Skip to main content
Formal Specification

Snatching the Scarf (High Jump Prize Snatch) — Formal Specification

Turkmenistan · dexterity-throw · Confidence: low

← Back to game page

Implementation Suitability

Suitability: adaptation-requiredFidelity: probabilistic-proxyGap-fills: 4
# Ýaglyga towusmak — AI-optimized specification ## General idea A traditional festival vertical leap and prize-snatching contest. A decorative silk scarf (*ýaglyk*) is suspended from an elevated horizontal bar at increasing heights; contestants take a running approach and execute an explosive vertical leap to snatch the prize cleanly in mid-air. The game centers on approach speed calibration, takeoff plant timing, and jumping stamina management across escalating height tiers. ## Formal model - **Abstraction taken:** Physical runway sprinting, plant biomechanics, and mid-air hand grasping are modeled as an escalating height threshold ladder with a stochastic jump-height resolver Hjump∼𝒩(μ(vapproach,spring_skill),σ2). An attempt succeeds if Hjump≥Htarget. - **Player/agent count:** 2 to n individual jumpers. - **Turn model:** Strictly sequential turns ordered by player rotation at each height tier. ## Formal rules R1. **Height Setup & Ladder:** - The contest begins at baseline height H1=2.40 m. - [gap-fill] Height increases by ΔH=+0.05 m (5 cm) each round after all active jumpers have completed their attempts. R2. **Jump Attempt Execution:** - On each attempt, the active jumper commits an approach speed v∈[vmin,vmax] and a takeoff timing parameter τ∈[0,1]. - [gap-fill] The effective vertical leap height is generated: Hjump=Hreach+α⋅v2⋅sin⁡(θplant)+ϵtiming, where ϵtiming∼𝒩(0,σ2(skill)). R3. **Snatch Success & Prize Capture:** - If Hjump≥Htarget, the attempt is **SUCCESS**; the jumper snatches the scarf, scores points equal to round(Htarget×100), and advances to the next height tier. - If Hjump<Htarget, the attempt is **FAIL**. R4. **Elimination Protocol:** - [gap-fill] Each jumper is allowed a maximum of 2 failed attempts at a given height tier before being eliminated from the competition (*oýundan çykmak*). R5. **Victory Condition:** - The competition concludes when all jumpers are eliminated, or when the final "Grand Scarf" (*Uly ýaglyk*) at height Hmax=3.20 m is captured. - The jumper with the highest cleared target height (and highest points) is crowned Champion Jumper (*Iň ýokary böküji*). - [gap-fill] Ties at the highest cleared height are broken by the fewest total failed attempts across the entire tournament ladder. ## Edge cases - Tie at the end condition: Handled by R5 — fewest total misses breaks ties; if still equal, a jump-off at +2.5 cm is conducted. - A player with no legal move: Not applicable — committing approach velocity and plant angle is always available on one's turn. - Repetition / cycles: Bounded strictly by monotonic height escalation Hk+1=Hk+ΔH and finite elimination limits (2 misses per height). - Simultaneous claims on the same resource or position: Not applicable — attempts are strictly individual and sequential. - Player elimination mid-round; turn order repair: Eliminated jumpers are removed from the active queue for subsequent height rounds; queue maintains original relative order. - Resource exhaustion (pieces, tokens, space, time): Height ladder has a physical ceiling (Hmax); human jump limits enforce inevitable elimination. - Termination guarantee: why the game provably ends: Guaranteed by strictly increasing height H→∞ combined with bounded human leaping capability P(clear|H)→0. ## State representation - Persistent state: Current target height Htarget, height tier index k, leaderboard `Dict[AgentID, Tuple[float, int]]` (best height, total misses), active jumper queue. - Per-player state: `standing_reach: float`, `skill_vertical_spring: float`, `misses_at_current_tier: int [0, 2]`, `is_eliminated: bool`, `highest_cleared: float`. - Per-turn ephemeral state: Active jumper ID, chosen approach speed, calculated leap height Hjump, clearance outcome `{SUCCESS, FAIL}`. - Observable information: Fully observable — all heights, attempt results, and score histories are public. ## Action space - `AttemptLeap(approach_speed, plant_angle)`: Execute approach run and vertical jump. Preconditions: Jumper is active and `misses_at_current_tier < 2`. ## Playability notes - Single-player-vs-AI: Natural — 1 human jumper competes in a tournament ladder against AI jumpers with varying physical attributes. - Group games (n>4): Sequential tournament format naturally scales from 2 to 30+ participants. - Physical-skill games: Explosive plant mechanics, leg spring, and mid-air grasp timing are modeled through the kinetic energy conversion equation with stochastic execution noise. - What is lost in the adaptation: The visual beauty of the colorful silk scarf fluttering in the desert wind, the rhythmic clapping of the festival crowd, and the physical thrill of the apex grab are converted into numerical threshold comparisons. ## Implementation hints Low complexity. Standard track-and-field high-jump ladder state machine. Branching factor is continuous (speed/angle) but easily discretized into discrete approach speed choices. Suggested first milestone: implement the sequential height escalation ladder with the R2 stochastic jump equation and miss-counter elimination logic.

Download as JSON

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

Download JSON
Cite this page

Snatching the Scarf (High Jump Prize Snatch) (Turkmenistan). (n.d.). Games of Countries. https://gamesofcountries.com/games/tm-turkmenistan/yaglyga-towusmak

@misc{gamesofcountries-tm-turkmenistan-yaglyga-towusmak,
  title = {Snatching the Scarf (High Jump Prize Snatch) (Turkmenistan)},
  year = {n.d.},
  publisher = {Games of Countries},
  url = {https://gamesofcountries.com/games/tm-turkmenistan/yaglyga-towusmak}
}