A question about the aimd calcuation by VASP

[LZYUCL]

When Usually, when we use the AIMD process of the vasp calculation system, a key parameter SMASS will be involved. I will first set this parameter to 0 in the general calculation process, and then adjust this parameter according to the SMASS output in the vasprun file. But the SMASS in INCAR_md I saw in the dpgen example is set to 0. I am not sure whether this value will be changed according to the output of vasp in the dpgen cycle. Will it affect the accuracy of the result if it is always set to 0?

[PolyuWeldingSpock]

In my opinion, examples of DPGEN just provided some reference not accurate parameter especially in fp steps. You should set your own parameter according to your own experience .