Conway’s Game of Life is a zero-player simulation where complex, lifelike patterns emerge from just four deterministic rules. Created by British mathematician John Conway in 1970, this model runs entirely on its initial setup without any further human input. You set up a grid of cells, press start, and watch simple math produce chaotic oscillations, moving spaceships, and self-replicating structures.
Understanding how to play and build patterns in this classic cellular automaton requires grasping its grid mechanics, neighborhood counts, and evolutionary phases. Whether you're coding your first simulation or manually tracking generations on paper, here is how the system works and how you can master it.
The Core Grid and Setup Mechanics
The Game of Life plays out on a two-dimensional orthogonal grid of square cells. Each individual cell exists in one of two states: alive or dead.
In theoretical mathematics, this grid stretches infinitely in every direction. Practical computer programs use finite grids, often wrapping the edges around like a torus so cells on the left edge neighbor those on the right.
Before a simulation begins, you create a starting configuration called a seed. This seed dictates every single future generation. Even a tiny change to the seed completely alters how the simulation evolves over time.
The Four Fundamental Rules
Every generation updates simultaneously across the entire grid. To figure out whether a cell lives, dies, or is born in the next step, you count its eight immediate neighbors—horizontally, vertically, and diagonally adjacent.
The four classic rules are:
- Underpopulation: Any live cell with fewer than two live neighbors dies as if by loneliness.
- Survival: Any live cell with two or three live neighbors lives on to the next generation.
- Overpopulation: Any live cell with more than three live neighbors dies as if by overcrowding.
- Reproduction: Any dead cell with exactly three live neighbors becomes a live cell as if by birth.
These rules use standard B3/S23 notation in cellular automaton literature, meaning a dead cell is born with 3 neighbors (B3) and a live cell survives with 2 or 3 neighbors (S23).
Classic Pattern Categories
Over decades of analysis, mathematicians and programmers have categorized thousands of recurring structures. Recognizing these patterns helps you predict how a grid will behave.
Still Lifes
Still lifes are stable patterns that never change from one generation to the next because every live cell maintains exactly two or three neighbors, and no dead cell ever reaches three neighbors. Common examples include the block (a 2x2 square), the beehive, and the tub.
Oscillators
Oscillators cycle through a fixed sequence of states before returning to their original configuration. The simplest is the blinker, a line of three cells that flips between horizontal and vertical every single generation (period 2). More complex oscillators include the toad and the pulsar.
Spaceships
Spaceships are patterns that translate across the grid over several generations, reappearing in a shifted position. The glider is the smallest and most famous spaceship, traveling diagonally across the board every four generations. Larger spaceships move orthogonally.
Guns and Infinite Growth
Guns are stationary structures that periodically shoot out spaceships across the grid. The Gosper glider gun, discovered by Bill Gosper in 1970, generates a continuous stream of gliders, proving that patterns can grow infinitely on a finite grid if fed enough space.
Common Mistakes When Playing Manually
If you're testing rules by hand on graph paper, it's easy to make calculation errors. Watch out for these pitfalls:
- Sequential updating: Updating cells one by one instead of simultaneously across the whole grid. Every cell must look at the current generation's state to calculate the next generation.
- Miscounting diagonal neighbors: Forgetting that corner-touching cells count as neighbors in the standard Moore neighborhood.
- Forgetting boundary conditions: On small manual grids, edge cells have fewer than eight neighbors, which alters local behavior unless you wrap the edges.
Frequently Asked Questions
Can the Game of Life simulate a real computer?
Yes. The Game of Life is Turing complete. By arranging gliders, guns, and logic gates, researchers have built working electronic circuits, binary adders, and even a simulated version of Tetris inside the grid.
How do I write code for it?
You need a 2D array representing the grid. In each loop iteration, count the neighbors for every coordinate using modulo arithmetic for wrap-around edges, apply the four rules, and write the results to a secondary array before updating the primary display.
Are there other rules besides B3/S23?
Thousands of them exist. Mathematician Stephen Wolfram and others explored alternative rules under the "Life-like" umbrella, where different birth and survival numbers create crystal growth, chaotic noise, or stable mazes.
What is the most famous pattern?
Aside from the glider and the Gosper glider gun, The Acorn is famous for a seed of just seven cells that evolves for 5,206 generations, generating 13 gliders before finally stabilizing.
