README
# evolve.life
Conway's Game of Life simulator — a pure-computation Native Gene for the Rotifer Protocol.
## Why
The Game of Life is the quintessential cellular automaton: dead-simple rules produce emergent complexity. It maps perfectly to Rotifer's biological metaphor — cells live, die, and evolve through generations, just like Genes in the Arena.
As a **Native Gene**, `evolve.life` performs zero I/O. It compiles to sandboxed WASM and runs deterministically, making it an ideal benchmark for IR compilation, Arena fitness scoring, and cross-Binding portability.
## Usage
```bash
# Run with a preset pattern
rotifer run evolve.life --input '{"preset": "r-pentomino", "generations": 500}'
# Benchmark: 512x512 grid, 1000 generations
rotifer run evolve.life --input '{"width": 512, "height": 512, "generations": 1000, "preset": "random", "seed": 7}'
# Custom rule (HighLife: B36/S23)
rotifer run evolve.life --input '{"preset": "random", "rule": "B36/S23"}'
```
## Presets
| Preset | Type | Description |
|--------|------|-------------|
| `glider` | Spaceship | Moves diagonally, period 4 |
| `blinker` | Oscillator | Period 2, simplest oscillator |
| `beacon` | Oscillator | Period 2, 2x2 block pair |
| `pulsar` | Oscillator | Period 3, highly symmetric |
| `r-pentomino` | Methuselah | 5 cells → 1103 generations to stabilize |
| `random` | Soup | Seeded 35% density fill |
## Output
- `final_grid` — grid state after simulation
- `alive_count` — live cells at end
- `peak_population` — maximum live cells observed
- `stabilized_at` — generation where grid reached equilibrium (-1 if not)
- `extinction` — whether all cells died
- `cells_processed` — total cell evaluations (for benchmarking)
- `ascii_snapshot` — ASCII art of final state (max 80×40)
## Fitness Characteristics
| Metric | Expected |
|--------|----------|
| Success Rate | 1.0 (deterministic, no failure mode) |
| Latency | <50ms for 64×64×100, ~2s for 512×512×1000 |
| Resource Cost | O(w × h × g) memory: O(w × h) |
| Robustness | Handles all edge cases (empty grid, max size, custom rules) |
Conway's Game of Life simulator — a pure-computation Native Gene for the Rotifer Protocol.
## Why
The Game of Life is the quintessential cellular automaton: dead-simple rules produce emergent complexity. It maps perfectly to Rotifer's biological metaphor — cells live, die, and evolve through generations, just like Genes in the Arena.
As a **Native Gene**, `evolve.life` performs zero I/O. It compiles to sandboxed WASM and runs deterministically, making it an ideal benchmark for IR compilation, Arena fitness scoring, and cross-Binding portability.
## Usage
```bash
# Run with a preset pattern
rotifer run evolve.life --input '{"preset": "r-pentomino", "generations": 500}'
# Benchmark: 512x512 grid, 1000 generations
rotifer run evolve.life --input '{"width": 512, "height": 512, "generations": 1000, "preset": "random", "seed": 7}'
# Custom rule (HighLife: B36/S23)
rotifer run evolve.life --input '{"preset": "random", "rule": "B36/S23"}'
```
## Presets
| Preset | Type | Description |
|--------|------|-------------|
| `glider` | Spaceship | Moves diagonally, period 4 |
| `blinker` | Oscillator | Period 2, simplest oscillator |
| `beacon` | Oscillator | Period 2, 2x2 block pair |
| `pulsar` | Oscillator | Period 3, highly symmetric |
| `r-pentomino` | Methuselah | 5 cells → 1103 generations to stabilize |
| `random` | Soup | Seeded 35% density fill |
## Output
- `final_grid` — grid state after simulation
- `alive_count` — live cells at end
- `peak_population` — maximum live cells observed
- `stabilized_at` — generation where grid reached equilibrium (-1 if not)
- `extinction` — whether all cells died
- `cells_processed` — total cell evaluations (for benchmarking)
- `ascii_snapshot` — ASCII art of final state (max 80×40)
## Fitness Characteristics
| Metric | Expected |
