a simple question on vfin.pl script

[hlzya]

Hi, Prof. Greame

One of my dimer calculations is stopped because of power cut. Then I have used the vfin.pl script to generate a new run. I found that the MODECAR was updated while the generated POSCAR was different from both POSCAR and CONTCAR as the previous run was shut down. Can you tell me why?

Thanks

Yi-An

[graeme]

Yes, I think this is expected.

The starting position for the continued calculation (POSCAR) will not be the same as the initial POSCAR since the dimer has moved. You would expect it to be the same as the CONTCAR, but it is not. The reason for this is that the CONTCAR always contains the latest configuration being calculated. Because of the finite-difference scheme, this can be displaced from the center of the dimer by dR. The correct position to restart the dimer from is the center of the dimer, which is always stored in the CENTCAR. This should be the geometry of the POSCAR is the restart run.

[hlzya]

Thank you very much.