Go is a two-player abstract strategy game played on a 19×19 grid (with smaller 9×9 and 13×13 variants for shorter games). Each player places stones to surround territory and capture enemy groups. Despite simple rules, Go has more legal positions than atoms in the observable universe — it's widely considered the deepest abstract strategy game ever invented.
Black plays first, then players alternate placing one stone of their color on any empty intersection. A stone or group with no adjacent empty points (no liberties) is captured. Score by counting territory (empty points surrounded by your stones) plus captured stones. Komi (a points adjustment) compensates White for going second.
Both are abstract strategy with perfect information, but Go favors positional accumulation over tactical exchange. The game tree is far larger; computer mastery only arrived in 2016 with AlphaGo.
You may not immediately recapture a single stone in a way that recreates the previous board position. Prevents infinite capture loops. Larger ko fights are central to high-level strategy.
9×9 — completes in 10–15 minutes and reveals territory and capture concepts cleanly. Move to 13×13 once captures feel natural, 19×19 for full-board strategy.
A points compensation given to White (the second player) at the end. Modern komi is 6.5 or 7.5, fractional to prevent draws.
Sign in with GitHub to share strategies, ask questions, or report a bug.