Comp Sci GCSE • A-Level • University ●○○ Beginner ★ Free

Conway's Game of Life

Click on cells to toggle them, or use the controls below.

Speed: 30
Cell Size: 10
Pattern:
Simulation running
Generation: 0 | Live Cells: 0

The Four Rules of Life (B3/S23)

Conway's Game of Life runs on just four rules applied simultaneously to every cell each generation. Each cell is alive or dead; its fate depends solely on its 8 neighbours:

  • Underpopulation: A live cell with < 2 live neighbours dies — isolation.
  • Survival: A live cell with 2 or 3 live neighbours lives on.
  • Overpopulation: A live cell with > 3 live neighbours dies — overcrowding.
  • Reproduction: A dead cell with exactly 3 live neighbours becomes alive.

Written as B3/S23 (Birth-3, Survive-2-3). Devised by British mathematician John Horton Conway in 1970, first published in Scientific American.

Turing Completeness & Emergence

The Game of Life is Turing complete — given an infinite grid and time, it can simulate any computation. Enthusiasts have built logic gates, calculators, prime-number generators, and even working computers inside the Life grid.

This emergence of unbounded computational complexity from just two states and four rules is one of the most surprising results in mathematics. It shows that complex behaviour does not require complex rules — an insight driving modern research in artificial life, systems biology, and complex systems.

Canonical Pattern Types

Life patterns fall into four categories:

  • Still lifes — stable, unchanging (block, beehive, loaf)
  • Oscillators — cycle with fixed period (blinker p-2, pulsar p-3, pentadecathlon p-15)
  • Spaceships — translate across the grid (glider, LWSS, MWSS)
  • Guns & Puffers — emit spaceships indefinitely (Gosper Gun: period 30)

The R-pentomino — just 5 cells — evolves for 1,103 generations before stabilising, producing 6 gliders and dozens of other structures along the way.

Cellular Automata in Science

Life's principles extend far beyond Conway's grid. Cellular automata model systems where local rules produce large-scale structure:

  • Forest fires — spread modelled cell-by-cell with ignition probability
  • Crystal growth — snowflake shapes from local bonding rules
  • Traffic flow — Nagel–Schreckenberg motorway-congestion model
  • Tumour growth — cancer-cell division and invasion modelling
  • Epidemics — spatial SIR models predict disease-spread patterns

In A New Kind of Science (2002), Stephen Wolfram argues cellular automata could be the computational substrate of physical reality itself.

Key Rules & Metrics

ConceptValue / RuleNotes
Birth ruleB3: exactly 3 neighboursDead cell becomes alive
Survival ruleS23: 2 or 3 neighboursLive cell persists
Glider velocityc/4 diagonal1 diagonal cell per 4 generations
Gosper gun period30 generationsFirst known infinitely-growing pattern (1970)
R-pentomino lifespan1,103 generations5 cells → complex debris before stabilising
Neighbourhood typeMoore (8 cells)Includes diagonals; contrast von Neumann (4 cells)

Curriculum Links

LevelTopics
GCSE Computer ScienceAlgorithms, data representation, iterative processes
A-Level Computer ScienceComputational thinking, Turing machines, complexity classes
A-Level BiologyPopulation dynamics, emergence, systems biology
University (Computer Science)Formal languages, computability, artificial life
University (Mathematics)Discrete dynamical systems, graph theory, combinatorics
PostgraduateComplexity theory, self-organisation, swarm intelligence

🔒 Unlock All 32 Simulations

Get unlimited access to all 32 interactive simulations including Game of Life, Lorenz attractor, quantum mechanics, evolution simulator, and more with MySimulator Premium.

View Premium Plans

Share this simulation

Send this page to students or colleagues.