 Re: ADPACK Ver. 2.2, Database 2011, and patch for OpenMX Ver. 3.5 ( No.1 ) |
- Date: 2011/09/29 18:22
- Name: Renato Miceli
- References: http://www.ichec.ie/
- Hi Mr. Ozaki,
I ran across an error while trying to compile OpenMX 3.5.3. It complains about an unknown function during the linking phase. The error is the following:
-------
XC_PBE.o: In function `XC_PBE':
/local/packages/openmx3.5/source/XC_PBE.c:93: undefined reference to `XC_PW92C'
make: *** [openmx] Error 1
-------
In the previous version of OpenMX, this call was being made to function XC_PW91C rather than XC_PW92C, which can be found in XC_PW91C.c. I wonder if there is a missing file XC_PW92.c or if this new function is actually a typo. Can you advise me on the issue, please?
Thanks in advance.
Looking forward to hearing back from you soon.
Kindest regards,
Renato Miceli
|
| Re: ADPACK Ver. 2.2, Database 2011, and patch for OpenMX Ver. 3.5 ( No.2 ) |
- Date: 2011/09/29 20:51
- Name: T.Ozaki
- Hi,
Sorry. This is my simple mistake.
I have uploaded a new patch (patch3.5.4.tar.gz) at
http://www.openmx-square.org/bugfixed/11Sep29/patch3.5.4.tar.gz
Thank you for your cooperation in advance.
Regards,
TO
|
| Re: ADPACK Ver. 2.2, Database 2011, and patch for OpenMX Ver. 3.5 ( No.3 ) |
- Date: 2011/09/29 23:10
- Name: Renato Miceli
- References: http://www.ichec.ie/
- Hi Mr. Ozaki,
the compilation and linking worked with this new patch. Thank you very much!
Kindest regards,
Renato Miceli
|
| Re: ADPACK Ver. 2.2, Database 2011, and patch for OpenMX Ver. 3.5 ( No.4 ) |
- Date: 2011/10/01 03:01
- Name: Lida
- Dear Dr. Ozaki,
I have a problem with the new version of PAO database for electronic transport calculations. In such calculations only the interaction of two adjacent unitcells should be considered, however, if you set the cutoff radius to the minimum value (i.e 7 (a.u) for Si in a nanowire structure) it will consider not only the interaction between two adjacent unitcells but also the interaction between the current unitcell with the unitcell next to the adjacent one, the following error message:
ERROR: PAOs of lead atoms can overlap only to the next nearest region.
Previously, I used lower cutoff radius in my calculations, so I did not face this problem. I was wondering if you have any advice in this regard.
Regards,
Lida
|
| Re: ADPACK Ver. 2.2, Database 2011, and patch for OpenMX Ver. 3.5 ( No.5 ) |
- Date: 2011/10/01 12:35
- Name: N.Kolchenko <nkolchenko@mail.ru>
- Dear prof. Ozaki,
1. Database 2011 -
The results of comparison with full-electron (L)APW (Wien2k) calculations are impressive. It's really fine !
2. Patch for OpenMX Ver.3.5 -
I tested new patch by "old problem". As I wrote earlier, calculations with GGA-PBE gave different results for OpenMP and hybrid MPI+OpenMP modes. For example, in system Mol_MnO.dat one could obtain (~ 0.2 eV) difference in total energy.
New patch version results for this case are identical up to many digits beyond tolerance level.
Regards,
NK
|
|
This thread is locked.Only browsing is available.
ADPACK Ver. 2.2, Database 2011, and patch for OpenMX Ver. 3.5