[molpro-user] Problem with maxit in FCI

[Peter Knowles]

There is a compiled limit of 35 iterations. If you have access to the  
source code, you can safely change this: src/fci/fci.f, subroutine  
davids, maxitr.
Peter
On 22 May 2008, at 03:46, John S. Sears wrote:

> Dear users
> I am having a problem with some FCI computations in molpro 2006.1.  
> The FCI
> calculation is not quite converged in 35 iterations. I set the maxit  
> keyword to
> 50 in the fci card and the program appears to recognize it. The  
> output even
> prints the maximum number of iterations
>
> Run Hamiltonian processor DAVIDSON
> Symmetry:                   2
> Trial vector:                  0.00
> Result vector:                 0.00
> Maximum iterations:           50
> Convergence threshold:         0.0000100
> RHS vector:                    0.00
> Hamiltonian shift:             0.0000000
> Output threshold:              0.0500000
> Options:                        0
>
>
> However, at the end of the 35th iteration the program stops
>
> 35   125044.6  0.00263340   -56.67373865
> +                                         0.06 -0.00000219 -0.00000210
> 0*** Convergence not achieved in max iterations
>
> Any advice here would be greatly appreciated.
> Thanks
> John
>
>
> --
> John S. Sears
> Postdoctoral Research Associate
> Center for Computational Molecular Science and Technology
> Center for Organic Photonics and Electronics
> School of Chemistry and Biochemistry
> Georgia Institute of Technology
> Atlanta, GA 30332-0400
>

--
Prof. Peter J. Knowles
School of Chemistry, Cardiff University, Main Building, Park Place,  
Cardiff CF10 3AT, UK
Telephone: +44 29208 79182 Fax: +44 2920874030 Email KnowlesPJ at Cardiff.ac.uk
WWW http://www.cardiff.ac.uk/chemy/contactsandpeople/academicstaff/knowles.html