Version Information

[graeme]

Ver 0.09c: Fixed a bug in the dipole calculation of cube files. The grid in the cube files are shifted by 1/2 a grid unit as compared to the vasp CHGCAR files.

[andri]

All executables on the Bader web site have now been updated with the current (v09c) version of the code.

[graeme]

The code and binary files have been significantly updated (to v0.20 10/8/06). The major differences are:

- The default algorithm has been changed to fix a bias in the original algorithm. The current algorithm is a little slower, but more accurate. The Bader volumes have smoother shapes now that they are not biased to be along the lattice vectors. There is no longer a systematic error in the charge of atoms in molecules for different orientations in the cell.

- Non-orthogonal cells are properly treated.

- The calculation is controlled with flags. A typical calculation can be run with the command 'bader CHGCAR'. The original algorithm can be run using 'bader -b ongrid CHGCAR'. To print the atomic volumes, use 'bader -p all_atom CHGCAR'. Options are listed by running 'bader -h'.

[graeme]

Wenjie fixed a bug in a part of the code which affects non-orthogonal cells. The code and binaries on the webpage have been updated. If you have downloaded these in the past few days and are analyzing non-orthogonal cells, you should update to the current version.

[graeme]

A new version (v0.21 11/08/06) has been released.

- Another bug has been fixed for non-orthogonal cells. The code now uses the (proper) Wigner-Seitz definition of the unit cell when calculating minimum distances between periodic images. Any results using non-orthogonal cells should be checked with this update.

[graeme]

v0.22 03/07 has been released.

Added the -ref option to allow for using a total charge density file for the Bader partitioning and a CHGCAR file to get the atomic charges. This option was added to work with the total charge density files that can be generated in the latest version of VASP.

[graeme]

v0.22a 03/07

Minimum distance reporting is fixed, as well as a bug for non-cubic cube files.

[graeme]

v0.22b 03/07

Fixed distances within cube files with a non-zero origin (again).
Added an OS X, Intel binary

[graeme]

v0.22c 03/07

Bader volumes (BvAtxxxx), when written in the CHGCAR format, now have single spaces between the values. This is consistent with how VASP prints these files.

Unified 64 bit linux binaries which are efficient on both AMD Opteron and Intel Woodcrest chips.

[graeme]

v0.22d 04/03

Added makefile, changed getarg functions, and fixed an unallocated array set to zero problem so that it builds and runs with gfortran.

[graeme]

v0.23 09/18/07

Johannes Voss added a cube file reference charge reading function.

[graeme]

v0.24 05/30/08

Wenjie added the printing of selected Bader and atomic volumes and fixed an error in the Voronoi analysis.

[graeme]

v0.25 06/09/08

A bug found by Carsten Rostgaard was fixed: the cube file data is now written in the correct rank-order.

Two output flags were added:
-p bader_index
-p atom_index
which writes the bader volume index or atomic volume index to a charge density file. Also, multiple -p output flags will be read.

The -o flag was never enabled properly, and has been removed. All charge density output files are written in the same format as the input file.

[graeme]

v0.25a 06/12/08

Unrolled some loops to fix a compiler bug with ifort/x86_32. Thanks to Carsten Rostgaard with help debugging this.

Some small changes to the printed output messages.

[graeme]

v0.25b 08/07/08

Wenjie added printing the volume of the atomic volumes.