Page 1 of 1

Suggestion on Mrgscr and Anaddb

Posted: Fri Oct 15, 2010 6:02 pm
by Robin
Dear developers,

While using mrgscr, I find it does not accept any path before the file names. For example, one can not specify “../test_SUS” but “test_SUS”. In mrgddb, there is no such problem.

For anaddb, I find the definition of rprim in the output file of eigenvector (LWF file) is inconsistent with that in the main output file. In the output by abinit, each row of rprim represents a cell vector, while in the LWF file, each column a cell vector. You can check the following example of Graphene.

Sincerely,
Guangfu Luo


#-----------------------------Output by abinit-----------------------------
rprim 1.0000000000E+00 0.0000000000E+00 0.0000000000E+00
-5.0000000000E-01 8.6602540378E-01 0.0000000000E+00
0.0000000000E+00 0.0000000000E+00 1.0000000000E+00

#-----------------------------LWF by anaddb------------------------------
rprim=
1.00000000 -0.50000000 0.00000000
0.00000000 0.86602540 0.00000000
0.00000000 0.00000000 1.00000000

Re: Suggestion on Mrgscr and Anaddb

Posted: Sun Oct 17, 2010 12:16 pm
by mverstra
Hello,

thanks for the debugging effort. The first point has been addressed by Matteo Giantomassi in another post. I am fixing the second one in 6.4.2. This is just an output error, nothing fundamental as far as I can tell.

cheers

Matthieu

Re: Suggestion on Mrgscr and Anaddb

Posted: Sun Oct 17, 2010 1:38 pm
by mverstra
Hi again,

ok, this is fixed in 6.4.2, along with a _lot_ of other outputs of rprim, which were erroneously transposed, but only present in debugging output.

cheers

matthieu