Author Topic: problem with "jobex -ex"  (Read 3618 times)

kankanxiaoya

  • Newbie
  • *
  • Posts: 1
  • Karma: +0/-0
problem with "jobex -ex"
« on: November 23, 2011, 08:07:47 am »
hello everyone:
     I have a problem with my excitation state optimization. I do a "jobex -ex" , it can be run for about 4 cycles ,after that it stoped.
     The warnning comes out "statpt.imput" , but opening that is :

                          this is S T A T P T


                     hessian and coordinate update for
                          stationary point search

                     by barbara unterreiner, marek sierka,
                           and reinhart ahlrichs

                          quantum chemistry group
                          universitaet  karlsruhe
                                  germany



        ***************  Stationary point options ******************
        ************************************************************
           Maximum allowed trust radius:           3.000000E-01
           Minimum allowed trust radius:           1.000000E-03
           Initial trust radius:                   1.500000E-01
           GDIIS used if gradient norm <           1.000000E-02
           Number of previous steps for GDIIS:       5
           Hessian update method:                  BFGS
                        *** Convergence criteria ***
           Threshold for energy change:            1.000000E-06
           Threshold for max displacement element: 1.000000E-03
           Threshold for max gradient element :    1.000000E-03
           Threshold for RMS of displacement:      5.000000E-04
           Threshold for RMS of gradient:          5.000000E-04
        ************************************************************

     Number of symmetry restricted degrees of freedom:   566
     Constituted by:     3 translational degrees of freedom
                         3 rotational degrees of freedom
                     560 internal degrees of freedom




.... that's over ,it cannot be the next "ATOM CARTESIAN COORDINATES".

SO  I am puzzled ,because I have run for 4 cycles, and it Suddenly Stops.  I need your help.

My structure is about 150 atoms ,is it too big or disk space low or some others?

Thank you very much

songk

« Last Edit: November 23, 2011, 08:09:22 am by kankanxiaoya »

uwe

  • Global Moderator
  • Sr. Member
  • *****
  • Posts: 448
  • Karma: +0/-0
Re: problem with "jobex -ex"
« Reply #1 on: December 09, 2011, 09:00:35 am »
Hello,

there should be an error message on screen or in the file where you directed the error output to. Try one of the following:

- remove the file hessapprox and rerun statpt directly. Sometimes GDIIS gets problems and by removing the approximate Hessians (and more) from the steps before (stored in the file hessapprox), GDIIS is being re-initialized. That often helps.

- if you run statpt from the command line as given in the line above, you should see the error message on your screen

- not very likely to happen earlier than in the fifth cycle, but still possible: check your user limits (stack size), see: http://www.turbo-forum.com/index.php?topic=23.0

Regards,

Uwe