Formal Specification
Traditional cross-country horse racing — Formal Specification
Mongolia · equestrian-team · Confidence: low
← Back to game pageImplementation Suitability
Suitability: engine-readyFidelity: probabilistic-proxyGap-fills: 2
# Khurdan moriny uraldaan — AI-optimized specification
## General idea
A long-distance cross-country horse racing simulation over 10 to 26 kilometres of natural steppe terrain. Each agent manages a horse-jockey pair, allocating stamina across discrete distance segments, choosing line trajectories over terrain features (undulating hills, badlands, flat pasture), and timing the final burst (*shavkh*) while managing heat and fatigue.
## Formal model
- **Abstraction taken:** The continuous open-steppe racecourse is discretized into `K = 100` distance segments (each representing 100–250 m). On each race tick `t`, all agents simultaneously choose a pace setting and trajectory path.
- **Player/agent count:** 2 to n competing pairs (standard simulation: 20 to 100 horses).
- **Turn model:** Synchronized simultaneous ticks until all horses cross the finish line.
## Formal rules
R1. Setup: Race distance `D_total` is set based on age category (e.g. `Soyolon = 24,000 m` -> 100 segments of 240 m). Each horse is initialized with attributes: `Base_Speed`, `Max_Stamina`, `Stamina_Current = Max_Stamina`, `Position = 0`, `Fatigue = 0`.
R2. On each race tick `t`, each jockey agent selects an action: `set_pace(mode)` where `mode in {CANTER, EXTENDED_CANTER, GALLOP, SPRINT_BURST}`.
R3. Speed & Stamina Dynamics:
- Speed on tick `t`: `V_t = Base_Speed * Pace_Multiplier(mode) * (1 - Fatigue/100) + Terrain_Modifier(segment_t) + ε_t`.
- Stamina burn: `Stamina_Burn = Pace_Cost(mode) * Terrain_Resistance(segment_t)`.
- `Stamina_Current = max(0, Stamina_Current - Stamina_Burn)`.
- If `Stamina_Current == 0`: Horse enters "Exhausted" state; speed drops to 40% of baseline, and fatigue accumulates rapidly.
R4. [gap-fill] *Giingoo* Chant Buff: A jockey may invoke the *Giingoo* chant once per race quadrant, recovering 10% stamina and reducing fatigue accumulation for 5 ticks.
R5. Advancement: Each horse's position advances by `Delta_Pos = V_t * dt`.
R6. Finish Line & Rankings:
- When a horse's `Position >= D_total`, its crossing timestamp is recorded.
- Top 5 finishers receive the *Airagdsan tavan mori* titles; 1st place is crowned *Tümniy ekh*.
- [gap-fill] The final horse to cross is recorded as *Bayan khodood*.
R7. Termination: The race terminates when all active horses have crossed the finish line or exceeded a maximum cutoff time.
## Edge cases
- Tie at the end condition: Sub-millisecond continuous position check breaks ties cleanly.
- A player with no legal move: Not applicable — pace choices are always available.
- Repetition / cycles: Positions increase monotonically (`V_t > 0` always), guaranteeing the finish line is reached.
- Simultaneous claims on the same resource: Open steppe allows multiple horses to run parallel without lane collision.
- Player elimination mid-round: If a horse suffers injury or unseats jockey, position progress halts; other horses continue uninterrupted.
- Resource exhaustion: Stamina depletion reduces horse speed to a slow trot but does not halt forward progress.
- Termination guarantee: Monotonic position advancement ensures `Position >= D_total` in finite ticks.
## State representation
- Persistent state: Per-horse structs `{id, pos, speed, stamina, fatigue, pace_mode, finished, finish_time}`, `race_distance`, `tick_counter`, `top_5_leaderboard`.
- Per-tick ephemeral state: Jockey pace commands, sampled terrain perturbations `ε_t`.
- Observable information: Positions of nearby horses, own horse stamina/fatigue gauges, remaining distance, and weather conditions.
## Action space
- `select_pace(pace_level)`: Pace level in `{CANTER, EXTENDED_CANTER, GALLOP, SPRINT}`.
- `chant_giingoo()`: Trigger cooldown-bounded stamina boost.
## Playability notes
- Single-player-vs-AI: Natural — 1 human jockey vs. a pack of simulated nomadic AI racers managing pace and energy.
- Group games (n>4): Highly scalable — multi-agent particle simulation running hundreds of competitors simultaneously.
- Physical-skill games: Energy management and endurance pacing optimization.
- What is lost in the adaptation: The raw physical dust, equine hoofbeats vibrating the earth, horse breath, and high-altitude steppe wind.
## Implementation hints
Discrete time-stepped physics integrator (`dt = 1.0` second per tick). Milestone 1: single-horse stamina-burn curve over 100 segments; Milestone 2: multi-agent race with dynamic pacing AI and leaderboard tracking.
Download as JSON
Download the full game corpus entry for this game as structured JSON.
Download JSONCite this page
Traditional cross-country horse racing (Mongolia). (n.d.). Games of Countries. https://gamesofcountries.com/games/mn-mongolia/khurdan-moriny-uraldaan
@misc{gamesofcountries-mn-mongolia-khurdan-moriny-uraldaan,
title = {Traditional cross-country horse racing (Mongolia)},
year = {n.d.},
publisher = {Games of Countries},
url = {https://gamesofcountries.com/games/mn-mongolia/khurdan-moriny-uraldaan}
}