Skip to main content
Formal Specification

Ducks and Hunters (Romanian Dodgeball) — Formal Specification

Romania · ball-team · Confidence: high

← Back to game page

Implementation Suitability

Suitability: engine-readyFidelity: discretizedGap-fills: 2
# Rațele și vânătorii — AI-optimized specification ## General idea An asymmetric, discrete dodgeball elimination game on a 2D corridor grid. Two hunters at opposite ends pass and target central ducks. Ducks can choose to evade or attempt high-risk aerial catches to earn extra lives or revive eliminated teammates. When one duck remains, a countdown phase initiates. ## Formal model - **Abstraction taken:** Court discretized into a 2D grid (L×W, e.g. 20×8). Hunters occupy end baselines x=0 and x=L−1. Ducks occupy the interior corridor 1≤x≤L−2. Ball trajectory is modeled as discrete line vectors with flight time and hit probabilities. - **Player/agent count:** 2 Hunters (H0,H1) and K Ducks (D1..DK, default K=4). - **Turn model:** Sequential throw events. Between throws, all active ducks position themselves simultaneously on the grid. ## Formal rules R1. State tracking: Per duck i: `(x, y)`, `lives` ∈ℕ (starts at 1), `status` ∈{active,eliminated}. Active hunter Hactive∈{0,1}. R2. Hunter Actions: - Active hunter holding ball selects target coordinate (xtarget,ytarget) and throw type ∈{strike,pass}. - Ball travels along line path from Hactive to opposing baseline. R3. Duck Reaction & Hit Resolution: - If ball trajectory passes through cell of active duck Di: - Duck chooses reaction: `dodge()` or `attempt_catch()`. - If `dodge()`: Hit check: Phit=max⁡(0.10,0.85−0.15×agilityi). - If hit: `D_i.lives -= 1`. If `D_i.lives == 0`, `D_i.status = eliminated`. - If `attempt_catch()`: - [gap-fill] Catch probability: Pcatch=0.50×skilli. - If catch succeeds: `D_i.lives += 1` ("Life gained"). Duck may immediately spend 1 life to revive one eliminated teammate (`status = active`). - If catch fails: Duck is struck directly (`D_i.lives -= 1`). R4. Turn Transition: - Ball reaches opposite baseline; the other hunter acquires the ball (Hactive←1−Hactive). R5. Final Survivor Countdown: - When exactly 1 active duck remains (`active_count == 1`): - A countdown counter `countdown_throws = 10` is initialized. - Each subsequent throw successfully dodged decrements `countdown_throws -= 1`. - If `countdown_throws == 0`: Final survivor wins match; all teammates are restored to active status. - If survivor is hit and has 0 extra lives before reaching 0: Hunters win the round. R6. [gap-fill] Turn Limit: Match capped at 200 total throws. If hunters fail to eliminate all ducks or trigger the countdown win within 200 throws, Ducks win on points. ## Edge cases - Tie at the end condition: Not applicable — countdown resolution or total elimination produces a strict binary winner. - A player with no legal move: Waiting / dodging is always legal. - Repetition / cycles: Bounded by R6 (hard 200-throw ceiling). - Simultaneous claims on the same resource or position: Ducks may share grid cells; if ball intersects multiple ducks in the same cell, the front-most duck resolves the interaction first. - Player elimination mid-round; turn order repair: Eliminated ducks leave the central corridor; when revived, they re-enter at the center cell (L/2,W/2). - Resource exhaustion: Extra lives are non-negative integers; ball is conserved. - Termination guarantee: Enforced by finite countdown (10 throws) and 200-throw overall ceiling. ## State representation - Persistent state: - `ducks`: Array of K structs: `[x, y, lives, status]`. - `active_hunter`: 0 or 1. - `countdown_remaining`: Integer (starts at -1, sets to 10 when 1 duck left). - `total_throws`: Integer counter. - Per-throw ephemeral state: - `ball_trajectory`: Ray vector from (x0,y0) to (x1,y1). - Observable information: Fully observable. ## Action space - Hunter: `throw_strike(target_x, target_y)`, `pass_ball()`. - Duck: `move(dx, dy)`, `dodge()`, `attempt_catch()`, `spend_life_revive(teammate_id)`. ## Playability notes - Single-player-vs-AI: Highly playable as a risk/reward dodgeball simulation. A human duck decides when to risk catching for extra lives. - Group games (n>4): Scales naturally for K=2..12 ducks. - Physical-skill games: Physical ducking and catching translated into stochastic probability trees conditioned on agent agility and timing. - What is lost: Physical adrenaline, stinging ball impacts, and playground roar. ## Implementation hints - Implement as a discrete raycast collision step. - Duck heuristic: If `lives > 1`, take higher risks attempting catches to generate teammate revivals; if `lives == 1`, prioritize pure dodging.

Download as JSON

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

Download JSON
Cite this page

Ducks and Hunters (Romanian Dodgeball) (Romania). (n.d.). Games of Countries. https://gamesofcountries.com/games/ro-romania/ratele-si-vanatorii

@misc{gamesofcountries-ro-romania-ratele-si-vanatorii,
  title = {Ducks and Hunters (Romanian Dodgeball) (Romania)},
  year = {n.d.},
  publisher = {Games of Countries},
  url = {https://gamesofcountries.com/games/ro-romania/ratele-si-vanatorii}
}