On 10/26/2017 02:41 PM, Phuc Hoang wrote:
Does anyone know if "synchronized-magnetic" is implicitly implemented when (add-flux) is called?
As described in the documentation <http://meep.readthedocs.io/en/latest/Introduction/#transmissionreflection-spectra>, the add-flux routine involves accumulating the discrete-time Fourier transform (DTFT) of the electric and magnetic fields at each time step for every point in the flux plane and then computing the flux at the end of the time stepping. add-flux does not synchronize the electric and magnetic fields when computing the DTFT fields since synchronizing the fields at every time step would greatly increase the cost of timestepping. This means that the Poynting flux computed by add-flux is not second-order accurate in time.
The electric and magnetic fields are currently only automatically synchronized when computing the instantaneous flux (via flux-in-box) or total energy (via field-energy-in-box). For reference, see the file src/energy_and_flux.cpp <https://github.com/stevengj/meep/blob/master/src/energy_and_flux.cpp> where these functions are defined.
_______________________________________________ meep-discuss mailing list meep-discuss@ab-initio.mit.edu http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
