Skip to main content
Formal Specification

Greenlandic Ring-and-Pin — Formal Specification

Greenland · dexterity-small · Confidence: high

← Back to game page

Implementation Suitability

Suitability: engine-readyFidelity: probabilistic-proxyGap-fills: 4
# Ajagaq — AI-optimized specification ## General idea A traditional Greenlandic Inuit cup-and-pin dexterity game where a perforated bone or antler target is swung on a sinew cord and caught on a pointed pin. Multiple holes of varying size and angle carry differentiated point values and difficulty tiers; a successful catch awards points and allows the player to swing again, while a miss passes the turn. ## Formal model - **Abstraction taken:** The continuous 3D pendulum swing, bone rotation, and pin spearing are replaced by a discrete probabilistic shot-selection model. [gap-fill] An agent selects a target hole category h∈{Hmain,Hside,Hangle}, each with a calibrated base success probability P(catch|h,skill) and point yield. - **Player/agent count:** 1 to n agents (commonly 2 to 4). - **Turn model:** Sequential turns with streak continuation (active player continues swinging until a miss occurs). ## Formal rules R1. Setup: Target winning score is W points (default W=20). An ordered circle of n agents is established 1,…,n, each starting with score 0. R2. Hole catalog and scoring: - `main_hole`: Large central aperture. Base success rate P0=0.70. Point value = 1 point. - `side_hole`: Medium peripheral hole. Base success rate P0=0.35. Point value = 3 points. - `narrow_eye`: Small apex hole / angled notch. Base success rate P0=0.15. Point value = 5 points. R3. Attempt resolution: On an agent's swing, the agent chooses a target hole h. The success probability is evaluated: P(success)=max⁡(0.05,min⁡(0.95,P0(h)+0.20×(Skill−0.5))) If successful, the agent receives the corresponding points and immediately takes another swing. R4. Turn passing: If the attempt fails (miss), the agent's turn ends immediately, and possession passes to agent (i(modn))+1. R5. [gap-fill] Round overflow cap: If an agent reaches score ≥W, the game concludes immediately without requiring equal turns for remaining players. R6. [gap-fill] Tie-break: Since the win threshold is reached sequentially during an active streak, ties are structurally impossible in standard play. R7. Win condition: The first agent to achieve a cumulative score ≥W wins the match. ## Edge cases - Tie at the end condition: Not applicable — turns are sequential; the first agent to reach W on their swing wins instantly. - A player with no legal move: Not applicable — swinging at one of the 3 hole types is always legal. - Repetition / cycles: Bounded by the positive point accumulation towards W; non-zero success probabilities guarantee W is reached in finite turns. - Simultaneous claims on the same resource or position: Not applicable — strictly sequential single-agent physical interaction with the apparatus. - Player elimination mid-round; turn order repair: If an agent withdraws, they are removed from the cycle; turn passes to next active index. - Resource exhaustion: Points accumulate; the bone and pin apparatus is non-depleting. - Termination guarantee: Provably terminates with probability 1 because ∑P(success)>0 on every turn, driving total score to W. ## State representation - Persistent state: Win threshold W, scores array [s1,…,sn], agent skill ratings [σ1,…,σn]. - Per-turn ephemeral state: Active agent index i, chosen hole h, sampled stochastic roll, current streak length. - Observable information: Fully observable — all player scores, hole options, and swing outcomes are public. ## Action space - `swing_target(hole_type)`: Choose `main_hole`, `side_hole`, or `narrow_eye`. ## Playability notes - Single-player-vs-AI: Natural — 1 human player vs. AI agents balancing risk-reward hole selection (consistent 1-point main hole vs. high-yield 5-point narrow hole). - Group games (n>4): Scales cleanly — turn rotation generalizes to any group size. [gap-fill] Solitaire scoring challenge mode (reach W in fewest misses) is also natively supported. - Physical-skill games: Dynamic hand-eye tracking and wrist flick are abstracted into the parameterized success function. - What is lost: The physical sensation of the ivory pin clicking into polar bear bone and the tactile swing tension of the sinew thong. ## Implementation hints Extremely simple Markov decision process with minimal state footprint. The AI decision problem is a classic bandit/stochastic knapsack: when to choose the risky 5-point hole to pass an opponent vs. safe 1-point grinding. First milestone: implement a 2-agent CLI game logging points and streaks to 20 points.

Download as JSON

Download the full game corpus entry for this game as structured JSON.

Download JSON
Cite this page

Greenlandic Ring-and-Pin (Greenland). (n.d.). Games of Countries. https://gamesofcountries.com/games/gl-greenland/ajagaq

@misc{gamesofcountries-gl-greenland-ajagaq,
  title = {Greenlandic Ring-and-Pin (Greenland)},
  year = {n.d.},
  publisher = {Games of Countries},
  url = {https://gamesofcountries.com/games/gl-greenland/ajagaq}
}