Gustavo Fernandes wrote:
I have been trying to use meep to calculate the modes of disk and ring resonators, so I thought I would begin by reproducing the results of the ring resonator example on the Meep tutorial. However, I have not been able to reproduce the harminv output quoted in the tutorial. In general, the output of harminv seems to depend on the resolution, as well as the simulation time. The results also seem to vary between different versions of Meep.
harminv is certainly sensitive to the number of data points that you give it (via the simulation time, and also indirectly by the resolution since the temporal resolution scales with the spatial resolution using the Courant stability criteria of S=(\delta t)/(\delta x) where S=0.5 (by default) in Meep). There increasing the resolution while keeping a fixed simulation time, will increase the number of time steps.
We identified a small bug in the cylindrical code relating to the PML implementation that was fixed in Meep 0.20.4, and so you should see slightly different results between Meep versions 0.20.4 and higher with those below.
The most important criteria for having confidence in your Meep installation is to ensure that 'make check' passes. The example ctl files have not been modified to produce outputs with the latest version of Meep (currently 1.0.3) and so are not a cause for concern.
Ardavan _______________________________________________ meep-discuss mailing list [email protected] http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
