[Wien] K-Points determination

[Stefaan Cottenier]

>    Can someone tell me how to estimate the number of k-points needed for a
calculation.  For example, if I am using a 2 x 2 x 2 supercell for Si with
64 atoms in the cell, how many k-points must I use to recover the accuracy
obtained for the simple Si cell (with a given number of k-points)?   Thanks
in advance.

Look at the top line of case.klist, at the right, to see Kx, Ky, Kz (or in
the screen output of kgen). If you double the lattice constant in one
direction, you have to go only to one half of the value of the corresponding
Ki to get the same sampling quality.

Stefaan