Spector has spoken about his affection for Oswald, Mickey's forgotten progenitor, and that warmth comes through in this game. The final chapters of Epic Mickey mix Oswald's growing admiration for Mickey with his bitter resentment and lingering sense of loss. Nobody will mistake Epic Mickey for The Cherry Orchard, but this game lives in a melancholy place few video games explore.
I also admire Spector and his team for expressing a kind of humane design in the game's combat and interaction systems. Players who don't wish to fight can befriend would-be enemies; flee bosses, and generally avoid good/evil binary options. A single virtuous choice often doesn't exist, and so the player must choose the path that seems best to him. I wish the stakes for these choices were attached to non-trivial situations and characters, but I applaud Spector's effort nonetheless.
Finally, I'm easy prey for stylishness, and Epic Mickey has it in spades. The worn-animation cutscenes are clearly labors of love; Mickey's in-game character animations are fluid and beautifully drawn; and Jim Dooley's score simultaneously conveys and deconstructs the Magic Kingdom's musical vibe. AAA 3rd-party games for the Wii are few and far between these days. Junction Point should be commended for building an ambitious family-friendly game amidst a flood of dreck targeting the same audience.
I wanted to like Epic Mickey more than I do, and I'm disappointed the game shipped with problems that should have been addressed before release. But flawed games can give us much to think about, and I believe that's the case with Epic Mickey. The post-credits ending suggests Spector has left the door open for a sequel. I hope so. I'd love to see him take another swing.