How it works

The board is a cube of eight stacked 8×8 layers, named A (bottom) to H (top). A cell is written layer, file, row: Ce4 is layer C, file e, row 4.

White starts on layer A along row 1. Black starts on layer H along row 8, at the opposite edge of the cube. Each side has its 8 usual pieces, plus 24 pawns: 8 in front, 8 on top, and 8 in the diagonal gap between them.

Moves

• Rook: any distance in a straight line. It can move along a row, along a file, or straight up and down through the layers.
• Bishop: any distance diagonally within a flat plane, so two coordinates change at once. It stays on its own colour, just like in normal chess.
• Queen: rook plus bishop, plus the corner-to-corner diagonals where all three coordinates change at once.
• King: one step to any of the 26 cells around it.
• Knight: two steps along one axis, then one step along a different axis (24 possible jumps). It leaps over pieces.
• Pawn: one step toward the enemy, either one row forward or one layer toward the enemy. It captures one step forward-and-sideways, up-and-sideways, or forward-and-up. It promotes on reaching the enemy's back row on the enemy's home layer. There is no double first step, no en passant and no castling.

Winning

Normal chess: checkmate wins. You can't make a move that leaves your king in check. Stalemate, threefold repetition and bare kings are draws.

Controls

Drag the cube to orbit it and scroll to zoom. Click a piece in 3D or on the layer panel. Legal moves light up in both views, and the small layer boards show how many targets each layer has. The arrow keys change layers and F shows only the current layer.