Formal Specification
Five Stones (Al-Khams Hijar) — Formal Specification
Jordan · dexterity-small · Confidence: low
← Back to game pageImplementation Suitability
Suitability: adaptation-requiredFidelity: probabilistic-proxyGap-fills: 2
# Al-Zaqtah — AI-optimized specification
## General idea
A physical dexterity game involving tossing and catching five stones. To adapt this game for AI, the physical act of throwing and catching is replaced by a stochastic resolver based on player skill and task difficulty. This preserves the turn-rotation logic, the progressive difficulty of the five stages, and the penalty for touching adjacent stones.
## Formal model
- **Abstraction taken:** The continuous physical throwing, sliding, and catching of stones is replaced by a discrete series of round-based choices and probability checks.
- **Player/agent count:** 2 to n players.
- **Turn model:** Sequential, with the active player executing stages until they fail, then passing to the next player.
## Formal rules
R1. **Setup:**
1.1. At start, players set their current stage to Stage 1 (*Al-Wahdah*) and their accumulated score to 0.
R2. **Turn Flow:**
2.1. On their turn, the active player attempts to execute their current stage.
2.2. Executing a stage consists of making a series of attempts (e.g. 4 pick-ups for Stage 1, 2 pick-ups for Stage 2, 1 pick-up for Stage 3 and 4).
2.3. Each pick-up is resolved stochastically based on:
- **Player Skill:** A scalar parameter representing dexterity.
- **Stage Difficulty:** Base difficulty increases progressively from Stage 1 to Stage 5.
- **Risk Choice:** [gap-fill] The player can choose a "conservative" or "risky" approach before each attempt, altering the success probability and score yield.
R3. **Fouls and Turn Transfer:**
3.1. If any pick-up attempt fails (e.g., failed catch, incorrect stone count picked up), or if an adjacent stone is touched, a foul (*Al-Harq*) occurs.
3.2. Upon a foul, the player's turn immediately ends, and their current stage progress is frozen. Play passes to the next player in rotation.
3.3. When the player's turn returns, they resume at the start of the stage they failed.
R4. **Progress and Scoring:**
4.1. If a player successfully completes all pick-ups of a stage, they advance to the next stage in their next turn.
4.2. **Stage 5 Resolution (Back of the Hand):**
- The fifth stage is resolved via a custom binomial-style check to determine how many of the 5 stones are caught on the back of the hand (0 to 5 stones).
- If 0 are caught, it is a foul (turn ends).
- Otherwise, those caught stones are tossed again and caught in the palm. Each stone successfully caught in the palm yields 10 points.
4.3. After completing Stage 5, the player's stage resets to Stage 1.
R5. **Winning:**
5.1. The first player to accumulate 100 points wins immediately.
5.2. [gap-fill] If multiple players reach 100 points in the same round rotation, the player with the highest total score wins.
## Edge cases
- **Tie at the end condition:** Handled by R5.2 (highest score wins; if scores are identical, players play a tie-breaker round of Stage 5).
- **A player with no legal move:** Not applicable; a player always has the option to attempt their current stage.
- **Repetition / cycles:** Players must progressively advance and fail; because success probabilities are bounded (P(success) > 0), players will eventually score and reach the target score, avoiding infinite stalemates.
- **Simultaneous claims on the same resource:** Not applicable; turns are sequential.
- **Player elimination mid-round:** Not applicable.
- **Resource exhaustion:** Not applicable.
- **Termination guarantee:** Guaranteed by the strictly increasing scores, a positive lower bound on success probability, and the fixed target score of 100.
## State representation
- **Persistent state:** Current stage (1 to 5) per player, cumulative score per player, turn rotation queue, active player index.
- **Per-turn ephemeral state:** Selected risk choice, attempt sequence, current round points.
- **Observable information:** Fully observed.
## Action space
- `attempt_stage(risk_level)`: Executes the stochastic check for the current stage with a chosen risk parameter ('CONSERVATIVE', 'NORMAL', 'RISKY').
## Playability notes
- **Single-player-vs-AI:** Highly suitable; the player plays against an AI opponent utilizing identical skill-based stochastic resolvers.
- **Group games:** Generalizes cleanly to n players.
- **What is lost in the adaptation:** The physical tactile feedback, muscle-memory speed, and visual tension of tossing real pebbles are fully lost, replaced by a strategic risk-management model.
## Implementation hints
Create a simple success formula: `P(Success) = skill * risk_modifier / stage_difficulty`. Implement the sequential turn flow and stage-progression checks as the first milestone.
Download as JSON
Download the full game corpus entry for this game as structured JSON.
Download JSONCite this page
Five Stones (Al-Khams Hijar) (Jordan). (n.d.). Games of Countries. https://gamesofcountries.com/games/jo-jordan/al-zaqtah
@misc{gamesofcountries-jo-jordan-al-zaqtah,
title = {Five Stones (Al-Khams Hijar) (Jordan)},
year = {n.d.},
publisher = {Games of Countries},
url = {https://gamesofcountries.com/games/jo-jordan/al-zaqtah}
}