[QE-users] How to optimize degauss and K point - reg

[singaravelan T R]

Dear all,
I am using occupations as "smearing".
Within that we have 'gaussian' , 'mathfessal Paxton' , 'mazari Vanderbilt'.
In each case we have degauss varying starting from 0.01.
Similarly we have different K points which also has to be optimised.
Qn 1: I know that we have to optimize degauss and k point at same time and
 the Thumb rule is degauss --> small and K point --> large. My question is
how to find which value of degauss and  K point is better.?

Qn2:  we have three sorts of smearing (Gaussian, mathfessal Paxton', mazari
Vanderbilt) without including fermi-dirac. My question how to find which
smearing is better. Or it depends on type of material which we use.

With thanks,
Singaravelan T R
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20201217/9bc7ef68/attachment.html>