Comment on "Calculation of microcanonical entropy differences from configurational averages"

We introduce a simple improvement on the method to calculate equilibrium entropy differences between classical energy levels proposed by Davis (S. Davis, Phys. Rev. E, 050101, 2011). We demonstrate that the modification is superior to the original whenever the energy levels are sufficiently closely spaced or whenever the microcanonical averaging needed in the method is carried out by importance sampling Monte Carlo. We also point out the necessary adjustments if Davis's method (improved or not) is to be used with molecular dynamics simulations.