A player may move a piece from square ( A ) to ( B ) in superposition only if both paths are legal classical moves from distinct board states. The piece exists on ( A ) and ( B ) simultaneously.
Classical chess is (Fraenkel & Lichtenstein, 1981). Quantum Chess, however, introduces non-deterministic branching without decoherence until measurement. quantum chess
In classical chess, a fork (e.g., a knight attacking two pieces) forces the opponent to choose which to save. In quantum chess, a fork allows the attacker to place their piece in superposition, attacking both simultaneously. The defender cannot block both because blocking collapses the wavefunction. A player may move a piece from square
A Quantum Chess position can be represented as a vector of ( 2^64 ) complex amplitudes. A move is a sparse unitary matrix of size ( 2^64 \times 2^64 ). Simulating the game classically requires exponential time. However, a quantum computer could store the state naturally, but the branching factor remains combinatorial. Therefore, the game is PQC-complete – no known polynomial-time quantum algorithm exists to solve optimal play. The defender cannot block both because blocking collapses
The central thesis of this paper is that Quantum Chess is not a stochastic analog of chess but a distinct mathematical structure. While classical chess belongs to (solved via brute-force search), Quantum Chess introduces non-classical correlations that preclude direct tree search, placing it in a unique category of PQC-complete .
| Variant | Deterministic? | Perfect Info? | Entanglement? | Complexity | |---------|----------------|---------------|---------------|-------------| | Classical | Yes | Yes | No | EXPTIME | | Three-check | Yes | Yes | No | EXPTIME | | Chess960 | Yes | Yes | No | EXPTIME | | Progressive | Yes | Yes | No | EXPTIME | | | No | Partial | Yes | PQC-complete |
The following rules are canonical for two-player Quantum Chess: