Chess is one of those games that allows for enormous complexity while having a comparatively small ruleset. The size of the possibility space of any game can be measured, assuming that the rules of the game are followed correctly. This conceptual space contains all possible outcomes and states and can be n-dimensional, where n is the number of variables to represent an outcome. This however does not mean a higher number of dimensions also makes for a bigger possibility space. In some games its relatively straightforward to measure the size of the space, like Tic-Tac-Toe, which has 765 different games. Scaling the 3x3 grid of Tic-Tac-Toe up to the 8x8 grid of Chess alone already increases the size of the space exponentially. When you add the different types of playing pieces and their movements into the equation… well, you may want to start over, because Chess is a completely different kind of game. The number of different legal board states in Chess is somewhere around 1040, and the amount of possible games is many, many orders of magnitude larger.
For a Chess playing artificial intelligence, considering the whole possibility space at once wouldn't bring any closer to winning. In the case of Stockfish — at the moment of writing the world's best open source Chess AI — it takes the current board state, calculates all the possible legal moves of said state and all legal moves the opponent could take for each of those, and so on and so on. This results a in gigantic tree through which the AI looks for the highest scoring series of moves.
This tree structure is something I wanted to utilise to visualise the possibility space of the game. Instead of only being able to see one board state, you see all possible board states at the same time. The example below is a short game resulting in a variant of Scholar's Mate:
Experience this tool for yourself at nielspoldervaart.nl/chess.