The ABINIT Users Mailing List ( CLOSED )

[Matthieu Verstraete <mjv500@york.ac.uk>]

Your setup looks ok: the only communication between datasets is through 
the getwfk -1, so it doesn't actually matter that much which order you do 
them in. As it stands, it may not work (try it though - it might) when it 
switches kpt grids (going from dtset 14 to 21).

The tsmear value datasets share a common set of kpoints, so you can more 
easily transfer the wfk. To be on the safe side you might specify

getwfk?2 -1
getwfk?3 -1
getwfk?4 -1

to force the first calculation for each kpt grid.

In a nutshell, the order has very little importance as you have to 
reconverge in every case and all the calculations are independent, apart 
from the starting wfk (which can just speed up the calculation, and 
shouldn't change the result).

Matthieu

--
================================================================
Dr. Matthieu Verstraete                 mailto:mjv500@york.ac.uk
Dept. of Physics, University of York,     tel: +44 1904 43 22 08
Heslington, YO10 5DD York, United Kingdom fax: +44 1904 43 22 14