babylon.engine.simulation_engine

Simulation engine for the Babylon game loop.

ADR032: Materialist Causality System Order

The step() function is the core of Phase 2. It takes a WorldState and SimulationConfig and returns a new WorldState representing one tick of simulation time.

The step function is: - Pure: No side effects, no mutation of inputs - Deterministic: Same inputs always produce same outputs - Transparent: Order of operations encodes historical materialism

Turn Order (materialist causality - base before superstructure): 1. Vitality - Biological cost + death (dead entities don’t work) 2. Territory - Land state updates (land conditions affect production) 3. Production - Value creation (value must exist before extraction) 4. Solidarity - Organization (affects bargaining power) 5. Imperial Rent - Value extraction (landlord eats after harvest) 6. Decomposition - LA decomposes on super-wage crisis (Terminal Crisis) 7. Control Ratio - Guard:prisoner ratio + terminal decision (Terminal Crisis) 8. Metabolism - Environmental degradation (ecological residue of production) 9. Survival - Risk assessment (P(S|A), P(S|R) from material state) 10. Struggle - Action/Revolt (agency responds to survival odds) 11. Consciousness - Ideological drift (ideology responds to material) 12. Contradiction - Tension aggregation (final systemic accounting)

Phase 2.1: Refactored to modular System architecture. Phase 4a: Refactored to use ServiceContainer for dependency injection. ADR032: Reordered systems for materialist causality.

Functions

step(state, config[, persistent_context, ...])

Advance simulation by one tick using the modular engine.

Classes

SimulationEngine(systems)

Modular engine that advances the simulation by iterating through Systems.
class babylon.engine.simulation_engine.SimulationEngine(systems)[source]

Bases: object

Modular engine that advances the simulation by iterating through Systems.

The engine holds a list of systems and executes them in sequence. Order encodes materialist causality (ADR032): 1. Vitality (death check) 2. Territory (land state) 3. Production (value creation) 4. Solidarity (organization) 5. Imperial Rent (extraction) 6. Decomposition (LA crisis) 7. Control Ratio (terminal decision) 8. Metabolism (ecology) 9. Survival (risk assessment) 10. Struggle (agency) 11. Consciousness (ideology) 12. Contradiction (tension) 13. ContradictionField (field computation) - Feature 002 14. FieldDerivative (derivatives + principal) - Feature 002 15. EdgeTransition (predicates + state machine) - Feature 002

Parameters:

systems (list[System])

__init__(systems)[source]

Initialize the engine with a list of systems.

Parameters:

systems (list[System]) – Ordered list of systems to execute each tick. Order matters! Economic systems must run before ideology.

Return type:

None

property systems: list[System]

Read-only access to registered systems.

run_tick(graph, services, context)[source]

Execute all systems in order for one tick.

All logs emitted during this method are automatically tagged with tick number and a unique correlation_id (UUID) for tracing.

Parameters:

  • graph – NetworkX graph (mutated in place by systems)

  • services – ServiceContainer with config, formulas, event_bus, database, metrics

  • context – TickContext or dict passed to all systems

Spec 008: Logs within run_tick() include tick and correlation_id.

babylon.engine.simulation_engine.step(state, config, persistent_context=None, defines=None, calculator_overrides=None)[source]

Advance simulation by one tick using the modular engine.

This is the heart of Phase 2. It transforms a WorldState through one tick of simulated time by applying the MLM-TW formulas.

Parameters:

  • state (WorldState) – Current world state (immutable)

  • config (SimulationConfig) – Simulation configuration with formula coefficients

  • persistent_context (dict[str, Any] | None) – Optional context dict that persists across ticks. Used by systems that need to track state between ticks (e.g., ConsciousnessSystem’s previous_wages for bifurcation mechanic).

  • defines (GameDefines | None) – Optional custom GameDefines. If None, loads from default defines.yaml location. Use this for scenario-specific calibration.

  • calculator_overrides (dict[str, Any] | None) – Optional dict of calculator instances to inject into ServiceContainer (e.g., melt_calculator, tensor_registry).

Return type:

WorldState

Returns:

New WorldState at tick + 1

Order encodes historical materialism:

  1. Economic base (value extraction)

  2. Consciousness (responds to material conditions)

  3. Survival calculus (probability updates)

  4. Contradictions (tension from all above)

  5. Event capture (log significant changes)