Skip to main content
Formal Specification

Golden Gourd (Mounted Archery Target) — Formal Specification

Uzbekistan · equestrian-team · Confidence: low

← Back to game page

Implementation Suitability

Suitability: adaptation-requiredFidelity: probabilistic-proxyGap-fills: 3
# Oltin qoboq — AI-optimized specification ## General idea Oltin qoboq is a martial equestrian archery contest where mounted archers gallop at full speed down a straight lane and shoot arrows upward at a suspended golden gourd or gilded target hanging from a high mast. Structurally distinct from stationary target archery or flat-trajectory horse archery, archers must balance gallop velocity, hands-free rein control, and an acute backward-overhead shooting release ("Parthian shot") within a tight 15-meter spatial shooting window. ## Formal model - **Abstraction taken:** The continuous 3D aerodynamics of arrow flight, bowstring release mechanics, and dynamic equestrian galloping sway are abstracted into a 1D track progression model where the archer chooses gallop pace, aim focus, and draw timing. The shot outcome is resolved via a stochastic target dispersion model centered on the elevated target coordinates. - **Player/agent count:** 1 to N individual mounted archer agents (P1…PN, formal default N=4). - **Turn model:** Sequential individual runs down the track across R=3 rounds [gap-fill: default 3 runs per archer]. ## Formal rules R1. **Track Setup & State:** - Track length: L=120m. Target mast located at Xtarget=60m, target height H=8m. - Shooting window: X∈[52.5m,67.5m] (a 15-meter trigger zone). - Maximum allowed run time: Tmax_run=8.0 seconds (enforcing full gallop; slower runs score 0 points) [gap-fill]. - Cumulative scores initialized: S(Pi)=0 for all i∈{1…N}. R2. **Run Execution:** - On an agent's run, the agent selects: - Gallop velocity v∈[15.0,20.0] m/s (faster gallop reduces run time but increases aim dispersion). - Aim elevation angle θ and draw timing tshot∈[0,Tmax_run]. - If position at shot release X(tshot)∉[52.5,67.5] or total run time L/v>Tmax_run, the shot is invalid → 0 points. R3. **Stochastic Shot Resolution:** - Valid shots generate a 2D impact dispersion (Δx,Δy) centered on the target: (Δx,Δy)∼𝒩(0,σ2(v,archery_skill,horsemanship)) - Distance from target center d=Δx2+Δy2: - **Severing Cord / Bullseye (d≤0.1m):** 5 points. - **Direct Body Hit (0.1m<d≤0.3m):** 3 points. - **Glancing Hit (0.3m<d≤0.5m):** 1 point. - **Miss (d>0.5m):** 0 points. - Run score added to cumulative total: S(Pi)←S(Pi)+points. R4. **Round Progression & Match Winner:** - Every archer completes exactly 1 run per round, sequentially in fixed index order. - After R=3 complete rounds (all archers have taken 3 runs), the archer with the highest cumulative score S(Pi) is declared the champion (*bosh mergan*). R5. [gap-fill] Tiebreaker:** - If multiple archers finish tied for highest score, a single sudden-death run is conducted; if still tied, the archer with the faster run time wins. ## Edge cases - **Tie at the end condition:** Resolved via single sudden-death run followed by run-time tiebreaker (R5). - **A player with no legal move:** Not applicable; an agent must submit a valid velocity and shot timing parameter for each scheduled run. - **Repetition / cycles:** Not applicable; runs progress strictly linearly across finite rounds (N×R total runs). - **Simultaneous claims on the same resource or position:** Not applicable; archers ride down the track strictly one at a time. - **Player elimination mid-round:** If an archer faults or misses, they score 0 for that run and remain in the tournament for subsequent rounds. - **Resource exhaustion:** Arrows and targets are modeled as permanent/reset for each run. - **Termination guarantee:** The contest provably terminates in exactly N×R runs (plus at most 1 sudden-death run). ## State representation - **Persistent state:** - Track parameters (L,Xtarget,H). - Competitor scores array S∈ℕN. - Round index r∈{1…R}, current archer index i∈{1…N}. - Agent skill attributes: archery_accuracy,horsemanship,composure∈[0.0,1.0]. - **Per-run ephemeral state:** - Chosen run speed v, shot timing tshot, sampled impact offset (Δx,Δy), run score. - **Observable information:** - Fully observable: all competitor scores, target hits, and run times are public. ## Action space - `execute_run(gallop_speed, shot_timing)`: Parameters: `gallop_speed` ∈[15.0,20.0], `shot_timing` ∈[2.5,4.5]s. Precondition: agent is the designated active runner. ## Playability notes - **Single-player-vs-AI:** Natural. Single-player mode functions as a high-score marksmanship trial; multi-agent tournament pits human against AI archers with realistic error variance. - **Group games reduction:** Naturally scalable for 1 to 20+ competitors without modification to individual run mechanics. - **Physical-skill games:** Dispersion standard deviation modeled as: σ=base_dispersion×(1.5−archery_skill)×(1.0+0.05×(v−15.0)×(1.0−horsemanship)) - **What is lost:** Physical equine gallop rhythm, muscular bow draw resistance, wind gust variability, and the visceral spectacle of coins bursting from a pierced pumpkin. ## Implementation hints - Model the shooting window as a simple timestamp check: tshot∈[52.5/v,67.5/v]. - Keep the physics simulation 1D along the track, evaluating 2D target offset only at the moment tshot. - First milestone: Implement a 2-archer, 3-round sequential loop with static dispersion to verify scoring and sudden-death tiebreaker logic.

Download as JSON

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

Download JSON
Cite this page

Golden Gourd (Mounted Archery Target) (Uzbekistan). (n.d.). Games of Countries. https://gamesofcountries.com/games/uz-uzbekistan/oltin-qoboq

@misc{gamesofcountries-uz-uzbekistan-oltin-qoboq,
  title = {Golden Gourd (Mounted Archery Target) (Uzbekistan)},
  year = {n.d.},
  publisher = {Games of Countries},
  url = {https://gamesofcountries.com/games/uz-uzbekistan/oltin-qoboq}
}