[Pw_forum] Projwfc.x for fatbands

[Matthieu Fortin-Deschênes]

Dear all,

I'm trying to understand the process of calculating fatbands like in  
PP/examples/example05.

Here is my problem:

When I use lsym=.TRUE. in the input of bands.x, I get the output file  
with 3 columns similar to feo_af.bands.all with the kpoints, energies  
and weights. However, some of the band crossings seem wrong.


When I use lsym=.FALSE. in the input of bands.x, the band crossing  
problem is solved (at least partially), but i dont get the weighted 3  
column output file of the bandstructure.

My question is:

Is there a physical reason why the weighted bands cannot be obtained  
when lsym=.TRUE. is used in the input of bands.x, or is it just not  
implemented?

Thank you

Matthieu Fortin-Deschênes
Polytechnique Montréal