On May 30, 2009, at 3:27 AM, matt wrote:
Thanks for your explanation. Just to clarify, could you elaborate as to why the convergence is first order with dispersion, but second order otherwise?
The standard FDTD method uses a center-difference approximation for derivative that is nominally second-order accurate. However, when you have a discontinuous interface this second-order accuracy breaks down, and you get only first-order accuracy (in the L2-norm sense). Meep's subpixel averaging restores second-order accuracy in the absences of sharp corners (and gets better than first-order accuracy even with sharp corners), but the subpixel averaging is currently only done for the nondispersive part of epsilon and mu.
Steven _______________________________________________ meep-discuss mailing list [email protected] http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
