“…Thus, for ≥ 5, we obtain the optimal rate of convergence for the homogenization of the Dirichlet problem, which is generically of scale −1 . This is consistent with the generically optimal rate −1 for the periodic setting (see the classical books [8,36] for the derivation, and [32,45,30] for discussions on the optimality of the rates.) It is not clear to us what the optimal rates are when 2 ≤ ≤ 4, which deserve further analysis.…”