Record of Lodoss War: Deedlit in Wonder Labyrinth

There is a particular kind of licensed game that exists to be forgiven. It coasts on nostalgia, ships with a stiff animation budget, and gets graded on a curve by the fandom it was aimed at. Deedlit in Wonder Labyrinth is not that game, and the gap between what it could have been and what it is remains the most interesting thing about it.

Developed by Team Ladybug with WSS playground and published by WSS playground, this is a 2D action-exploration game built around Deedlit — the high elf from Ryo Mizuno’s Record of Lodoss War, a property that has been anchoring Japanese high fantasy since the late ‘80s and hit Western consciousness through the 1990 OVA. Mizuno supervised the project directly, and it shows in the narrative positioning: this isn’t a greatest-hits retread but a bridge story, filling in untold events leading into Diadem of the Covenant. Deedlit wakes in a labyrinth with no memory and no explanation. You explore to find out who put her there.

Commercially, it landed in an awkward but advantageous spot when it launched on March 27, 2021. The Metroidvania space that year was saturated with long indie epics competing on scope. Team Ladybug went the other direction — a tight, short run, elegantly signposted, with a boss rush and higher difficulty tiers bolted on for replay value. That decision defines both its reception and its ceiling. The game has drawn a warm player reception and a cooler, more divided critical one, with the argument centering on the length rather than the craft. Both camps are looking at the same game. They just weigh brevity differently.

• Spirit-Swapping (Sylph & Salamander): The core system. Deedlit toggles between a wind spirit and a fire spirit mid-fight. The active spirit alters her attributes and slashes incoming damage of that element — swap into Salamander when a fire boss winds up, swap back for Sylph’s mobility.
• Spirit Levels as a Healing Economy: Killing enemies levels the active spirit. Max it out and Deedlit regenerates HP. Health isn’t a resource you hoard; it’s a resource you earn by staying aggressive and switching correctly. This is the single best design decision in the game.
• Melee and Ranged, Freely Equipped: Swords for close work, bows for distance, with new weapons scattered through the maze and swappable on the fly. No skill-tree gating, no crafting tax.
• Interconnected Labyrinth with Ability Gating: Traversal upgrades retroactively open earlier rooms — genre orthodoxy, executed cleanly, with a map built to respect your time.
• Presentation Package: Hand-drawn pixel art and animation at a level most studios can’t afford, animated cutscenes, and a full orchestral-style score.
• Replay Modes: Boss rush plus higher difficulty settings.

The Gameplay Loop

Most Metroidvanias treat elemental affinity as a lock-and-key nuisance: fire enemy, bring fire sword, move on. Deedlit treats it as a rhythm instrument.

The Sylph/Salamander toggle is bound to a single input and carries no cooldown penalty worth respecting, which means the game is built to create encounters that expect you to flip states two or three times inside one exchange. A boss telegraphs a fire sweep; you switch to Salamander, eat the hit at reduced cost, switch back to Sylph for the movement advantage during the punish window. It’s the defensive logic of a parry system without the twitch tax of a parry system, and it makes the game legible to players who bounce off Souls-adjacent timing.

Then the healing economy layers on top and quietly changes everything. Because damage-dealing with your active spirit charges it toward a max-level HP regen, the optimal play is never to turtle. Retreat and you starve. The system creates forward pressure without ever writing “be aggressive” on the screen — it just makes cowardice unprofitable. That’s real design confidence, and it’s the reason the combat holds up across a full run despite a relatively shallow move set.

Where It Flattens Out

The bosses are the soft spot, and it’s not close. Encounter design is readable — patterns are clear, tells are honest, and the elemental swap always has a correct answer. The problem is that the correct answer is usually the only answer, and it’s rarely hard to find. Once the swap logic has been internalized in the opening stretch, most fights resolve into pattern recognition rather than pressure. Players coming from Hollow Knight’s Pantheon or the harder corners of the genre will likely clear most bosses on the first or second attempt. The higher difficulty modes address this by raising numbers, not by rewriting behavior, which is a patch rather than a fix.

