Strict kinetic energy test.

Only the VR algorithm samples the expected kinetic energy distribution. This table lists the p-values of the kinetic energy validation. All results are from a water system of 900 TIP3P molecules. The temperature was controlled using either the weak-coupling (WC) or the velocity-rescale (VR) algorithm, complemented in the NPT case by a weak-coupling and a Parrinello-Rahman (PR) pressure-control algorithm, respectively. For all state points, the null hypothesis that the simulations controlled by the WC algorithms sampled the expected kinetic energy distribution can easily be rejected. For the simulations controlled by the VR/PR algorithms, however, the p-values indicate a high similarity between the sampled and the expected distribution.