another qmd problem


Click here for full thread
Forum Regular
In principle, this would work since the qmd module is essentially just repeated calls to task <theory> gradient and then updates of atomic positions based on those gradients. However, the qmd module was never tested for use with the plane wave part of NWChem, so I can make no guarantees concerning accuracy or stability. A small test I performed finished without crashing and the output looked reasonable, but I would proceed with caution and carefully check your results.