Progression has a similar softness. The labyrinth is described as sprawling and interconnected, and structurally it is — but the critical path is considerably more linear than the marketing implies. Ability gates open in a near-fixed sequence. Sequence breaking is minimal. You will rarely be lost, which is either a mercy or a missing organ depending on your tolerance for cartography. Compared against the 1997 Castlevania that named the genre — the benchmark critics reached for repeatedly — this is a guided tour of a labyrinth rather than a labyrinth.

Onboarding and Friction

Here’s the counterargument, and it’s strong: this is one of the most approachable good Metroidvanias of its generation. Onboarding friction is close to zero. There’s no resource management, no punishing death loop, no obscure item interactions, no wiki dependency. Weapons swap instantly. The map is honest. A player who has never finished a game in this genre can finish this one, and finish it feeling competent rather than coddled.

Against the short runtime, that accessibility reads as intent, not compromise. Team Ladybug built something with almost no dead space — no padding corridors, no backtracking-as-content, no fetch quests inflating a completion counter. Every room does a job. The trade is that when the credits roll you’ll want more game that doesn’t exist, and the boss rush isn’t a substitute for a second act.

The Story Layer

The amnesia framing is the oldest trick in the drawer, and the game knows it — it uses the setup as a delivery mechanism for lore rather than as a mystery worth solving. For Lodoss readers, the payoff is real: Mizuno’s supervision produced continuity that fits, and the connective tissue to Diadem of the Covenant lands. For everyone else, the narrative is pleasant background radiation. Neither audience is badly served, but only one is actually being courted.

This is the part where the game stops arguing and just wins.

The pixel art is exceptional — not “good for indie,” not “charmingly retro,” but among the best sprite work of the last decade. Deedlit’s animation set carries an absurd frame count for a game this size: hair and cloth react to her movement, the spirit swap has real visual weight, and enemy tells are readable from silhouette alone. The art direction isn’t chasing a 16-bit pastiche either; it’s using modern resolution and color depth to draw something the SNES could never have rendered, which is the correct way to use the medium.

The animated cutscenes are competent rather than remarkable, and they know to stay brief. The score is the honest surprise: full, orchestral in texture, and mixed loud enough to carry rooms that would otherwise feel empty.

Technically, the design leaves little room for trouble. The engine is doing nothing exotic, and a 2D game of this scope is light on hardware — the swap mechanic is built to feel instant, with no visible negotiation between input and action.

Nintendo Switch is arguably the better fit, and the critical split between versions mostly reflects that. Not because of raw performance; both builds are designed to be stable. It’s a fit question. A tightly authored short action game with clean checkpointing and no resource management is close to the platonic ideal of handheld play, and the pixel art is made for the OLED panel. Docked, it should hold up fine, but this game wants to be held. The Joy-Con d-pad situation is the usual Switch caveat — a Pro Controller or a third-party d-pad face is a meaningful upgrade for the swap-heavy fights.

Steam is the more configurable version and the one to buy if you own a good pad. Play it with a controller; keyboard-and-mouse works but the spirit toggle plus directional bow aiming is clearly designed around analog-and-shoulder-buttons ergonomics. The PC build spent time in early access ahead of its 1.0, worth knowing before reading any early impressions as a verdict on the port. Uncapped frame rate and resolution scaling are yours here; on Switch they aren’t.

Cross-buy doesn’t exist between them. If you want both, you pay twice, and there’s no save transfer — a real annoyance for a game whose ideal use case is handheld-primary with desktop replay.

Deedlit in Wonder Labyrinth is a game that made a choice most studios are too anxious to make: it decided how long it should be and then stopped. There’s no padding here, no artificial difficulty spike to inflate playtime, no backtracking dressed up as content. What’s left is a short run of some of the finest 2D craft in the genre, wrapped around an elemental system that’s smarter than the game ever bothers to brag about.

Its limits are real. The bosses need teeth. The labyrinth needs more corners you can get lost in. When the credits roll you’ll feel the absence of a second act that the runtime never had room for, and at full price that will sting some buyers more than others.

We rate it a solid, worthwhile effort with real flaws: the craft is beyond dispute, but the bosses lack pressure and the structure is more guided tour than labyrinth. Buy it on sale if you’re price-sensitive, buy it on Switch if you have the choice, and don’t expect it to keep you busy. Expect it to be good.