physim worlds — a visual guide

Each physim task drops an agent into a procedurally generated world with hidden laws behind an anonymous port interface: it can set input ports in [−1,1] and read noisy, unnamed sensors — nothing else. These pages show what those worlds actually are, from two perspectives: the god view (evaluator-only: the hidden lattice, the wiring, the true collective modes, films of the dynamics) and the agent view (exactly what the agent gets). Under the hood every world is a grid of simple units (24×24 up to 96×96 cells) updated by one local rule per track family; everything an agent encounters — materials, chemistry, organisms, evolution — emerges from that rule. Time advances in discrete ticks; experiments spend a tick budget.

D — bulk matter

Coupled tanh lattices: collective bistability, hysteresis, per-region timescales, slow fatigue. The tutorial track that ends in a moody, 8-region material (D4) frontier agents still fail.

C — chemistry

Reaction–diffusion: self-sustaining "atoms", movable measurement apparatus, drifting objects, two coupled species, and an excitable wave medium (C4) whose timing physics breaks every frontier model.

B — ecology

Make the food finite and chemistry becomes ecology: carrying capacity, competition, natural selection, the exclusion boundary — and B2, where excitable waves are the weather that feeds the world.

E — evolution

Add heredity: a mutable trait field new tissue copies from its parents. Gene pools shift under selection the world itself applies — storms, famines, port-driven eras — and the population's history is written in its genotypes.

How the worlds work — the shared machinery

All four tracks are instances of one template: a multi-channel lattice field theory (diffusion/mixing + a pointwise nonlinear reaction + input fields + noise). The track pages describe each reaction family; everything below is common.

The idea: why the worlds are built this way

The benchmark asks one question: can an agent do science? Not recall science — do it. That requires a world where:

What the agent can see and do

The agent never sees the grid, the wiring, or any panel marked "god view" on these pages. Its entire universe is:

Everything an agent can access runs in a separate process from the world engine; hidden state, wiring, and ground-truth ensembles never cross that boundary. Worlds are procedurally generated from a seed (plus an optional evaluator-side salt, so public code cannot reproduce a live evaluation world).

The difficulty ladder, track by track

bulk matter (D): D0 tutorial magnet → D1 sensor opacity → D2 multi-region → D3 budget pressure → D4 slow/moody regions + fatigue

chemistry (C): C0 static objects → C1 movable apparatus → C2 drifting objects → C3 two species + cascades → C4 excitable waves

ecology (B): B0a carrying capacity → B0b competition → B0 selection → B1 exclusion boundary → B2 wave-fed ecology

evolution (E): E0 heredity laboratory → E1 storm world (evolution as weather) → E2 enzyme economics (trade-offs as theorems of conservation)

Each track page carries god-view films of the dynamics, static panels, and the per-world parameter cards.