[molpro-user] FCI excited states

Peter Knowles KnowlesPJ at Cardiff.ac.uk
Mon Nov 27 05:53:51 GMT 2006


There is indeed a hard-wired internal limit of 35 iterations. If you  
have source code, you can safely edit src/fci/fci.f at line 1318 to  
change this.
Peter

On 24 Nov 2006, at 14:09, ahauser at sbox.tugraz.at wrote:

> Dear Molpro Users,
>
> I´m trying to calculate several excited states with Molpro calling  
> the Full CI routine,
>
>
> ---------------------------INPUT------------------------
> ...
> {hf;wf,2,5,2,open,1.1,1.5;}
> {fci;state,6;maxit,50;}
> ...
> ---------------------------END--------------------------
>
>
> The number of iterations steps can be set by the MAXIT option, but  
> if I go above 35, e.g to 50,  it still stops after 35 iterations,  
> even though the FCI subroutine is telling me that the maximum  
> iteration number is correctly set to the higher value:
>
> ---------------------------OUTPUT------------------------
> ...
> 1PROGRAM * FCI (Full CI)     Author: P.J. Knowles, 1984
>
>  ***  Initialisation ***
>
>  Molecular orbitals read from record     2100.2  Type=RHF/CANONICAL  
> (state 1.5)
>
>  Frozen orbitals:              0 (  0  0  0  0  0  0  0  0)
>  Active orbitals:             92 ( 21 10 10  5 21 10 10  5)
>  Active electrons:             2
>  Spin quantum number:          1.0
>  Orbital pairs:                   712     520     520     410      
> 666     520     520     410
>  Strings:                         620     520     520     410      
> 666     520     520     410
>                                     1       0       0       0        
> 0       0       0       0
>  Determinants:                    620     520     520     410      
> 666     520     520     410
>
>  Load integrals               15.7 sec
>  Transform integrals          19.4 sec
>  Storage for integrals:        2385320
>  Remaining memory:            40597652
>
>  Core energy:               0.17596723
>  Run Hamiltonian processor DAVIDSON
>  Symmetry:                   5
>  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
>  Number of roots:             6
> 0Hamiltonian diagonaliser entered at time     62.63
> 0Initial configuration generated:
>        1      1.0000000     -0.2809057
>
>  It  Tr    CPU  Convergence   Energy       Pop     dE(next)
>
>   1   1   63.1  1.00000000    -0.28090572
>                                           0.15 -0.00182823 -0.00166338
>   2   2   63.5  0.99531357    -0.28305479    -0.05323067
>                                           0.13 -0.03932300 -0.03608153
>   3   3   63.9  0.81136687    -0.28317567    -0.13449595      
> 0.08384091
>                                           0.11 -0.04844164  
> -0.045038730
>                                           0.26 -0.00109344 -0.00094682
>  ...
>
>
>  33   3   79.6  0.07432732    -0.28329003    -0.20134778     
> -0.18656720    -0.18307272    -0.18086455
>  0
>                                           0.22 -0.00508214 -0.00450035
>  34   3   80.1  0.19842390    -0.28329786    -0.20151232     
> -0.19377751    -0.18313991    -0.18086548
>  0
>                                           0.28 -0.00121914 -0.00109017
>  35   3   80.8  0.08954343    -0.28330231    -0.20156398     
> -0.19547164    -0.18315472    -0.18087653
>  0
>                                           0.26 -0.00027774 -0.00024506
> 0*** Convergence not achieved in max iterations
>
> ...
> ---------------------------END--------------------------
>
>
> I do not know if this is either a bug or just a reasonable,  
> internally preset limit for the Davidson routine. The Molpro manual  
> gives no answer here.
>
>
> Thanks in advance,
> Andreas Hauser
>
>
>

--
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/ 
staff/knowles.html






More information about the Molpro-user mailing list