[Pw_forum] space group bug?

[Tommaso Francese]

Dear all QE users,
i tried to simulate a crystal structure of silicon with symmetry space group Fd-3m, where the input was:

&CONTROL
                 calculation = 'scf' ,
                      outdir = '/home/***/espresso-5.1/tmp/' ,
                  pseudo_dir = '/home/***/upf_files/' ,
                      prefix = 'si' ,
 /
 &SYSTEM
                       ibrav = 2,
                           A = 5.43 ,
                           B = 5.43 ,
                           C = 5.43 ,
                       cosAB = 0 ,
                       cosAC = 0 ,
                       cosBC = 0 ,
                         nat = 8,
                        ntyp = 1,
                     ecutwfc = 60 ,
 /
 &ELECTRONS
 /
ATOMIC_SPECIES
   Si   28.08550  Si.pz-vbc.UPF 
ATOMIC_POSITIONS crystal 
   Si      0.000000000    0.000000000    0.000000000    
   Si     -0.000000000    0.500000000    0.500000000    
   Si      0.500000000    0.500000000    0.000000000    
   Si      0.500000000   -0.000000000    0.500000000    
   Si      0.750000000    0.250000000    0.750000000    
   Si      0.250000000    0.250000000    0.250000000    
   Si      0.250000000    0.750000000    0.750000000    
   Si      0.750000000    0.750000000    0.250000000    
K_POINTS automatic 
  1 1 1   1 1 1 

i don’t know why, but when i try to visualize both the Input file and Output file with XcrySden or J-ICE, the structure which shown is completely different from the structure introduced, resulting with a=b=c= 3.84 Angstroms instead of being 5.43, and the angles alpha=beta=gamma= 60°, instead of being 90°. 

Does anyone know why it happens? Is it an internal bug?  

I attach also the output, saw you can see directly what happens.

Thanks in andvance,
Tommaso Francese,
Università Cà Foscari di Venezia.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: silicon.pw.out
Type: application/octet-stream
Size: 11379 bytes
Desc: not available
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20140828/0752e8da/attachment.obj>