Apparent bug in MP3 portion of MP4 calculation


Jump to page 1Prev 16
Forum Vet
8:26:16 PM PST - Mon, Nov 30th 2015
I have an molecule containing several atoms, using GAMESS geometry, RHF and 6-31+G**, and treated by both softwares, where GAMESS uses its default convergence limit, etc., and find the differences of both ccsd and ccsd(t) energies are around 0.018 hartrees.

Forum Vet
11:05:40 PM PST - Tue, Dec 1st 2015
Dear Jhammond
Sorry, I did not notice you used bohr as the unit, now the results from GAMESS agree well
with those from NWCHEM6.6.

But my molecule still has a bit large discrepancy, i.e., 0.018 hartrees.

Any suggestions?

Forum Vet
8:28:46 PM PST - Sat, Dec 5th 2015
Now I know the difference might be caused by the NWCHEM input is not in the desired situation.


Forum >> NWChem's corner >> Running NWChem
Jump to page 1Prev 16