Hello,

I work on evaluation of the Feynman path integral of the density matrix using a diagonalization approach.

I have to compute eigenvalues and eigenvectors of a symmetric matrix, for this purpose I use the subroutines tred2 and tqli.

I encounter some problems:
1.A shift in the eigenvalues which starts from a certain eigenvalue and continues.
2.Negative small eigenvalues (this maybe related to the previous problem)
3.Sometimes I get the error massage: "too many iterations in tqli".

I want to mention that:
1.I work with large matrices (400*400 or so) and small eigenvalues.
2.I work with Double Precision notation in FORTRAN 77 and ifort compiler.

Thanks,
Tali.