Formal Specification
Eritrean Traditional Stone Pitching Target Game — Formal Specification
Eritrea · dexterity-throw · Confidence: low
← Back to game pageImplementation Suitability
Suitability: engine-readyFidelity: discretizedGap-fills: 3
# Kudo — AI-optimized specification
> **Reconstructed source.** The entry carries a provenance warning: its rules were written from
> general knowledge, not read from the two works it cites (Pankhurst 1990; Trevaskis 1960).
## General idea
A traditional Eritrean target pitching game where youths throw flat stones (*Kudo*) at a row of
three upright wooden pegs from an 8-meter distance. Each turn consists of pitching three stones in
succession, scoring two points for a knockdown and one point for a glancing touch. The race to 16 points
rewards consistent projectile precision. The entry leaves unstated whether downed pegs are immediately
restored between throws within a turn.
## Formal model
- **Abstraction taken:** Discretized physical skill. Continuous trajectory and target physics are
abstracted into discrete random outcome distributions over peg hits, knockdowns, and misses.
- **Player/agent count:** 2 to 6 players (entry §Players).
- **Turn model:** Round-robin sequential turns; each turn consists of a sequence of 3 stone casts.
## Formal rules
R1. Setup (entry §1): **3 upright wooden pegs** are placed in a row **8 meters** from the throwing line.
R2. Pitching turn (entry §2): on their turn, a player pitches **3 stones** sequentially at the pegs
[gap-fill: the entry does not specify whether a knocked-down peg is stood back up immediately for
subsequent throws in the same turn, or only reset between player turns. Immediate reset after each
throw is assumed for independent trials].
R3. Scoring (entry §3):
- Knocking down a peg: **2 points**.
- Touching a peg without knocking it down: **1 point**.
- Missing all pegs: **0 points**.
[gap-fill: continuous physical throwing accuracy is modelled via categorical outcome probabilities
for knockdown, touch, and miss].
R4. Victory (entry §4): the **first player to reach 16 points** wins the game
[gap-fill: the entry does not specify whether the round is completed so all players have equal turns
if player 1 reaches 16 first, or whether sudden death applies immediately].
## Edge cases
- Tie at the end: unaddressed; R4 states first to 16 points wins, but if players complete full rounds
and multiple players cross 16 points simultaneously, no tiebreak procedure is defined.
- A player with no legal move: not applicable; every active player on their turn always possesses the
`pitch_stone` action with 3 available casts.
- Repetition / cycles: not applicable; score accumulation is strictly non-decreasing and each cast is an
independent trial with no cyclic board positions.
- Simultaneous claims on one position: not applicable; turns are strictly sequential, one player pitching
at a time.
- Elimination mid-round and turn-order repair: no mid-game elimination; all 2-6 players remain active in
rotation until one reaches 16 points.
- Resource exhaustion: pitching stones are retrieved and reused after each turn; point totals accumulate
monotonically toward the target threshold of 16.
- Termination guarantee: guaranteed with probability 1; because score is strictly cumulative and non-decreasing,
and every throw has a positive probability of scoring >= 1 point, the score crosses 16 in a finite number
of turns almost surely via hazard rate.
## State representation
- Persistent state: `scores[p]` ∈ {0..∞}, `active_player` ∈ {0..n-1}, `turn_number`.
- Per-turn ephemeral state: `stones_thrown` ∈ {0..3}, `turn_score`.
- Observable information: fully observable public scores and peg status.
## Action space
- `pitch_stone()`: active player executes a stone throw, generating a discrete outcome {knockdown, touch, miss}
and updating `scores[active_player]` accordingly (R2, R3).
## Playability notes
- Single-player-vs-AI: fully playable with stochastic agents parameterized by skill distributions.
- Reducing n > 4 to 2 agents: direct and lossless; the mechanics are identical for 2 players.
- Physical-skill games: throwing mechanics are discretized into success probabilities; competitive play
reflects risk/reward calibration.
- What is lost: the physical tactile feel of selecting polished river stones and ground-level reading of
dirt clearings in highland Eritrea.
## Implementation hints
Parameterize throwing accuracy as a 3-outcome categorical distribution `(P_knockdown, P_touch, P_miss)`
that can be adjusted per AI difficulty level. Handle peg resets cleanly: resetting pegs before each individual
cast keeps the three throws within a turn independent and identically distributed. Implement strict sudden-death
or round-completion tiebreaking with a documented flag. Suggested first milestone: 2-player stochastic pitching
simulation verifying monotonic score growth and finite termination.
Download as JSON
Download the full game corpus entry for this game as structured JSON.
Download JSONCite this page
Eritrean Traditional Stone Pitching Target Game (Eritrea). (n.d.). Games of Countries. https://gamesofcountries.com/games/er-eritrea/kudo
@misc{gamesofcountries-er-eritrea-kudo,
title = {Eritrean Traditional Stone Pitching Target Game (Eritrea)},
year = {n.d.},
publisher = {Games of Countries},
url = {https://gamesofcountries.com/games/er-eritrea/kudo}
}