On Wed, 13 Jun 2007, James Andrews wrote: > I like Meep. > > So I'm putting this question out with some embarassment and > trepidation, but one of my colleagues has strongly requested that I > look into buying a commercial FDTD package that will do what Meep does, > but with a gui interface. We are primarily interested in finite 1-d and > 2-d nonlinear (both the real and imaginary parts) photonic crystals > with interesting dispersive properties.
If you want a GUI code, most people I know seem to use Rsoft, although I haven't used it myself so I can only recommend it by hearsay. For my own part, I'm a long-time Mac user and like GUIs. They are quicker to learn and quicker to get a simulation started than with a scripting interface. However, for scientific simulations I normally don't just run one simulation -- I want to run many simulations, trying different parameters, exploring various designs, optimizing something with respect to something else. And so for me, a scripting interface is more efficient in the long run because it allows me to express many structures with a single input file, parameterized by variables that I can easily change (or loop over, or optimize over...). And furthermore, it lets me put scripts to process the output in custom ways directly in the input file (as opposed to a separate post-processing script in Matlab etc). So, it's a question of long-term laziness (scripting) vs. short-term laziness (GUIs). But to each his/her own. Steven _______________________________________________ meep-discuss mailing list [email protected] http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
