[Pw_forum] IFCs and supercell

[Lorenzo Paulatto]

On 11/07/2014 05:18 PM, stefano de gironcoli wrote:

>      One could/should actually compute what is the safe range needed 
> include all relevant points... this was not done but we just included 
> a "VERY VERY SAFE RANGE!" of +/-2*N that proved safe for all system 
> examined so far.

I'll add one small comment just for discussion purposes.

If the atoms are put sufficiently far away from the origin of the 
Cartesian axes, i.e. with crystal coordinates much larger than one, no 
value is large enough to be safe.

Hence it would be better to take the atoms back inside the first unit 
cell (which is a trivial operation in crystal coordinates) before doing 
this operation.

However, I have yet to see somebody do that and it adds annoying phase 
factors in a code that's already complicated enough.

cheers


-- 
Dr. Lorenzo Paulatto
IdR @ IMPMC -- CNRS & Université Paris 6
+33 (0)1 44 275 084 / skype: paulatz
http://www.impmc.upmc.fr/~paulatto/
23-24/4é16 Boîte courrier 115, 4 place Jussieu 75252 Paris Cédex 05