Formal Specification
Namur stilt jousting — Formal Specification
Belgium · wrestling-combat · Confidence: low
← Back to game pageImplementation Suitability
Suitability: engine-readyFidelity: probabilistic-proxyGap-fills: 3
# Joutes sur échasses de Namur — AI-optimized specification
## General idea
A multi-agent team combat elimination game where elevated agents mounted on 84 cm stilts maneuver on a 2D arena grid to unbalance and eliminate opposing agents using directional collision strikes (shoulder charges, stilt sweeps, elbow parries, and balance drops). When one brigade is fully eliminated, the surviving agents immediately transition into a free-for-all battle (*Bout-a-tot*) until exactly one champion remains standing.
## Formal model
- **Abstraction taken:** The continuous physical mechanics of stilt balance (center of mass over narrow stilt points, physical grappling, momentum exchange) are discretized into a 2D grid with discrete balance/stability gauges () and simultaneous turn-based tactical action selection.
- **Player/agent count:** 2 teams (), each with agents.
- **Turn model:** Simultaneous action commitment per tick, resolved in discrete sub-phases (movement collision balance check elimination cleanup).
## Formal rules
R1. **Arena & Initial Setup:**
- Arena is a bounded 2D grid .
- Each agent starts with position , balance points , and status .
R2. **Action Selection:**
- Each standing agent selects 1 action per tick:
- `advance(dx, dy)`: Move 1 cell (cost: 5 balance points).
- `charge_shoulder(target_id)`: High-damage forward balance attack.
- `sweep_stilt(target_id)`: Stilt-to-stilt trip attack targeting lower body.
- `parry_elbow()`: Defensive stance reducing incoming strike damage by 70%.
- `genuflection()`: Drop center of gravity (evades upper-body charge; vulnerable to stilt sweep).
R3. **Collision & Damage Resolution:**
- Damage matrix resolves conflicting actions:
- `charge_shoulder` vs standing target: Deals damage to target balance.
- `sweep_stilt` vs `genuflection`: Deals critical damage to target balance.
- Strike vs `parry_elbow`: Deals only damage; attacker suffers recoil damage.
- Balance degrades from impacts: .
R4. **Elimination & Turn Order Repair:**
- If or agent is pushed out of bounds, agent state becomes .
- Eliminated agents are removed from the grid immediately and take no further actions.
R5. **Brigade Victory & Bout-a-tot Transition:**
- If all agents of are eliminated while has standing agent, wins the Brigade Joust.
- If , the game enters Phase 2 (**Bout-a-tot**):
- Team affiliations are dissolved; every surviving agent treats all other standing agents as hostile.
- [gap-fill] Standing agents have their balance restored to at the start of Bout-a-tot.
- The last single standing agent with wins the *Échasse d'Or*.
## Edge cases
- **Tie at the end condition:** If the last two standing agents eliminate each other on the exact same tick in Bout-a-tot, [gap-fill] both are restored to and fight a 1v1 sudden-death overtime round.
- **A player with no legal move:** Not applicable — defensive stance or movement is always available.
- **Repetition / cycles:** Bounded by strict monotonic decay: every strike or movement permanently drains balance; passive stalling is penalized by [gap-fill] a 1-point balance drain per tick for inactive agents.
- **Simultaneous claims on the same resource or position:** Both agents bounce back to origin cells and suffer 10 collision balance damage.
- **Player elimination mid-round; turn order repair:** Handled natively by R4 (eliminated agents transition to inactive status and are ignored in subsequent tick resolution).
- **Resource exhaustion:** Not applicable — balance points are non-renewable without active recovery.
- **Termination guarantee:** Provably terminates due to balance degradation bounded by and non-zero damage exchange per interaction.
## State representation
- **Persistent state:** Phase , active brigade counts , overall victor ID.
- **Per-agent state:** 2D position , balance gauge , team ID, stilt orientation .
- **Observable information:** All agent positions and visible balance tiers (High, Medium, Critical) are public knowledge.
## Action space
- `act(type, target_id, direction)`:
- `type` .
- `target_id`: Integer ID of adjacent enemy agent.
- `direction`: Movement/aim unit vector .
## Playability notes
- **Single-player-vs-AI:** Natural as human controlling 1 jouster within a brigade of cooperative AI teammates against an opposing brigade.
- **Group games (n>4):** Reduces to 2-agent form by having 1 commander agent per brigade issue tactical formation and focus-fire orders.
- **Physical-skill games:** Physical stilt balance abstracted into balance gauge with stochastic stumble chances when .
- **What is lost in the adaptation:** The physical vertigo of standing 84 cm high on thin wooden poles, crowd chants, and the bone-jarring impact of shoulder charges.
## Implementation hints
- Implement the combat resolution as a simultaneous matrix evaluator (Rock-Paper-Scissors-Lizard-Spock style with Charge > Genuflect > Sweep > Parry > Charge).
- Suggested first milestone: Build a 2-agent (1v1) duel engine with balance gauges and action resolution before scaling to full 30v30 mêlées.
Download as JSON
Download the full game corpus entry for this game as structured JSON.
Download JSONCite this page
Namur stilt jousting (Belgium). (n.d.). Games of Countries. https://gamesofcountries.com/games/be-belgium/joutes-sur-echasses-de-namur
@misc{gamesofcountries-be-belgium-joutes-sur-echasses-de-namur,
title = {Namur stilt jousting (Belgium)},
year = {n.d.},
publisher = {Games of Countries},
url = {https://gamesofcountries.com/games/be-belgium/joutes-sur-echasses-de-namur}
}