Hello all,

I've run into some convergence problems with a 216 atoms supercell calculation, and a fellow user suggested modifying the Diag.Memory value from its default of 1 to a higher one. It has helped, however now I'm perplexed as to the proper use of this parameter, for which the documentation seemed minimal.

Does anyone know of a rule of thumb for properly setting this Diag.Memory parameter, according perhaps to system size or number of processors used?

Thanks in advance,

Kevin

Reply via email to