For the option of solving 1st variation eigen equation iteratively. why the default setting is：maxitefv=4，minitefv=6？What is the reasonable value of maxitefv？
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The number of iterations runs from 1 to maxitefv. The code also checks convergence of the residual against a tolerance of epsefvit. If the residual is smaller than this then the loop exits, but only if the number of iterations is greater than minitefv. This ensures that some iterations will be performed even if the residual is accidentally less than the tolerance.
The purpose of iterative diagonalisation is for very large systems for which a direct diagonalisation is too expensive. These days however, you can use MKL LAPACK with many cores to diagonalise large matrices, which makes Elk's iterative procedure less useful.
Regards,
Kay.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
For the option of solving 1st variation eigen equation iteratively. why the default setting is：maxitefv=4，minitefv=6？What is the reasonable value of maxitefv？
Hi,
The number of iterations runs from 1 to maxitefv. The code also checks convergence of the residual against a tolerance of epsefvit. If the residual is smaller than this then the loop exits, but only if the number of iterations is greater than minitefv. This ensures that some iterations will be performed even if the residual is accidentally less than the tolerance.
The purpose of iterative diagonalisation is for very large systems for which a direct diagonalisation is too expensive. These days however, you can use MKL LAPACK with many cores to diagonalise large matrices, which makes Elk's iterative procedure less useful.
Regards,
Kay.