Tears of the Kingdom: The Physics Engine as Permission

Nintendo shipped The Legend of Zelda: Tears of the Kingdom on 12 May 2023, for Switch only, as a direct sequel to Breath of the Wild rather than a new engine or a new Hyrule. Same map, mostly. Same physics rig underneath it, mostly. The thing that changes everything is a set of four new verbs — Ultrahand, Fuse, Ascend, Recall — and the fact that Nintendo built them as a general physics toy that the whole game then gets built out of, rather than as scripted set-piece tools. That’s the interesting design story here: the same world, made out of parts you can pick up.

What Ultrahand actually lets you do

Ultrahand grabs any physics object in the world and lets you glue it to any other physics object. A plank to a wheel. A fan to a raft. A rocket to a shield, angled downward, so that firing it launches you into the sky. The build sticks together with a rough, slightly wobbly joint logic that clearly runs on the same rigid-body solver as everything else in the game — there’s no separate “vehicle mode,” no menu of prefab machines. You’re manipulating the same physics objects the game uses for its rockslides and its rafts, and the game just lets you keep gluing them together past the point most designers would call a build “done.”

This is the actual design bet, and it’s a bigger one than the marketing beat around it (the Twitter clips of jet-propelled tanks). A game that gives players a general-purpose physics toolset is a game that has given up controlling the solution space. Breath of the Wild already flirted with this — bomb-and-metal-object physics puzzles, ice-block towers — but Tears of the Kingdom commits to it as the whole verb set. Shrine puzzles are built assuming you’ll solve them with a plank and two fans in a configuration nobody on the design team pre-tested. That’s a genuinely unusual thing to ship at Nintendo’s scale, where playtesting budgets exist specifically to catch the unintended solution and patch it out.

Fuse is the quieter half of the same idea, applied to combat instead of construction: any material object can be welded onto a weapon or a shield, changing its stats and often its entire function — a boulder on a stick turns a twig into a sledgehammer, an eyeball fused to an arrow turns it into a homing shot. It’s the same “everything is a physics object with properties” logic, just aimed at the fight rather than the traversal. Ascend, by contrast, is the one pure convenience verb: swim up through solid ceilings to the room above. It looks like magic and reads mechanically as a fast-travel-within-a-level tool, there mainly so vertical level design doesn’t turn into a chore of finding the one staircase.

The Depths are the tell

Underneath Hyrule sits the Depths, a mirror-image map lit only by the glowing plants your Brightbloom Seeds coax awake and by lightroots that permanently clear a radius of the dark. It’s the part of the game that gets the least credit and deserves the most scrutiny, because it’s where the design philosophy runs into its own limits. The Depths are large, visually consistent — the eerie, spore-lit dark works as mood for a long time — and, mechanically, thinner than the surface. Zonai device farms and a scattering of mining and combat encounters repeat across a map roughly the size of Hyrule itself, and thirty hours in, the novelty of “it’s dark and I have a torch” has done most of what it’s going to do.

That’s worth dwelling on because it’s the honest cost of Tears of the Kingdom’s scale. Making a physics sandbox this deep and this reliable took an enormous amount of engineering effort, and the Depths read like content built to fill a footprint that size rather than content that earned its footprint on its own terms. Compare it to the Sky Islands, which are smaller, front-loaded early, and mostly there to teach Ultrahand before the surface opens up — a good tutorial space dressed as spectacle. The Depths, structurally, are the same idea run three times longer than the idea supports.

The sky islands as a soft tutorial

What the Sky Islands do well is invisible by design, which is the highest compliment a tutorial space can earn. The very first Ultrahand puzzle you meet — building a raft to cross a floating gap — has almost no wrong answer, because nearly anything that floats and has a fan on it will get you across. By the fourth or fifth island, the game is quietly demanding tighter tolerances: a build that needs to be stable enough to survive a landing, or light enough that two fans can lift it. Nobody tells you the tolerances have tightened. You just notice your first-island solution stopped working and build something better, and that’s the whole tutorial, taught entirely through failure states rather than text.

