Formal Specification
Powder play (Fantasia) — Formal Specification
Morocco · animal-contest · Confidence: high
← Back to game pageImplementation Suitability
Suitability: adaptation-requiredFidelity: probabilistic-proxyGap-fills: 3
# Tbourida — AI-optimized specification
## General idea
A high-synchronization equestrian troop performance game where a team of mounted cavalry agents must execute a coordinated acceleration, maintain strict linear formation along a 200-meter course, and execute a simultaneous trigger pull within a millisecond-precision timing window. Scoring maximizes synchronization, formation rigidity, and shot coincidence while severely penalizing outliers.
## Formal model
- **Abstraction taken:** The continuous biomechanics of equine galloping, flintlock ignition delay, and sand footing are abstracted into a multi-agent discrete-time synchronization model ( ticks across 3 phases: Entry, Gallop/Aim, Discharge).
- **Player/agent count:** 1 troop of agents (, default ), led by designated agent (*m'qaddem*). Competing troops take sequential runs.
- **Turn model:** Synchronous real-time ticks (, 200 ticks total run) where all agents submit speed/alignment/aim inputs simultaneously per tick.
## Formal rules
R1. **Setup & Initialization:**
- agents are placed in a horizontal line at , spaced evenly at interval .
- Each agent begins with position , velocity , aim state , and musket loaded status = true.
R2. **Phase 1: Entry (*Hadda*):**
- Ticks : Troop advances at steady parade speed (). Agents must match velocity and maintain line alignment within tolerance .
R3. **Phase 2: Charge & Aim (*Talqa*):**
- At tick signaled by , all agents transition to full acceleration ().
- Between and , agents must execute the `raise_and_aim_musket()` action within a synchronized window of ticks.
R4. **Phase 3: Firing Command & Salvo (*El-Baroud*):**
- At , agent triggers the global command `fire_salvo()`.
- Every agent must execute `pull_trigger()` at timestamp .
- Firing discrepancy .
- [gap-fill] If tick (), salvo is classified as *Perfect Unified Blast* (+50 pts). If ticks, *Minor Discrepancy* (+20 pts). If ticks, *Fractured Salvo* (−30 pts).
R5. **Deceleration & Completion:**
- Ticks after firing: All agents execute `decelerate()` to reach before the end barrier at .
R6. **Score Evaluation:**
- .
- [gap-fill] Any horse collision, horse leaving lane (), or misfire gives a run penalty of pts.
## Edge cases
- **Tie at the end condition:** In tournament mode between competing troops, ties are broken by lowest firing time variance ; if still tied, broken by highest formation rigidity score.
- **A player with no legal move:** Not applicable — every agent always has legal acceleration, aim, and trigger actions.
- **Repetition / cycles:** Impossible — the course position increases monotonically; maximum run length is fixed at 200 ticks.
- **Simultaneous claims on the same resource or position:** Lane boundaries prevent spatial overlap; collision resolver applies repulsive lateral vector if .
- **Player elimination mid-round; turn order repair:** If an agent suffers a misfire or fall, [gap-fill] that agent's individual score is nullified, a deduction applies to the troop, and the remaining agents complete the run.
- **Resource exhaustion:** One single powder charge per musket; once fired, status becomes unloaded.
- **Termination guarantee:** The simulation strictly terminates at ticks or when all horses cross and decelerate to zero.
## State representation
- **Persistent state:** Troop ID, run number, overall tournament score table.
- **Per-agent state:** Position , velocity , aim angle , musket status , trigger timestamp .
- **Per-tick ephemeral state:** Leader signal state , current line variance .
- **Observable information:** Fully observable state among all troop members (positions, velocities, and leader voice cues are broadcast).
## Action space
- `adjust_velocity(acceleration)`: float .
- `steer_lane(delta_x)`: float .
- `aim_musket(elevation, pitch)`: target orientation.
- `pull_trigger()`: instantaneous binary event (valid only when musket is aimed).
## Playability notes
- **Single-player-vs-AI:** Highly engaging: the human controls the *m'qaddem* (issuing pace and firing cues) or a wing rider, while AI agents dynamically attempt to match alignment and firing commands.
- **Group games ():** Natural cooperative multi-agent task: agents optimize a joint reward function prioritizing group synchronization over individual speed.
- **Physical-skill games:** Physical barrel stability and trigger timing modeled as stochastic function .
- **What is lost in the adaptation:** The sensory majesty of charging Barb stallions, smell of burnt black powder, and equestrian physical balance.
## Implementation hints
- Implement as a discrete 2D physics runner with a high tick-rate ( or ) to ensure precise detection of sub-second trigger pull synchrony.
- First milestone: Implement 11-agent linear formation maintenance along a 200m track with constant acceleration, then layer the synchronized trigger event.
Download as JSON
Download the full game corpus entry for this game as structured JSON.
Download JSONCite this page
Powder play (Fantasia) (Morocco). (n.d.). Games of Countries. https://gamesofcountries.com/games/ma-morocco/tbourida
@misc{gamesofcountries-ma-morocco-tbourida,
title = {Powder play (Fantasia) (Morocco)},
year = {n.d.},
publisher = {Games of Countries},
url = {https://gamesofcountries.com/games/ma-morocco/tbourida}
}