Formal Specification
Knurr and Spell ("Poor Man's Golf") — Formal Specification
United Kingdom · dexterity-throw · Confidence: high
← Back to game pageImplementation Suitability
Suitability: engine-readyFidelity: probabilistic-proxyGap-fills: 1
# Knurr and Spell — AI-optimized specification
## General idea
A turn-based long-distance striking game where agents trigger a spring trap to launch a small dense ball (*knurr*) into the air and attempt to hit it on the rise with a long flexible club (*pommel*). The core mechanic balances swing timing, power, and trajectory angle against aerodynamic carry and roll across open moorland.
## Formal model
- **Abstraction taken:** The physical spring release and swing trajectory are parameterized by two decision variables: `release_height in [1.0, 2.0]` meters and `swing_timing_offset in [-0.1, 0.1]` seconds relative to the apex of the rise. Impact mechanics yield launch velocity `V_0` and launch angle `theta`, which feed a 2D ballistic aerodynamic model with ground roll.
- **Player/agent count:** 2 agents (`P1`, `P2`).
- **Turn model:** Sequential blocks: each agent completes a fixed series of 10 knocks (or turns alternate knock-by-knock).
## Formal rules
R1. **Match Structure:** Each agent is allocated exactly `N = 10` knocks (hits).
R2. **Knock Execution:** On each knock, the active agent chooses parameters:
- `strike(spring_tension: float in [0.5, 1.0], swing_power: float in [0.5, 1.0], aim_elevation: float in [20.0, 45.0])`
R3. **Impact Resolution:**
- Timing window: True strike timing error `dt ~ Normal(0, sigma_skill)`.
- Impact efficiency `eta = max(0.0, 1.0 - (abs(dt) / 0.05)^2)`.
- If `eta == 0.0`: Hit is a complete miss / "duck" -> `distance = 0.0`.
- If `eta > 0.0`:
- Initial velocity `V_0 = 70.0 * swing_power * eta` (m/s).
- Ballistic flight distance `d_air = (V_0^2 * sin(2 * aim_elevation)) / 9.81 * Wind_Factor`.
- Ground bounce and roll `d_roll = 0.3 * d_air * Terrain_Factor`.
- Total distance for knock `d_knock = d_air + d_roll` (in yards).
R4. **Aggregate Yardage Update:**
- `Total_Score_Agent = Total_Score_Agent + d_knock`.
R5. **End Condition & Victory:**
- When both agents have completed all 10 knocks, the agent with the strictly greater `Total_Score` wins.
- [gap-fill] Tie resolution: If total yardages are identical to within 0.1 yards, a 1-knock sudden-death playoff determines the winner.
## Edge cases
- **Tie at the end condition:** Resolved by R5 (sudden-death single knock playoff).
- **A player with no legal move:** Not applicable; `strike()` is always legal.
- **Repetition / cycles:** Bounded by strict knock counter `N = 10` per agent.
- **Simultaneous claims on the same resource:** Not applicable; sequential turns on an open range.
- **Player elimination mid-round:** Not applicable; 2-agent closed match.
- **Resource exhaustion:** Exactly 10 knocks per player; non-depletable knurr and trap.
- **Termination guarantee:** Guaranteed by finite knock count decrementing to 0.
## State representation
- **Persistent state:** `knocks_taken_P1: int`, `knocks_taken_P2: int`, `total_distance_P1: float`, `total_distance_P2: float`, `active_agent: int`.
- **Per-knock ephemeral state:** Current knock distance `d_knock`, impact efficiency `eta`, air carry vs. ground roll breakdown.
- **Observable information:** Complete transparency of all scores and previous shot parameters.
## Action space
- `strike(spring_tension, swing_power, aim_elevation)`: Continuous 3-tuple parameterizing the mechanical setup and human swing.
## Playability notes
- **Single-player-vs-AI:** High fidelity as a golf-style driving range challenge where the agent seeks optimal launch parameters under fluctuating wind conditions.
- **Group games reduction:** Naturally scales to n-player tournament leaderboards by comparing aggregate distances.
- **Physical-skill games:** Physical swing timing and sweet-spot precision are abstracted into the stochastic timing error `dt`.
- **What is lost in the adaptation:** The sensory vibration through the cane handle and reading the micro-contours of Yorkshire heather.
## Implementation hints
- Pure mathematical physics function; execution takes microseconds.
- Branching factor is 3 continuous parameters (easily discretized into a grid of 5x5x5 = 125 actions).
- First milestone: Implement ballistic trajectory and wind multiplier function, running 1,000 trials to verify realistic distance distribution (150–250 yards).
Download as JSON
Download the full game corpus entry for this game as structured JSON.
Download JSONCite this page
Knurr and Spell ("Poor Man's Golf") (United Kingdom). (n.d.). Games of Countries. https://gamesofcountries.com/games/gb-united-kingdom/knurr-and-spell
@misc{gamesofcountries-gb-united-kingdom-knurr-and-spell,
title = {Knurr and Spell ("Poor Man's Golf") (United Kingdom)},
year = {n.d.},
publisher = {Games of Countries},
url = {https://gamesofcountries.com/games/gb-united-kingdom/knurr-and-spell}
}