Recall and the durability argument nobody won

Recall is the fourth verb, and the strangest: point at any object with a recent physics history and rewind its motion, so a boulder that fell off a cliff floats back up to where it started, with you riding it. It’s mostly used for the obvious trick — ride a falling rock back up to a sky island — but its deeper function is patching a hole Breath of the Wild left open. That earlier game’s weapon durability system was the most argued-over decision Nintendo made all decade: swords snap after a dozen fights, forcing you to constantly loot and rotate your kit, and a large chunk of the fanbase never made peace with it. Tears of the Kingdom doesn’t remove durability. It leans into it, because Fuse means a snapped stick is never actually a dead end — it’s a fresh canvas for whatever material you found five minutes ago — and Recall means a bad physics interaction is never a permanent loss either, just a mistake you can wind back. The durability system stops being a punishment and becomes the reason the other two verbs have something to chew on. Take Fuse away and durability is still an annoyance; keep it, and durability is the thing that makes you experiment instead of hoarding a favourite sword for the final boss the way Breath of the Wild quietly encouraged you to.

That’s a genuinely clever piece of systems design — solving a five-year-old community complaint by giving the original decision a second system to lean on, rather than reversing it — and it’s the kind of fix that only shows up when a sequel is built by the team that lived with the original’s reception rather than a team starting fresh. It also means the loudest criticism of Breath of the Wild becomes, in Tears of the Kingdom, one of the stronger arguments for why the sequel needed Ultrahand and Fuse in the first place rather than a bigger map or a new region.

Ganondorf and the shape of the story

Where the story of Breath of the Wild was mostly ambient — you pieced Calamity Ganon together from memories scattered across a hundred years of ruin — Tears of the Kingdom front-loads its villain. Ganondorf appears in the opening minutes, sealed for centuries under Hyrule Castle, and the plot’s job for the next forty hours is closing the gap between that cold open and the final confrontation. It’s a more conventional structure, and a slightly worse fit for the exploratory design around it: the sense of dread the game wants you to feel about Ganondorf doesn’t build the way the mystery of Calamity Ganon did, because you’ve already met him. What the story does instead, and does with more craft than the marketing let on before launch, involves Princess Zelda and a plot thread that only pays off in the last act — covered below the line, because it’s the one genuine spoiler in a game that otherwise wears its structure on its sleeve.

Where it sits against the run of open worlds it’s not trying to be

The interesting comparison isn’t to other open-world action games, most of which solve the “what do I do next” problem with a map full of icons. Tears of the Kingdom solves a different problem — “what tools do I have” — and lets the world stay comparatively icon-light because the tools generate the content rather than the map doing it. That’s a design lineage that runs back through Breath of the Wild’s own break from the Ubisoft-style checklist, and it’s worth reading alongside pieces on why every open world puts a tower on the map and the map screen as an admission of failure — Tears of the Kingdom is one of the few big-budget open worlds actively arguing against both habits, and mostly winning the argument.

The shrines carry less individual charm than Breath of the Wild’s did — fewer of them land as a standalone idea worth remembering a year later — but the aggregate physics toy they sit inside is the more significant achievement, because it’s not a content pipeline, it’s a rule set, and rule sets generalise in ways individual puzzle rooms don’t.

If you’ve played Breath of the Wild and are wondering whether the sequel is worth the ticket given how much geography repeats: it is, on the strength of Ultrahand and Fuse alone, and the Depths are the one stretch you’re allowed to rush through without guilt.

Spoilers below

The Ganondorf cold open isn’t the whole story. Zelda, pursuing the source of Ganondorf’s imprisonment seal deep in the past, is transformed into the Light Dragon — the same dragon that circles Hyrule’s skies throughout the entire game, glimpsed constantly and never explained until the very end. Every dragon sighting in the previous forty hours was her, already sacrificed, already searching. It’s a properly grim twist for a Nintendo game to commit to, and it recontextualises a background detail (a slow, silent dragon in the far distance) that most players will have logged as ambient wildlife rather than the emotional crux of the plot. The finale restores her, but the game doesn’t rush to undo the horror of the transformation on the way there, and that restraint is the best writing choice in it.

What to play next: if the appeal is the physics-as-puzzle idea specifically, Cocoon’s puzzle design with no fat on it scratches the same itch at a fraction of the scale and length, no glue required.