Roguelikes and card battlers have been having a ‘moment’ for years now. Early classics like Hoplite and Dream Quest were ground-breaking in their canny mix of challenge and simplicity. Anything halfway decent becomes common, and these sort of games were something truly special. Lose sight of the lineage, forget the forefathers, though, and you’ll be pleasantly shocked by the quality of recent roguelikes.
Chief among these is the recently-released Dimension of Dreams, standing apart like a very good fusion of the aforementioned games. It takes some familiar mechanics while totally inventing its own card pool and class system, and while the progression is slow and the difficulty perhaps slightly over-tuned, it is great and by all signs will only get better.
Don’t let the 3D art style fool you; this game’s action unfolds within turn-based battles interspersed with overland travel on a 2D grid. Three levels filled with shops, mystery events, treasure chests and monsters. After every battle, you gain experience, gold and most crucially, the chance to add a new card to your deck. That’s the big picture, macro-level strategy of managing health and coins as long-term resources, but the game’s chief challenge lay within those battles. Each turn, the player draws two (or sometimes more) cards and has a few actions to spend playing those cards. There are four color-coded categories of cards: red for offense, blue for defense, green for utility and purple for status effects.
The game’s combos and effects largely revolve around these colorful card types. For now the game has three classes, roughly corresponding to that holy trinity of warrior, rogue and mage. (With the ‘mage’ being more of a ‘holy cleric’ sort). As an example, the rogue-type Demon Samurai has plenty of dodge, critical and poison-based keyword effects.
If all this sounds curiously parallel to Slay the Spire, that is quite accurate and, moreover, quite deliberate. But the variety and detail of both class mechanics and enemy traits are unique to this game and reminiscent of Dream Quest’s utter ex nihilo creativity. In lieu of attributes or spells, player characters have equipment and relics, the former selected from a fixed pool of options every third level, the latter purchased from shops or awarded after boss battles. The synergy between these relics and the deck’s own archetypes becomes a delicate chance for the player to apply their wits. So it possesses the requisite opportunity for a first-rate mind to distinguish itself.
The game’s not a dreamy playpen, though, and has plenty of practical constraints to limit theoretical min-maxing. This tension is necessary because roguelikes are also about building the most elegant, efficient machine out of whatever scraps are available. On this front, the game is a wee bit scanty with its options, I’d say, a little bit beyond the norm of what’s ‘fair’ to make the average run winnable through wise decisions alone. There are so many different viable deck archetypes, but only a few of them thrive in the early stages of the game where it is most crucial to build momentum. So there’s a bit of false richness here, insofar as things like a Parry-based warrior archetype simply isn’t going to cut it at max difficulty, barring some exceptional events.
One more criticism is about those same ‘events’ and the unusually high-variance nature of the level-up rewards. When landing on an event space, a short vignette of text spells out a situation then presents the player with two possible reactions. Some of them are totally binary outcomes, with one ‘right’ choice giving a pure reward and the other pure punishment. This is titillating storytelling but bad gameplay, to present an opaque risk with zero information and no trade-off decisions. And the quality of these boons also varies wildly, from trivial passive boosts to a free level-up. Oh, and upon levelling up, your character can very rarely gain extra draws or energy, which are x10 more effective and useful than, say, three extra max HP. So the game’s not as finely tuned as it should be in these edge cases. Best fix would be to tell the player possible odds upfront for random events and lessen either the swingy wide-ranging level-up options or somehow make them more uniform.
The criticism above sounds harsh & dire, but these deficiencies only stand out because the rest of the game is so nice and polished. There are ten levels of difficulty, with each one throwing a new monkey wrench into things. The scaling here is pretty much ideal; so far, the best I’ve managed is the eighth. (The scalable challenge mode is also pretty much a direct call back to Slay the Spire). There are already three classes to start, with another three in the pipeline and two others coming later down the road. The game is only a dollar because these other classes are locked behind either ‘loot boxes’ or instant payment unlocks.
Fortunately even the ‘box’-based reward system is tastefully implemented. Each class has exactly 50 boxes available for purchase for in-game currency earned after each run, and each box will contain one of those 50 things to reveal. So what is random is the order in which things become available for the player, but everything is guaranteed to become available after the whole lot is purchased. Because the game is excruciatingly difficult right now, this might still mean a 10-hour crusade to unlock the next class without shelling out more cash, but any roguelike worth its salt ought to make 10-hours (5 runs or so) melt away in light of the player’s focus and enjoyment. Flow-state, basically.
Honestly, aside from the wonky difficulty in early stages, the wacky random events & level-ups and rather generic 3D models (which make the game a battery-chugger), the game is a treasure, and one that’s only bound to improve with further balance tweaks and content. Needs a few bugfixes, too. It’s neither wholly derivative nor original but remains a roguelike to watch and try if you’ve the slightest interest, and for this price it’s a steal.