[Pw_forum] unexpected force_symmorphic behavior with explicit k grid

[Giovanni Pizzi]

Dear Paolo,
thanks for your answer.
Indeed, I am using Yambo which requires that flag.

The point is that Andrea Marini and I are working on a Wannier90 
interface for Yambo.
While for the generation of the automatic grids the increased number of 
k points is perfectly ok, the problem arises when we have to generate 
the full k grid that is needed by Wannier90.
In this case, in fact, if for instance I specify the 64 k points for a 
4x4x4 grid in the nscf calculation, and I leave the force_symmorphic 
flag set to true, QE will calculate the band energies on 136 k points 
instead.
Of course for this nscf calculation I can simply remove the 
force_symmorphic flag, I was just wondering if it could make more sense 
not to increase the number of k points if they are given as an explicit 
list (but maybe there are cases for which instead this is the intended 
behavior).

Giovanni


On 06/20/2012 06:27 PM, Paolo Giannozzi wrote:
> On Wed, 2012-06-20 at 17:46 +0200, Giovanni Pizzi wrote:
>> >  I was using the force_symmorphic flag of pw.x.
> you shouldn't, unless you have a good reason to use it
>
>> >  I noticed that even when I provide an explicit list of k points
>> >  (e.g. with
>> >  K_POINTS crystal
>> >  or even
>> >  K_POINTS crystal_b
>> >  for a band structure) pw.x increases the number of k points.
> "force_symmorphic" just removes the symmetry operations having
> an associated fractional translation from the list of allowed
> symmetries. Any change in the number of k-points, if any, is
> a consequence of the reduced symmetry of the system.
>
>> >  Is this the intended behavior, or should the list of K points be
>> >  modified only for K_POINTS automatic?
> hard to say what the "intended" behavior is: there isn't any.
> That option was added (by me) in order to allow PWscf to produce
> data for other codes (don;t remember which ones) that do not
> implement nonsymmorphic symmetry groups.
>
> P.
> -- Paolo Giannozzi, IOM-Democritos and University of Udine, Italy


-- 
Giovanni Pizzi
Post-doctoral Research Scientist
EPFL STI IMX THEOS
MXC 319 (Bâtiment MXC)
Station 12
CH-1015 Lausanne (Switzerland)
Phone: +41 21 69 31159