Hi, Sir~
I have recompiled vasp with your VTST 2.0 code. But I found the total energy has changed greatly compared to the original results.
For example , I have calculated the Pt/NH3 system. The old calculated total energy is about -101.ev. But with the new version, I obtain an energy of about 130eV. Is this phenomenon normal ?
The influence of VTST code on total energy
Moderator: moderators
The VTST code should have no effect on normal vasp runs.
The default values of ICHAIN=0 (don't use neb/dimer/lanczos/dynmat) and IOPT=0 (don't use our optimizers) turn off the vtstcode.
The only change you should see in the OUTCAR of a normal vasp run, are the printed lines:
VTST: version 2.03a (07/10/07)
CHAIN: initializing optimizer
and then the force average each iteration (for example):
FORCES: max atom, RMS 2.058488 0.404837
FORCE total and by dimension 2.560414 2.021172
If there is any difference in the calculation, something is going very wrong. Our code is designed to sit along-side vasp and provide additional methods, but not alter how vasp runs. Please check your calculation. Also, a large positive energy of 130eV doesn't make any sense.
The default values of ICHAIN=0 (don't use neb/dimer/lanczos/dynmat) and IOPT=0 (don't use our optimizers) turn off the vtstcode.
The only change you should see in the OUTCAR of a normal vasp run, are the printed lines:
VTST: version 2.03a (07/10/07)
CHAIN: initializing optimizer
and then the force average each iteration (for example):
FORCES: max atom, RMS 2.058488 0.404837
FORCE total and by dimension 2.560414 2.021172
If there is any difference in the calculation, something is going very wrong. Our code is designed to sit along-side vasp and provide additional methods, but not alter how vasp runs. Please check your calculation. Also, a large positive energy of 130eV doesn't make any sense.