Formal Specification
Canal Vaulting / Far-Leaping — Formal Specification
Netherlands · water · Confidence: low
← Back to game pageImplementation Suitability
Suitability: adaptation-requiredFidelity: discretizedGap-fills: 4
# Fierljeppen — AI-optimized specification
## General idea
A multi-stage athletic vaulting contest where an agent executes sequential decisions across four distinct phases: sprint speed commitment, jump catch height, climb rate vs. pole tilt trade-off, and exit trajectory. The core strategic challenge is managing kinetic risk: climbing higher up the pole yields greater landing distance but increases the probability of pole stalling or water immersion failure (*nassprong*).
## Formal model
- **Abstraction taken:** The continuous physical aerodynamics and biomechanics of pole vaulting across water are abstracted into a 4-stage discrete decision pipeline parameterized by agent attributes (speed, climb agility, timing) and stochastic risk checks. [gap-fill] The physical mechanics are modeled via four parametric phase functions.
- **Player/agent count:** 1 to n agents competing sequentially in individual rounds.
- **Turn model:** Strictly turn-based across fixed tournament rounds (3 preliminary attempts + 3 final attempts per agent).
## Formal rules
R1. Setup: Each agent receives an allotment of `MAX_ATTEMPTS` (default = 3 preliminary, 3 final for qualifiers). A leaderboard tracks the single maximum distance `max_d(i)` recorded by agent `i`.
R2. On each attempt, the active agent selects a strategy tuple `(sprint_intensity, catch_height_target, climb_effort)` from discrete bounded intervals `[0.0, 1.0]`.
R3. Phase 1 (Sprint & In-jump): The agent's speed and approach accuracy determine initial momentum `M` and actual catch height `H_0`. If approach accuracy check fails [gap-fill: P(miss) = f(1.0 - speed_control)], the attempt fails immediately with distance 0.00 m.
R4. Phase 2 (Climb & Pole Arc): As the pole rotates through angle `theta` from vertical (90 deg) toward horizontal (0 deg), the agent climbs upward at rate `C = f(climb_effort, agility)`. Higher climb increases final height `H_final` but slows forward rotation. If the pole fails to pass the critical angle `theta_crit` before forward momentum expires, the pole falls backward into the water, resulting in a fault (0.00 m).
R5. Phase 3 (Exit Jump & Landing): When the pole reaches optimal exit angle [gap-fill: theta in [25 deg, 35 deg]]`, the agent pushes off. Final landing distance `D` is computed as `D = base_distance(pole_length, theta) + projectile_extension(H_final, push_force) + noise`.
R6. Score Recording: If the attempt successfully lands in the sand bed, `attempt_score = D`; otherwise, `attempt_score = 0.00`. The agent's best valid score across all attempts is updated: `max_d(i) = max(max_d(i), attempt_score)`.
R7. Win Condition: The agent with the highest `max_d(i)` at the conclusion of all scheduled rounds wins the match.
## Edge cases
- Tie at the end condition: If two agents have identical maximum distances, the tie is broken by comparing their second-best valid attempts. [gap-fill] If still tied, both agents execute a single sudden-death vault attempt.
- A player with no legal move: Not applicable; an agent always possesses the action to submit an attempt parameter tuple.
- Repetition / cycles: Not applicable; the game operates on a strictly finite, decremented attempt counter.
- Simultaneous claims on the same resource or position: Not applicable; attempts are strictly sequential.
- Player elimination mid-round; turn order repair: If an agent withdraws or faults all attempts, their rank is fixed at 0.00 m and the turn rotation advances to the next listed agent.
- Resource exhaustion (pieces, tokens, space, time): Bounded by the fixed attempt limit `N_attempts * N_players`.
- Termination guarantee: The total number of state transitions is strictly bounded by `N_players * Total_Attempts`, guaranteeing provable termination.
## State representation
- Persistent state:
- `agents`: List of agent IDs and skill profiles (speed, agility, power, composure).
- `attempt_history`: Map of `agent_id -> list of float distances`.
- `current_round`: Integer in `[1, Total_Rounds]`.
- `active_agent_index`: Integer in `[0, N_players - 1]`.
- Per-turn ephemeral state:
- Current attempt parameters: `(sprint_intensity, catch_height_target, climb_effort)`.
- Simulated trajectory state: `(momentum, apex_height, pole_angle, water_cleared_bool)`.
- Observable information:
- Complete public information: all scores, distances, leaderboard standings, and attempt histories are fully observable.
## Action space
- `vault_attempt(sprint_intensity: float, catch_height_target: float, climb_effort: float)`:
- Preconditions: Active agent's turn; remaining attempts > 0.
- Parameters: Float values normalized in `[0.0, 1.0]`.
## Playability notes
- Single-player-vs-AI: Natural; human player submits vault parameters aiming to beat an AI opponent's recorded benchmark or leaderboard threshold.
- Group games (n>4): Generalizes seamlessly; leaderboard sorting naturally accommodates arbitrary field sizes without interaction breakdown.
- Physical-skill games: Physical climbing velocity and spatial timing are modeled via non-linear risk/reward probability density functions: `P(water_immersion) = exp(climb_effort * 2.5 - momentum)`.
- What is lost: The visceral, real-time vestibular sensation of dangling 10 meters above water on a swaying composite pole and executing split-second leg extensions.
## Implementation hints
Low state complexity; the primary design task is tuning the mathematical parameters of the multi-stage vault physics model so that conservative strategies guarantee modest scores (e.g. 14–16 m) while aggressive climbing introduces realistic ~20–30% fault risks to reach world-class distances (e.g. 21+ m). First milestone: Implement deterministic scoring with fixed distance tables, then integrate the stochastic risk-failure curves.
Download as JSON
Download the full game corpus entry for this game as structured JSON.
Download JSONCite this page
Canal Vaulting / Far-Leaping (Netherlands). (n.d.). Games of Countries. https://gamesofcountries.com/games/nl-netherlands/fierljeppen
@misc{gamesofcountries-nl-netherlands-fierljeppen,
title = {Canal Vaulting / Far-Leaping (Netherlands)},
year = {n.d.},
publisher = {Games of Countries},
url = {https://gamesofcountries.com/games/nl-netherlands/fierljeppen}
}