Light hadron masses with a tadpole-improved next-nearest-neighbour lattice fermion action

Calculations of hadron masses are done in quenched approximation using gauge field and fermion actions which are both corrected for discretization errors to $O(a^2)$ at the classical level and which contain tadpole improvement factors. The fermion action has both nearest-neighbour and next-nearest-neighbour couplings in the kinetic and Wilson terms. Simulations done at lattice spacings of $0.27$ and $0.4$fm yield hadron masses which are already quite close to experimental values. The results are compared to Wilson action calculations done at comparable lattice spacings.