Formal Specification
Greasy Bowsprit Walking — Formal Specification
Netherlands · water · Confidence: low
← Back to game pageImplementation Suitability
Suitability: engine-readyFidelity: discretizedGap-fills: 4
# Sprietlopen — AI-optimized specification
## General idea
A sequential balance traversal contest across a 1D greasy tapered pole suspended over water. Agents advance in discrete steps, selecting movement speeds that balance forward momentum against exponential slip probabilities that increase as the pole narrows toward the tip.
## Formal model
- **Abstraction taken:** Dynamic barefoot friction on a soap-lubricated wooden cylinder is modeled as a 1D spatial traversal of length meters. Slipperiness increases non-linearly with position as the pole diameter narrows from cm to cm.
- **Player/agent count:** 1 to individual agents competing sequentially.
- **Turn model:** Sequential attempts; each agent executes discrete balance/step ticks until reaching the goal or falling into the water.
## Formal rules
R1. **Pole Setup:** Linear track of length meters. Start position . Target flag at .
R2. **Attempt Initialization:** Active agent starts at , status `on_pole`, step counter , elapsed ticks .
R3. **Action Selection:** On each step tick, the agent chooses `step(speed)` where `speed` (representing m, m, or m strides) [gap-fill].
R4. **Slip Resolution:**
- At current position , base slip hazard is .
- Instantaneous slip probability is:
where is agent balance skill [gap-fill].
R5. **State Transitions:**
- If slip occurs: Agent status `fallen`, recorded distance , attempt terminates.
- If no slip: Position updates , elapsed ticks .
R6. **Flag Retrieval:** If , the agent successfully retrieves the flag (`flag_captured = true`), and recorded distance is with completion time .
R7. **Ranking & Win Condition:**
- Agents who captured the flag rank ahead of all agents who fell.
- Among flag capturers, the agent with the lowest completion time wins.
- If no agent captured the flag, the agent with the greatest distance wins.
## Edge cases
- **Tie at the end condition:** If two agents achieve identical maximum distance or identical completion time, a second tie-breaker attempt is run with increased soap multiplier () [gap-fill].
- **A player with no legal move:** While `on_pole`, all 3 step speeds are always available; when `fallen`, the attempt has concluded.
- **Repetition / cycles:** Impossible; is strictly monotonically increasing () per successful step, bounded by maximum 24 steps [gap-fill].
- **Simultaneous claims on the same resource or position:** Not applicable; attempts are conducted strictly one at a time.
- **Player elimination mid-round; turn order repair:** Falling ends only the current agent's attempt; next agent in rotation starts afresh from .
- **Resource exhaustion (pieces, tokens, space, time):** Not applicable; pole and soap conditions reset for each attempt.
- **Termination guarantee:** Because each step strictly advances , the agent must either slip or reach within at most 24 steps, guaranteeing absolute finite termination.
## State representation
- **Persistent state:** Current contestant index , leaderboard of attempts .
- **Per-attempt state:** Position , status , elapsed ticks .
- **Observable information:** Fully observable; contestant position, leaderboard rankings, and attempt results are public.
## Action space
- `step(speed)`: Parameter `speed` . Precondition: status .
## Playability notes
- **Single-player-vs-AI:** Natural; human competitor tries to optimize stride cadence against leaderboard scores of simulated AI competitors.
- **Group games (n>4):** Scales cleanly without interactive complexity across dozens of participants.
- **Physical-skill games:** Neuromuscular barefoot grip and dynamic balance are mapped to a stochastic hazard survival curve.
- **What is lost:** Real-time lateral ankle adjustments, pole flex/whip dynamics, and cold water splash sensations.
## Implementation hints
- Very low state complexity; state is a single floating-point coordinate .
- First milestone: Implement the 1D slip hazard curve and simulate 100 runs across varying speed policies to verify that cautious and sprinting strategies have viable strategic niches.
Download as JSON
Download the full game corpus entry for this game as structured JSON.
Download JSONCite this page
Greasy Bowsprit Walking (Netherlands). (n.d.). Games of Countries. https://gamesofcountries.com/games/nl-netherlands/sprietlopen
@misc{gamesofcountries-nl-netherlands-sprietlopen,
title = {Greasy Bowsprit Walking (Netherlands)},
year = {n.d.},
publisher = {Games of Countries},
url = {https://gamesofcountries.com/games/nl-netherlands/sprietlopen}
}