VASP couldn't identify the symmetry of a large supercell correctly?

[user]

Dear All,

My input structure was a simple cubic supercell of 64 atoms. VASP was supposed to find a bcc cell as the primitive cell (2 bcc primitive cells build up the sc cell), BUT it could not.

It returned somthing like

----------------------------------

LATTYP: Found a simple cubic cell.
ALAT = 12.0000000000

Lattice vectors:

A1 = ( 12.0000000000, 0.0000000000, 0.0000000000)
A2 = ( 0.0000000000, 12.0000000000, 0.0000000000)
A3 = ( 0.0000000000, 0.0000000000, 12.0000000000)
Subroutine PRICEL returns:
Original cell was already a primitive cell.

Analysis of symmetry for initial positions (statically):

Routine SETGRP: Setting up the symmetry group for a
simple cubic supercell.

Subroutine GETGRP returns: Found 48 space group operations
(whereof 24 operations were pure point group operations)
out of a pool of 48 trial point group operations.

The static configuration has the point symmetry T_h .
The point group associated with its full space group is O_h .
---------------------------------------------

Could you please tell me what could be the problem?

Thanks.

[Veronika]

If you want vasp to find a bcc unit cell you have to set your lattice vectors as for a bcc cell and NOT for simple cubic!!!
eg:
-0.5 0.5 0.5
0.5 -0.5 0.5
0.5 0.5 -0.5

The only other possiblility is that the cubic structure is unstable and you converge towards a bcc structure.

greetings
<span class='smallblacktext'>[ Edited Tue May 03 2005, 05:21PM ]</span>

[admin]

the symmetry of the primitive cell is found by analyzing
1) the Bravais lattice
2) the ions' positions
(and facultatively, if set)
3) the local magnetic moments
4) the initial velocities of the ions.
all 2 (3, 4) sets of data have to match the symmetry criteria. Therefore,
--) please check whether all 64 ions' positions are given correctly within the
precision of SYMPREC (default = 1e-05), such that none of the ions (moments,...)
breaks the symmetry of the superlattice composed of 2 bcc cells.
--) if necessary, change SYMPREC to the required precision (this has to be done in INCAR).

[user]

Thank you very much for your replies. I have double checked and found that my structure (a quaternary compound) does not have the bcc symmetry. Sorry for the false alarm.