Formal Specification
Human Rooster Fight (Morog Lorai / Morog Juddho) — Formal Specification
Bangladesh · wrestling-combat · Confidence: high
← Back to game pageImplementation Suitability
Suitability: engine-readyFidelity: probabilistic-proxyGap-fills: 3
# Morog Lorai — AI-optimized specification
## General idea
A traditional Bengali folk combat and balance game simulating a rooster fight (*Morog Juddho*). Competitors adopt a mandatory one-legged stance: tucking one leg behind the back and gripping the ankle with a hand, hopping across a circular arena to deliver forceful shoulder checks and torso bumps. Competitors are eliminated if their raised foot touches the ground, if they release their grip, or if they step outside the ring, until only one standing champion survives.
## Formal model
- **Abstraction taken:** Physical single-leg hopping, collision momentum, and balance recovery are modeled as 2D spatial positioning with a dynamic `Stability` float (). Collisions between hopping agents transfer impulse force that drains stability; if stability drops below zero, a foot-drop fall occurs.
- **Player/agent count:** 6 to 12 agents in free-for-all battle royale (or 2 agents in a 1v1 duel).
- **Turn model:** Real-time ticked simultaneous movement (1 tick = 0.25 sec). Combat continues until competitor remains standing.
## Formal rules
R1. Setup:
- Circular arena of radius centered at .
- competitors start standing on one leg with opposite ankle firmly gripped behind back.
- Each competitor starts with `stability = 100` and `stamina = 100`.
R2. Hopping Movement:
- On each tick, active agents submit a hop vector with max hop speed .
- Hopping consumes 0.5 stamina per tick. If stamina , movement speed is reduced by .
R3. Collision & Shoulder Check (*কাঁধের ধাক্কা*):
- [gap-fill] When two agents and move within contact distance (), a shoulder collision resolves:
- If both agents charge each other, mutual stability loss occurs proportional to opposing momentum.
R4. Elimination Triggers (*আউট*):
- Foot Touch / Fall: If an agent's `stability <= 0`, their raised foot touches the ground or they fall; they are eliminated immediately.
- [gap-fill] Grip Slip: On taking heavy impulse (), there is a chance of grip release (), triggering instant elimination.
- Ring Out: If an agent's coordinates exceed radius , they are eliminated immediately.
R5. Posture Recovery:
- When an agent is out of contact ( from all opponents), they recover stability per tick up to a max of 100.
R6. Ring Shrinking Mechanism:
- [gap-fill] To prevent passive stalling, if players remain after 120 ticks (30 seconds), the arena radius shrinks at a rate of down to a minimum of .
R7. Win Condition: The last remaining active competitor standing inside the arena is declared the tournament champion.
## Edge cases
- Tie at the end condition: If the final two competitors knock each other down or step out on the exact same tick, a 1v1 sudden-death rematch round is held.
- A player with no legal move: Hopping vectors within speed limits are always legal for active competitors.
- Repetition / cycles: Bounded strictly by the shrinking ring radius (R6) and stamina/stability decay.
- Simultaneous claims on the same resource or position: Multi-agent collision resolves pairwise momentum vectors simultaneously.
- Player elimination mid-round; turn order repair: Eliminated players are removed from active physics collision lists immediately.
- Resource exhaustion (pieces, tokens, space, time): Ring boundary shrinking enforces spatial compression, guaranteeing terminal collisions.
- Termination guarantee: Arena boundary reduction to forces physical contact, guaranteeing stability depletion and a single winner.
## State representation
- Persistent state:
- `active_competitors_count`: Integer .
- `current_radius`: Float .
- `tick_counter`: Integer .
- `champion_id`: Player ID of the surviving winner.
- Per-turn ephemeral state:
- `competitor_states[N]`: Array of `{id, x: float, y: float, stability: float, stamina: float, is_active: bool}`.
- `collision_events`: List of pairwise impacts resolved on current tick.
- Observable information: Fully observable 2D arena geometry, opponent positions, velocities, and estimated stability levels.
## Action space
- `hop_move(vx, vy)`: Submit directional hop velocity vector within max speed bounds.
- `brace_impact()`: Reduce movement speed to zero to increase effective balance against incoming bumps ( stability resistance).
## Playability notes
- Single-player-vs-AI: Thrilling arcade physics battle royale; player dodges aggressive chargers, flanks unstable opponents, and controls arena center.
- Group games (n>4): Naturally supports 6 to 12 autonomous agents in a free-for-all melee.
- Physical-skill games: Single-leg balance and shoulder collision leverage are parameterized through 2D rigid-body impulse equations.
- What is lost in the adaptation: The hilarious physical wobble and intense burning burn in the supporting calf muscle.
## Implementation hints
Moderate complexity. The game is essentially a 2D circle-clamped particle physics simulation with impulse collision checks and stability state tracking. Milestone 1: implement 2D boundary clamping, hop velocity integration, and pairwise collision impulse; Milestone 2: build an AI agent that navigates away from ring edges and targets low-stability opponents.
Download as JSON
Download the full game corpus entry for this game as structured JSON.
Download JSONCite this page
Human Rooster Fight (Morog Lorai / Morog Juddho) (Bangladesh). (n.d.). Games of Countries. https://gamesofcountries.com/games/bd-bangladesh/morog-lorai
@misc{gamesofcountries-bd-bangladesh-morog-lorai,
title = {Human Rooster Fight (Morog Lorai / Morog Juddho) (Bangladesh)},
year = {n.d.},
publisher = {Games of Countries},
url = {https://gamesofcountries.com/games/bd-bangladesh/morog-lorai}
}