Formal Specification
Shoulder Stone Put — Formal Specification
Serbia · strength-endurance · Confidence: low
← Back to game pageImplementation Suitability
Suitability: engine-readyFidelity: abstractedGap-fills: 2
# Bacanje kamena s ramena — AI-optimized specification
## General idea
A sequential athletic strength and trajectory optimization game. Each agent executes 3 stone-putting attempts, selecting release parameters (effort level, launch angle, and approach aggressiveness) to maximize ground distance while managing the risk of overstepping the baseline foul line. The highest single valid throw wins.
## Formal model
- **Abstraction taken:** Physical biomechanics (muscular explosive force, shoulder angle, foot plant) are modeled via a ballistic range equation with parameterized skill, effort, and foul probability distributions:
, where .
- **Player/agent count:** individual agents (, default ).
- **Turn model:** Strictly sequential flights of 3 rounds; in each round, every agent throws once.
## Formal rules
R1. Initialization: Agents are indexed . Each agent maintains a record of throw distances: `attempts[i] = [0.0, 0.0, 0.0]`. Total rounds . Stone mass .
R2. Throw Action: In each round , active agent submits action parameters:
- `effort \in [0.5, 1.0]` (percentage of maximum power).
- `angle \in [35.0, 50.0]` (launch angle in degrees; optimal ).
- `risk_mode \in [conservative, standard, aggressive]` (proximity to scratch line).
R3. Execution and Foul Check:
- Foul check: .
- If foul occurs (stepping over line): `d = 0.0` (foul recorded).
- If valid: Launch velocity . Distance calculated: .
- Record attempt: `attempts[i][r-1] = d`.
R4. Best Mark Evaluation: Each agent's score is `best_score[i] = max(attempts[i])`.
R5. [gap-fill] Tiebreaking: If two or more agents tie on their best throw, their second-best valid throws are compared. If still tied, a 1-throw sudden-death shoot-off is executed.
R6. Victory: The agent with the highest `best_score` after 3 rounds (and tiebreakers) is declared the winner (*Arambaša*).
## Edge cases
- Tie at the end condition: Fully resolved by R5 (comparison of 2nd best throws, then sudden death).
- A player with no legal move: Not applicable; an agent must submit an action tuple `(effort, angle, risk_mode)`.
- Repetition / cycles: Bounded; the competition consists of exactly discrete throws.
- Simultaneous claims on the same resource or position: Not applicable; strictly sequential throws.
- Player elimination mid-round / turn order repair: [gap-fill] If an agent withdraws mid-competition, their recorded throws remain on the leaderboard; future rounds skip their slot.
- Resource exhaustion: Not applicable; the stone is retrieved after each throw.
- Termination guarantee: Guaranteed by the strict 3-round execution limit.
## State representation
- Persistent state: `attempts: dict[AgentID, list[float]]`, `round_index: int`, `active_agent: int`.
- Per-throw ephemeral state: `sampled_distance: float`, `foul_flag: bool`.
- Observable information: Fully observable public leaderboard.
## Action space
- `put_stone(effort: float, angle_deg: float, risk_mode: enum[conservative, standard, aggressive])`:
- `effort`: Float .
- `angle_deg`: Float .
- `risk_mode`: Controls trade-off between release distance boost and scratch-line foul probability.
## Playability notes
- Single-player-vs-AI: Natural; player balances risk vs reward across 3 attempts while monitoring AI leaderboard marks.
- Group games (n>4): Natively supports flight sizes up to 20 competitors.
- Physical-skill games: Muscular power and scratch line spatial margin are mapped to ballistic equations and logistic foul functions.
- What is lost in the adaptation: The visceral bodily exertion, grip on cold stone, and crowd cheering at the Guča stadium are abstracted into mathematical parameters.
## Implementation hints
Extremely clean athletic simulation model. First milestone: implement 3-round deterministic leaderboard tracking with static ballistic calculations, then calibrate the stochastic noise and foul curve.
Download as JSON
Download the full game corpus entry for this game as structured JSON.
Download JSONCite this page
Shoulder Stone Put (Serbia). (n.d.). Games of Countries. https://gamesofcountries.com/games/rs-serbia/bacanje-kamena-s-ramena
@misc{gamesofcountries-rs-serbia-bacanje-kamena-s-ramena,
title = {Shoulder Stone Put (Serbia)},
year = {n.d.},
publisher = {Games of Countries},
url = {https://gamesofcountries.com/games/rs-serbia/bacanje-kamena-s-ramena}
}