[Wien] problems in lapw3

[B. Yanchitsky]

Dear wien users,

There are some problems in lapw3.
Firstly, in rotdef.f the array pos is declared as
REAL(8),INTENT(IN)  :: pos(3,ndif)

but in lapw3.f allocated as
allocate ( POS(nato*48,3))

in rotdef.f this array is referenced as
pos(1:3,index1)

but in atpar.f as
POS(INDEX,j),J=1,3

Second problem is that for some cubic structures atomic X-ray factors are
different in magnitude for symmetry equvalent atoms. Below is simple example.
The structure is L12, Cu3Au,
Cu atoms (all equivalent) at (1/2,1/2,0) (0,1/2,1/2) (1/2,0,1/2)
Au atom at (0,0,0)
In *output3 we can see the following

Equiv.atom            1  at   0.500000000000000       0.500000000000000
   0.000000000000000E+000

   0  0  0  0.0000  27.7707  27.7707  0.00000  0.00000  0.00000  0.00000  0.00000
  -1  0  0  0.1320 -25.7150 -25.7203  0.00462  0.00004  0.00057  0.00000  0.00000
                    ^^^^^^^^
-----
  Equiv.atom            2  at   0.000000000000000E+000  0.500000000000000
   0.500000000000000

   0  0  0  0.0000  27.7707  27.7707  0.00000  0.00000  0.00000  0.00000  0.00000
  -1  0  0  0.1320  25.7294  25.7203  0.00925 -0.00012  0.00000  0.00000  0.00000
                    ^^^^^^^^
i.e. the terms with lm=00 have equival values, but contributions from lm!=00 are different.
I'm not sure, but might be this comes from the point
that k-vector (-1,0,0) is transformed in local frame of the first atom into (1,0,0)
but in local frame of the second atom into (0,0,1). As a result lists of spherical harmonics
(array YL) for both atoms are different, but from another point of view this should not do matter.

Regards,

Bogdan Yanchitsky
Institute of Magnetism
Vernadsky Blvd., 36-b
03142  Kiev
UKRAINE

Tel. (+380-44) 444 34 20
Fax. (+380-44) 444 10 20