Skip to main content
Formal Specification

Syrian Leapfrog and Arched Horse — Formal Specification

Syria · strength-endurance · Confidence: medium

← Back to game page

Implementation Suitability

Suitability: adaptation-requiredFidelity: probabilistic-proxyGap-fills: 2
# Al-Hisan al-Muqawwas — AI-optimized specification ## General idea A physical team endurance and balance game. One team forms a structural human "bridge" (the Horse) and the other team leapfrogs and piles onto their backs (the Riders). The strategic element consists of the Riders deciding their landing positions to balance weight distribution, while the Horse manages their fatigue, followed by a hidden finger-guessing riddle that can trigger a role-swap. ## Formal model - **Abstraction taken:** The physical structure of the Horse is modeled as a 1D array of M back segments (where M is the team size, 3≤M≤6). Segment 0 is adjacent to the wall (sturdiest), and segment M−1 is at the rear (weakest). Jumping is represented as a selection of target segments, and stability is determined by a stochastic structural-load resolver. - **Player/agent count:** 2 team agents (Team A and Team B), each managing a squad of M players. - **Turn model:** Phase-based: Team B sequentially mounts Team A, then Team A attempts a structural stability check, followed by a simultaneous riddle phase. ## Formal rules R1. Setup: Team A is designated the Horse; they form a human line of M segments. Segment i has a maximum weight capacity Ci=(M−i)×100 units (nodes closer to the wall have higher capacity). Segment weight loads Li are initialized to 0. R2. Mounting Phase: For each rider j∈[0,M−1] on Team B, they select a jump action `mount(target_segment)` where target_segment∈[0,M−1]. 2.1. [gap-fill] Jump Accuracy: The actual landing segment Sland is resolved stochastically. If target_segment=0, there is a probability Pmiss=0.2 that they overshoot or undershoot and fall off. 2.2. If a rider falls off (or misses all segments), the mounting phase fails immediately. Team B loses the round, and roles swap. 2.3. If they land on Sland, the load on that segment increases by the rider's weight: LSland←LSland+weightj. 2.4. A segment can only support riders whose combined weight does not exceed its capacity, and at most 2 riders can occupy a single segment simultaneously [gap-fill: occupancy constraint]. R3. Structural Integrity Check: 3.1. Once all M riders are mounted, the game engine calculates the collapse probability Pcollapse=∑i=0M−1max⁡(0,(Li−Ci)/Ci). 3.2. A stochastic check is performed. If Pcollapse>rand(0,1), the Horse collapses. Team A loses the round, and Team B wins the right to remain riders in the next round. 3.3. If the Horse does not collapse, the riders are deemed stable. R4. Finger Riddle Phase: 4.1. The rider captain selects a secret integer Fsecret∈[1,5]. 4.2. The horse captain selects a guess integer Fguess∈[1,5]. 4.3. If Fguess==Fsecret, Team A (Horse) wins the round, and roles swap for the next round. 4.4. If Fguess≠Fsecret, Team B (Riders) wins the round, and retains the rider role for the next round. R5. Win / Loss / Draw: 5.1. The game ends after R rounds (e.g. 5 rounds). 5.2. Each team scores 1 point for every round they spend in the "Riders" role. The team with the most points wins. ## Edge cases - Tie at the end condition: Standard, handled by R5.2 (draw declared, or team with fewer collapses wins). - A player with no legal move: In R2, a rider can always choose a valid segment to aim at. - Repetition / cycles: Provably bounded by the finite round limit R. - Simultaneous claims on the same resource: Handled by R2.4 (maximum of 2 riders per segment, resolved in order of jump sequence). - Player elimination mid-round: If a rider falls off, the round ends immediately, and roles swap (R2.2). - Resource exhaustion: None. - Termination guarantee: Guaranteed by the finite round limit R. ## State representation - Persistent state: Capacities array {Ci}, load array {Li}, segment occupancy counters, team roles, round counter, cumulative scores, current phase. - Per-round ephemeral state: Jump landing outcomes, collapse calculation, secret riddle choice, and guess choice. - Observable information: Both teams see capacities, current loads, occupancies, and scores. The secret riddle selection Fsecret is strictly hidden from the horse team until the guess is submitted. ## Action space - `mount(segment_id)`: Select target segment 0..M−1 to land on. - `choose_fingers(num)`: Select Fsecret∈[1,5] (Riders only). - `guess_fingers(num)`: Select Fguess∈[1,5] (Horse only). ## Playability notes - Single-player-vs-AI: Excellent. An AI Rider team can run simple optimization heuristics to distribute their player weights evenly across the Horse segments to minimize Pcollapse while avoiding over-crowding. An AI Horse team can guess the riddle using uniform probability or by analyzing historical human choices. - Group games (n>4): Natively supported. The squad structure generalizes to any group size 6≤N≤12. - Physical-skill games: Physical balance, strength, and leaping are modeled by the segment capacities, player weights, and accuracy probabilities. - What is lost: The visceral sensation of impact, the intense strain of holding weight, and the physical laughter during a collapse. ## Implementation hints - Simple turn-based state machine. The physics are completely abstracted into segment capacities and a collapse probability equation. - First milestone: Implement the mounting loop, verify segment weight accumulation, and run the collapse/riddle resolution logic.

Download as JSON

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

Download JSON
Cite this page

Syrian Leapfrog and Arched Horse (Syria). (n.d.). Games of Countries. https://gamesofcountries.com/games/sy-syria/al-hisan-al-muqawwas

@misc{gamesofcountries-sy-syria-al-hisan-al-muqawwas,
  title = {Syrian Leapfrog and Arched Horse (Syria)},
  year = {n.d.},
  publisher = {Games of Countries},
  url = {https://gamesofcountries.com/games/sy-syria/al-hisan-al-muqawwas}
}