Mode
1D Wolfram
2D Life
Rule number
Speed
0
Generation
—
Live cells
—
Pattern type
0
FPS
Explore all 256 1D Wolfram rules and Conway's Game of Life. Observe how simple local rules give rise to extraordinary complexity.
The core of a cellular automaton is its update rule, a function that maps a cell's local neighborhood to its new state. For a 1D elementary CA (like Wolfram's), the rule is defined for all 8 possible configurations of a 3-cell neighborhood.
$$Rule: \quad (s_{i-1}^t, s_i^t, s_{i+1}^t) \rightarrow s_i^{t+1}$$Here, $s_i^t$ is the state (0 or 1) of cell $i$ at time $t$. The rule number (0-255) is a decimal encoding of the 8-bit output pattern for these neighborhoods.
For Conway's Game of Life in 2D, the rule for a cell at position $(x, y)$ depends on its 8 surrounding neighbors (the Moore neighborhood). The update rule is:
$$ s_{x,y}^{t+1}= \begin{cases}1 & \text{if }(s_{x,y}^t = 1 \text{ and }N = 2 \text{ or }3) \text{ or }(s_{x,y}^t = 0 \text{ and }N = 3) \\ 0 & \text{otherwise}\end{cases} $$Where $N$ is the sum of the states in the 8-cell neighborhood. This simple conditional captures the famous rules of survival, death, and birth.
Cryptography & Random Number Generation: The unpredictable, complex patterns generated by certain rules (like Rule 30) are used to create pseudo-random sequences. These sequences are deterministic from the seed but appear random, useful in stream ciphers.
Modeling Natural Systems: CAs excel at simulating phenomena where global behavior arises from local interactions. For instance, they model forest fire spread, where each cell's state (unburned, burning, burned) depends on its neighbors, or the patterning of animal coats like a zebra's stripes.
Parallel Computing & Unconventional Architectures: The CA model is inherently parallel—every cell updates simultaneously based on local information. This inspires designs for massively parallel computing architectures and studies in emergent computation, like using Rule 110 to build a universal Turing machine.
Traffic Flow & Urban Planning: Simple CA models (e.g., the Nagel-Schreckenberg model) treat each road segment as a cell and cars as moving states. Local rules for acceleration, braking, and randomization can realistically simulate traffic jams and inform road network design.
First, you might think that "if the initial state is random, the result will also be random each time," but this is not necessarily true. Especially in 1D cellular automata, depending on the rule number, the system often converges to similar stable patterns (e.g., an "all-zero dead world" or a certain periodic pattern) even from different random initial states. This is a characteristic of rules with low "sensitivity to initial conditions." Conversely, if you want to observe complex behavior like that of Rule 110, a useful tip is to start from an initial state with some "live" cells scattered to a certain degree.
Next, a common misunderstanding in 2D Game of Life is that "patterns like gliders will move forever no matter where you place them". In reality, they can collide with other patterns and vanish or transform into completely different structures. If they reach the edge of the simulation area, they will disappear unless the boundaries are toroidal (connected left-right and top-bottom). When analyzing patterns in practice, ensure you have a sufficiently large field and consider the effects of the boundaries.
Finally, misconfiguring the "speed" parameter. Setting the speed to maximum makes the animation too fast to follow the detailed changes of patterns. Especially when you want to verify the operation of complex devices like glider guns, slowing down the speed and visually tracking the changes generation by generation is a shortcut to understanding. Conversely, when you want to know the global trend after long-term execution, increase the speed to run it quickly. The key is to use the appropriate setting for your purpose.
The concept of cellular automata is directly applied in simulations of phase transformations and microstructure formation in materials engineering. For example, the model for grain growth during metal cooling (the Potts model) is a type of cellular automaton. By simply treating each cell as a crystal grain orientation and applying a simple rule like "become the same orientation as the majority of neighboring cells," you can reproduce a realistic grain growth process.
Furthermore, in the field of fluid dynamics, lattice gas automata (LGA) and their evolved form, the lattice Boltzmann method (LBM), fall into this category. They model the collision and movement of fluid particles with simple rules on a cellular space, giving rise to macroscopic flow that obeys the Navier-Stokes equations. They excel at simulating flow in complex-shaped channels and multiphase flow.
Additionally, traffic flow simulation is a classic application. By viewing a road as a sequence of cells and cars as "live" cells, and imposing rules like "move forward if the cell ahead is empty, otherwise stop," you can capture complex phenomena such as traffic jam generation and propagation. For instance, you can observe "phase transition"-like behavior where flow suddenly deteriorates beyond a certain density.
First, try systematically exploring the "rule space" with this simulator. Wolfram classified 1D cellular automata into four classes based on their behavior: Class I (uniform convergence) to Class IV (complex, persistent structural patterns). Trying all rules from 0 to 255 and classifying which class each belongs to yourself is excellent training. For example, Rule 30 produces chaotic patterns (Class III), while Rule 90 generates fractal-like triangle patterns (Class II).
To learn the mathematical background, "computability theory" or "automata theory" is the next step. Understanding what it means for a cellular automaton to be Turing complete (computationally universal) and how computation is actually emulated in Rule 110 will deepen your understanding of the nature of computation. Also, when you learn that "gliders" in the Game of Life can be seen as 0 and 1 signals, and that you can construct "AND gates" or "OR gates" to manipulate them, you'll be amazed at how the fundamentals of digital circuits are contiguous with complex systems.
As a practical next topic, look into "cellular automata with three or more states" or "asynchronous-update cellular automata". For example, forest fire simulation (cell states: tree, burning, burned) or infectious disease spread models (cell states: susceptible, infected, recovered/immune) enable modeling of phenomena closer to reality. These are excellent subjects for experiencing how realistic, complex behavior "emerges" from simple rule sets.