As a follow up to my last post I’m going to talk about another modelling package…

Meep

Having said what I said last time (my last post) about finite element method packages I have come across a very nice tool. It’s finite difference time domain as opposed to finite element method, but it works in a kinda similar way but also provides some temporal information.  This is very useful if you are working with pulsed systems.  This package is called meep (http://ab-initio.mit.edu/wiki/index.php/Meep).

There is no pretty user interface compared to the FEM solutions but a quite nice language for configuration files. It’s got some nice features as well as having the time information it allow the user to use materials which are not only birefringent but have off axis permittivities and permeabilities, so allowing for modelling of things such as Kerr rotation.  I’m not aware of a commercially available solution that does this, and certainly not one that does it so easily.

It runs pretty quickly, it can run over command line on a linux server, you can set it up to run nicely on multiple cores (allowing for great use of multiprocessor systems). It’s just a simple modelling tool.  The temporal information also gives some nice insight into what goes in things like scattering.  There’s a good website with some results of this (although not using Meep, there’s no reason these couldn’t be produced in meep and I’ve done some of them)…

http://www.met.reading.ac.uk/clouds/maxwell

Next time I’ll probably start talking about some data analysis tools…