[Pw_forum] pair correlation function and the choice of fnosep

[Paolo Giannozzi]

Bertrand SITAMTZE wrote:

> when sampling the canonical ensemble in cp.x we should choose the
> parameter "fnosep" [...] "should be chosen to be comparable with the
> center of the vibrational spectrum of the system" [...]. Does it mean
> that we have first to compute the phonon spectrum of the system? 

no, it means that you need an idea of what the typical vibrational
frequencies of the system are. It is a much simpler task that
calculating the entire phonon spectrum

P.
-- 
Paolo Giannozzi, Democritos and University of Udine, Italy