Eigenvalue computation in FEMLAB - problem

[eirp]

tm waves in femlab

Hi guys!

I'm computing the eigenvalues and eigenfunctions (solving Helmholz equation in elmag. module - "In plane TE/TM waves") of planar shapes in FEMLAB.
I need to compute as many eigenvalues as I can but there's problem.
It's no problem to compute 50-55 eigenvalues in a short time but when I increase this number a little (to 70-80) the computing time is gigantic...
It's due to some physical limitation of finite element method (or eigenvalue problem) or I need to improve something in FEMLAB settings???
I'd like to compute some statistic of eigenvalue differences for various shapes and study so called "trace formula" which connects eigenpairs with orbits of bouncing ball in billiard...

Thanks in advance,
Eirp

 

[deadsilence]

Hi eirp,

I once had the same problem. And yes, I think it has something to do with the FEM.

 

[eirp]

That's bad because FEMLAB's great GUI and postprocessing.. I can turn to solving the PDE using finite difference method but I don't have time to make such program.
Thank for your suggestion,
Eirp