Formal Specification
Tulip Gathering — Formal Specification
Tajikistan · chase-tag · Confidence: low
← Back to game pageImplementation Suitability
Suitability: engine-readyFidelity: discretizedGap-fills: 1
# Lolachinakon — AI-optimized specification
## General idea
A synchronized multi-agent chase-and-tag game. Child agents form rigid 3-player trios holding hands, advancing in cadence toward a boundary line during a singing phase. When the sleeping catcher (*Mavluda*) triggers an awakening sprint, all trios must retreat across the safety baseline without breaking their hand-linked formation or being tagged.
## Formal model
- **Abstraction taken:** Trio formation is modeled as a 3-agent composite kinematic entity with a maximum inter-agent distance constraint and speed penalty for broken links.
- **Player/agent count:** 1 Catcher agent + Trios (each consisting of 3 agents).
- **Turn model:** Two-phase state machine (Phased Singing / Rhythmic Advance Real-time Retreat & Chase).
## Formal rules
R1. Setup: Safety baseline at , Ditch line at . Catcher rests at . Trios initialize along .
R2. Phase 1 (Singing & Advance): Trios advance at fixed cadence . Inter-agent distance within a trio must satisfy units.
R3. Trigger Event: When the lead trio reaches , the singing phase ends. The catcher becomes active and executes `awaken_sprint()`.
R4. Phase 2 (Retreat): All trios must turn and sprint toward . Catcher sprints at ; unbroken trios sprint at .
R5. Tagging: If the catcher reaches within distance unit of any agent before that agent crosses , that agent's entire trio is captured.
R6. Formation Penalty: If any trio's inter-agent distance exceeds 2.0 units during retreat, the formation breaks, reducing their speed to .
R7. [gap-fill] Captured trio provides the next round's catcher. Game runs for 5 rounds; trio with zero captures wins.
## Edge cases
- **Tie at the end condition:** Team with lowest capture count wins.
- **A player with no legal move:** Movement is always legal.
- **Repetition / cycles:** Bounded by round structure.
- **Simultaneous claims on the same resource:** First agent reached by catcher triggers capture of that trio.
- **Player elimination mid-round:** Capture resets round state.
- **Resource exhaustion:** Not applicable.
- **Termination guarantee:** Guaranteed by 5-round limit.
## State representation
- Global state: Phase (`singing_advance`, `retreat_chase`), round timer, catcher state, trio formation status.
- Per-agent state: Position , trio ID, role (`catcher`, `trio_member`), formation integrity flag.
- Observable information: Fully observable 2D state.
## Action space
- Catcher: `awaken_sprint()`, `pursue(target_trio_id)`.
- Trio agents: `advance_rhythm()`, `sprint_retreat(direction)`, `maintain_grip()`.
## Playability notes
- **Single-player-vs-AI:** Player controls central runner of a trio coordinating steering and grip retention.
- **Group games (n>4):** Excellent cooperative multi-agent test bed.
- **Physical-skill adaptation:** Rigid linkage constraint simulates physical hand-holding.
- **What is lost:** Vocal singing unison and flower plucking gestures.
## Implementation hints
Treat trios as rigid bodies with spring-damper constraints between the 3 agents.
Download as JSON
Download the full game corpus entry for this game as structured JSON.
Download JSONCite this page
Tulip Gathering (Tajikistan). (n.d.). Games of Countries. https://gamesofcountries.com/games/tj-tajikistan/lolachinakon
@misc{gamesofcountries-tj-tajikistan-lolachinakon,
title = {Tulip Gathering (Tajikistan)},
year = {n.d.},
publisher = {Games of Countries},
url = {https://gamesofcountries.com/games/tj-tajikistan/lolachinakon}
}