Tier 4 Series 303

303F: The Poisson Disc

Pilot Record
Student Profile
"Randomly placing trees usually results in clumps and overlaps. It looks digital and fake. Nature has personal space. A Navigator uses Poisson Disc Sampling to create organic, evenly spaced distributions."

The Concept: Poisson Disc Sampling

A specific algorithm that picks a point, then tries to pick another point that is *at least* "r" distance away. If it fails, it tries again.

* **Result:** A pleasing, natural distribution of objects without overlaps.
* **Use Case:** Forests, crowd placement, loot spawning.
Red Flag Detected

The AI Trap: "The Overlap Clump"

You ask the AI: "Scatter 50 coins on the floor."

// AI-Generated Code: Lazy Random
for(int i=0; i<50; i++) {
    // Audit Fail: Coins will spawn inside each other.
    Vector3 pos = new Vector3(Random.Range(0,10), 0, Random.Range(0,10));
    Instantiate(coin, pos, rot);
}

This is "Collision Chaos." Physics objects spawning inside each other will explode apart or look glitchy.

Elite Telemetry

Research shows "Elite" teams achieve 15% faster lead times by keeping AI on a "very tight leash."

  • Small Batches Solving one problem at a time prevents logic drift.
  • Modular Design Localizing the "blast radius" of AI changes.
  • Tight Loops Rapid iteration with constant code review.

The Navigator's Correction

Corrective Protocol
// Corrected: Organic Spacing
List<Vector2> points = PoissonDisc.Generate(radius: 1.5f);
foreach(var p in points) {
    Instantiate(coin, new Vector3(p.x, 0, p.y), rot);
}
Your Pilot Command
> A skilled Navigator directs the AI to use Poisson logic.
Next Mission
The Procedural Exam