<br> Dear espresso community,<br><br> I have a question. I relaxed a surface (periodic in x and y direction) with criteria for forces 4.0d-4. I wanted to calculate a polarization via Berry phase in x-direction, however the code (berry phase) works in parallel only for z-direction ( I need to use a parallel version due to large number of atoms). So, I had to exchange the axes (z-> x , x->z) for both atomic coordinates and for cell parameters:
<br><br>before:<br>CELL_PARAMETERS cubic<br> 5.65402582 0.00000000 0.00000000<br> 0.00000000 5.65402582 0.00000000<br> 0.00000000 0.00000000 30.74000000<br><br>and after:<br>CELL_PARAMETERS cubic<br> 30.74000000
0.00000000 0.00000000<br> 0.00000000 5.65402582 0.00000000<br> 0.00000000 0.00000000 5.65402582<br><br>The code found 4 symmetry operations as before. The strange thing - after those operations I performed scf calculations with the same parameters and output forces are huge ~
0.02 Ry/au. So, it looks like a system is not relaxed. Did I do something wrong?<br><br>cheers,<br> jess<br><br><br>-- <br>===================================== <br>Jess Kondor<br><br>PICYT, Av. Venustiano Carranza 2425-A,
<br>San Luis Potosí 78210, México<br>=====================================