generalized eigenvalues

[Patrick Alken]

Hello,

  I am solving generalized eigensystems:

  A x = s B x

using the qz(A, B) routine. It is working perfectly
for a 625 x 625 matrix system (and smaller). However when I
try to use it on a 676 x 676 system, it gives screwy eigenvalues
that I know are false. The matrix system comes from a finite
differenced differential equation, so increasing the matrix
sizes should give better and better approximations to the 
eigenvalue, but at a certain point, it stops converging to
the correct value and jumps to some strange eigenvalue which
is certainly wrong.

Is this a known problem with octave? What is the largest
matrix system it can safely handle?

Patrick Alken



-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.

Octave's home on the web:  http://www.octave.org
How to fund new projects:  http://www.octave.org/funding.html
Subscription information:  http://www.octave.org/archive.html
-------------------------------------------------------------