Game of Life Simulator screenshot

Game of Life Simulator

Run Conway's Game of Life on a bounded grid: play, pause, step, randomise from a seed, or clear. The rule is applied in your browser and nothing is uploaded.

Privacy First
0 Runs
Updated 2026-11-07
Runs in your browser
Simulation only, no prediction
Nothing sent to a server
Works offline after load

How this simulation works

Conway's Game of Life is a four-line rule applied to a grid of live and dead cells: a live cell with two or three live neighbours survives, a cell with exactly three live neighbours is born, and everything else dies. This page applies that rule to a fixed grid whose border is dead rather than wrapped, so a glider leaves the field instead of reappearing on the far side and confusing what you are watching. The grid size and the starting density are controls; Randomise fills the grid from the seed, and Clear empties it so you can start from a blank board.

How to use it

Set the cell grid and the starting density, then press Play. Step advances exactly one generation, Randomise refills the grid from the current seed, and Clear empties it. Pause holds the current generation on screen.

Why it runs in your browser

The simulation runs on this page with JavaScript and the canvas API. Nothing is uploaded, no account is needed, and the page works offline after it loads.

This is a browser simulation of Conway's Game of Life on a bounded grid. Live cells survive on two or three neighbours and a dead cell is born on exactly three; every other cell dies in the next generation.

The seed and the fill density decide the starting pattern. A seed of 1234 at 28 percent fill starts the same way every time, which is what makes a run reproducible rather than merely pretty.

You can play the generations on a timer, advance one at a time, randomise to the next seed, or clear the grid. The board does not wrap: a glider that leaves the right edge is gone rather than reappearing on the left.

The grid, the seed and the rules

The column count runs from 20 to 120 and the row count is a fixed 40, so the grid grows wider rather than taller as you raise the columns. Fill density sets how many cells start alive, from 5 to 60 percent.

Every step computes the whole next generation from the current one, so a cell's neighbours are read before anything is written. That is what keeps a blinker a blinker rather than a smear.

Because the border is dead rather than wrapped, the edge acts as an absorbing boundary. Patterns near an edge behave differently from the same pattern in the middle, and that is deliberate.

  • Birth: exactly three live neighbours
  • Survival: two or three live neighbours
  • Grid: 20 to 120 columns, bounded, not wrapped

What the picture shows and what it does not

The canvas is a view of the state, not a measurement. The page reports no population count, no period and no pattern classification, so a still life and a long transient look equally interesting.

The clock only paces the drawing. The state after a given number of steps depends on the seed and the density and on nothing else, so pausing and resuming cannot change the outcome.

How to use the board controls

Set the Seed, the Columns and the Fill density, 1234 at 28 percent is the reproducible example, and the board fills the same way every time. Rules are Conway's on a bounded grid that does not wrap.

Press Play for generations on a timer, Step for one at a time, Randomise for the next seed, or Clear for an empty board. A glider leaving the edge is gone.

What to Expect

Best for

  • Watching a known rule produce gliders, blinkers and still lifes
  • Demonstrating why the border rule matters to a pattern
  • A reproducible canvas example for a lesson or a test

Not ideal for

  • Measuring population dynamics or classifying patterns
  • An unbounded or wrapping universe, which this grid is not
  • Loading a pattern file, which the page cannot import

What this tool keeps

  • The starting pattern for a seed and a density
  • The current generation until Clear or Randomise
  • A paused start when reduced motion is requested

What may need cleanup

  • A pattern that dies at the edge, which an unbounded grid would keep alive
  • The lack of any population figure, which you would have to count yourself
  • A cleared grid, which Reset restores only from the seed

Common errors

  • Expecting a glider to reappear after crossing an edge
  • Changing the density and expecting the same starting cells
  • Reading a stable pattern as a stalled simulation

Example use cases

  • A lesson on Conway's rule
  • A demo of why the border convention matters
  • A reproducible example for a canvas test

Sample input

Seed 1234, 60 columns and 28 percent fill, then Step ten times.

Sample output

Ten generations of a pattern fixed by that seed and density, with cells near the edge dying at the boundary.

Who this is for

  • Teachers and students of cellular automata
  • Developers wanting a canvas simulation example
  • Anyone curious how three-neighbour birth behaves on a finite grid

Frequently asked questions

Which rules does the Life simulator apply?

Live cells survive on two or three neighbours, a dead cell is born on exactly three, and every other cell dies in the next generation.

Can I import a pattern file?

No. The page cannot load a pattern file; a pattern is drawn on the grid or produced from a seed.

Why did the pattern die at the edge?

The grid is bounded rather than wrapping, so cells at the edge lose neighbours an unbounded universe would keep.

Developer program

Build custom tools for your store

Join the SDK program to create useful ecommerce utilities for your store.