The Game of Life is a cellular automaton devised by mathematician John Conway in 1970. It is a "zero-player" game: from a single starting pattern, an entire universe of gliders, oscillators and spaceships unfolds purely from local rules, with no further input. It is the classic demonstration of emergence — how astonishingly complex, lifelike behaviour can arise from a handful of trivial rules — and it inspired whole fields from artificial life to self-replicating machines.
B3/S23 — Conway's classic rule: a dead cell with exactly
3 live neighbours is Born; a live cell with
2 or 3 neighbours Survives;
all other cells die. Each digit before the slash is a birth count, each
digit after it a survival count, so variants like
B36/S23 (HighLife) tweak those thresholds.
The Game of Life is Turing-complete — anything a computer can calculate can, in principle, be built inside it. Enthusiasts have constructed working clocks, calculators, and even a full simulation of the Game of Life running inside the Game of Life itself.
Conway's Game of Life in 3D: a zero-player cellular automaton where extraordinarily complex, self-organising structures emerge from two simple rules about birth and survival.
Each 3D cell follows two rules: a living cell survives with the right number of neighbours; a dead cell is born if it has exactly the right count. Simple rules produce self-replicating structures, gliders, and stable oscillators.
Click cells to toggle alive/dead. Press Play to run. Load famous presets like gliders or spaceships. Adjust generation speed and birth/survival neighbour thresholds with sliders.
John Conway proved that the 2D Game of Life is Turing-complete — it can compute anything a real computer can. In 2009, a player built a working model of Conway's Life running inside the game itself.
This is Conway's Game of Life rendered in 3D, a zero-player cellular automaton in which an entire universe of motion unfolds from one starting pattern. Every cell on a torus-wrapped grid counts its eight Moore neighbours and updates simultaneously each generation according to a birth/survival rule such as B3/S23. From these trivial local rules emerge gliders, oscillators and self-replicating structures, making the simulation a vivid demonstration of how complex, lifelike order can arise from utter simplicity.
It runs a cellular automaton on a wrap-around grid. Each cell is alive or dead; the next state depends only on its eight neighbours. A live cell with a survival count stays alive, a dead cell with a birth count is born, all others die. Rules like Conway B3/S23, HighLife B36/S23, Replicator B1357/S1357 and Day&Night B3678/S34678 are selectable, plus a true 3D variant using a 20×20×20 grid.
Sliders set grid size (20–80), random seed density (5–60%) and speed (1–30 generations per second). Dropdowns pick a preset (Gosper gun, pulsar, acorn, R-pentomino, diehard), the rule, and the projection: flat, stacked by generation, wrapped on a sphere, or full 3D. Restart, Step and Pause control playback; Save/Load RLE import or export patterns. Drag to rotate, scroll to zoom.
Conway's Game of Life is Turing-complete: anything a computer can calculate can, in principle, be built inside it. Enthusiasts have constructed working clocks, calculators and even a copy of the Game of Life running entirely within the Game of Life.
It is a cellular automaton devised by mathematician John Conway in 1970. It is a zero-player game: you set an initial pattern of live and dead cells, and the whole evolution then proceeds automatically from a fixed set of rules. Despite that simplicity it can produce moving gliders, repeating oscillators and remarkably intricate behaviour.
The digits after B are the neighbour counts at which a dead cell is born, and the digits after S are the counts at which a live cell survives. So B3/S23 means a dead cell with exactly 3 live neighbours becomes alive, and a live cell with 2 or 3 neighbours stays alive; every other cell dies. Changing those digits gives variants such as HighLife B36/S23 or Day and Night B3678/S34678.
Grid size sets the width and height of the square lattice, density sets the fraction of cells seeded alive on a random start, and speed sets how many generations are computed each second. The preset menu loads famous patterns, the rule menu switches the birth/survival logic, and the projection menu shows the world flat, stacked by generation, on a sphere, or as a genuine 3D automaton.
Yes. The 2D modes apply the standard rules exactly: cells wrap around the edges like a torus, all eight neighbours are counted, and every cell updates simultaneously. The flat, height and sphere views are just different ways of drawing the same 2D grid. The 3D Life mode is a separate, distinct automaton on a cubic grid using a 26-neighbour rule, so its behaviour differs from the classic 2D version.
Because the rules are applied locally and in parallel across many cells, small differences propagate and interact, producing emergent structures that no single rule describes. This is the core idea of emergence: global complexity arising from purely local interactions. The Game of Life is famous precisely because it shows that even a two-rule system can support behaviour rich enough to be Turing-complete.