rng-seeding-patterns

Deterministic random number generation uses a managed seed to ensure predictable, reproducible outcomes for testing and debugging specific gameplay mechanics. This skill provides patterns for implementing seeded RNG to generate identical sequences across game sessions.

Use seeded RNG when you need predictable, reproducible outcomes for testing or deterministic mechanics like maze generation, and unseeded RNG for non-deterministic gameplay elements. This skill defines clear guidance on choosing between them based on your feature requirements.

Effective seed management patterns include utilizing scene-level configurations, URL parameters, and dedicated test modes. This skill illustrates how to implement these strategies to maintain deterministic logic for entity placement and maze generation.

Testing RNG behavior involves verifying deterministic outputs and documenting seed regeneration patterns. This skill provides test patterns to validate that your random number generation produces expected, reproducible results during automated testing.

Yes, you can use Phaser's built-in capabilities alongside custom seeded RNG implementations. This skill provides code examples covering both approaches for managing deterministic game logic and entity spawning.

Procedural generation differs on reload when random number generation is unseeded or the seed is not persisted. Implementing deterministic logic with seed management patterns ensures maze generation and entity placement remain reproducible across sessions.