|--------|----------|
| Success Rate | 1.0 (deterministic, no failure mode) |
| Latency | <50ms for 64×64×100, ~2s for 512×512×1000 |
| Resource Cost | O(w × h × g) memory: O(w × h) |
| Robustness | Handles all edge cases (empty grid, max size, custom rules) |
表型
输入
| 属性 | 类型 | 描述 |
|---|---|---|
| grid | array | Custom initial grid (2D array of 0s and 1s). Ignored if preset is set. |
| rule | string = B3/S23 | Life-like rule in B/S notation (default: B3/S23 for Conway's Life) |
| seed | number | Random seed for 'random' preset (deterministic output) |
| width | number = 64 | Grid width in cells |
| height | number = 64 | Grid height in cells |
| preset | glider | pulsar | blinker | beacon | r-pentomino | random | Initial pattern preset. Overrides grid if provided. |
| generations | number = 100 | Number of generations to simulate |
输出
| 属性 | 类型 | 必填 | 描述 |
|---|---|---|---|
| extinction | boolean | ✓ | Whether all cells died before final generation |
| final_grid | array | ✓ | Grid state after all generations |
| alive_count | number | ✓ | Number of live cells in final grid |
| stabilized_at | number | ✓ | Generation at which the grid stabilized (-1 if not stabilized) |
| ascii_snapshot | string | ASCII art of the final grid state (max 80x40) | |
| cells_processed | number | ✓ | Total cell evaluations (width * height * generations_computed) |
| peak_population | number | ✓ | Maximum live cells observed across all generations |
| generations_computed | number | ✓ | Actual number of generations computed (may be less if stabilized/extinct) |
原始 JSON Schema
inputSchema
{
"type": "object",
"required": [],
"properties": {
"grid": {
"type": "array",
"items": {
"type": "array",
"items": {
"enum": [
0,
1
],
"type": "number"
}
},
"description": "Custom initial grid (2D array of 0s and 1s). Ignored if preset is set."
},
"rule": {
"type": "string",
"default": "B3/S23",
"description": "Life-like rule in B/S notation (default: B3/S23 for Conway's Life)"
},
"seed": {
"type": "number",
"description": "Random seed for 'random' preset (deterministic output)"
},
"width": {
"type": "number",
"default": 64,
"maximum": 512,
"minimum": 4,
"description": "Grid width in cells"
},
"height": {
"type": "number",
"default": 64,
"maximum": 512,
"minimum": 4,
"description": "Grid height in cells"
},
"preset": {
"enum": [
"glider",
"pulsar",
"blinker",
"beacon",
"r-pentomino",
"random"
],
"type": "string",
"description": "Initial pattern preset. Overrides grid if provided."
},
"generations": {
"type": "number",
"default": 100,
"maximum": 10000,
"minimum": 1,
"description": "Number of generations to simulate"
}
}
}outputSchema
{
"type": "object",
"required": [
"final_grid",
"alive_count",
"peak_population",
"stabilized_at",
"extinction",
"generations_computed",
"cells_processed"
],
"properties": {
"extinction": {
"type": "boolean",
"description": "Whether all cells died before final generation"
},
"final_grid": {
"type": "array",
"items": {
"type": "array",
"items": {
"enum": [
0,
1
],
"type": "number"
}
},
"description": "Grid state after all generations"
},
"alive_count": {
"type": "number",
"description": "Number of live cells in final grid"
},
"stabilized_at": {
"type": "number",
"description": "Generation at which the grid stabilized (-1 if not stabilized)"
},
"ascii_snapshot": {
"type": "string",
"description": "ASCII art of the final grid state (max 80x40)"
},
"cells_processed": {
"type": "number",
"description": "Total cell evaluations (width * height * generations_computed)"
},
"peak_population": {
"type": "number",
"description": "Maximum live cells observed across all generations"
},
"generations_computed": {
"type": "number",
"description": "Actual number of generations computed (may be less if stabilized/extinct)"
}
}
}Arena 历史
| 日期 | 适应度 | 安全分 | 调用数 |
|---|---|---|---|
| 3月17日 | 1.0000 | 1.00 | 1 |