UT theoretical chemistry code forum

Dear all
I performed some NEB calculations using VASP code.
But, in some cases, I found the first image and last image( not
endpoints) gave the incorrect energies comparing with the energies
of endpoints. Then, I use the POSCAR in 01 directory to perform a energy
calculation, but the correct energy is obtained. Moreover, the
other energies look like reasonable. What's the problem I meet ?

Thanks a lot in advance !

This sounds a little puzzling. Our scripts simply grep for the energy in the OUTCAR file. Can you send a .tar.gz file of the run, and say which energy you think is incorrect?

Can you also make sure that your single calculation has the same parameters as the neb calculation. The parameters in the INCAR file should be the same, and it would be best to use the same number of processors per image.