[Wien] Latest WIEN2k compiled on different machines - different :ENE for TiC

[Peter Blaha]

A couple of years ago I had very similar problems with the pgi compiler
(on an Opteron system).
Since then I do not use pgf90 anymore.
On the other hand, ifort 8.1 is also known to be very buggy ....

I suggest to use a recent ifort version.

TiC.tar was probably created a few years ? ago.

Jorissen Kevin schrieb:
> Dear WIEN users,
>  
> I've installed the latest WIEN2k on two different machines :
> * intel xeon + ifort 8.1 + mkl 7.0
> * opteron + pgf90 5.2-4 + acml 2.0
>  
> To the best of my knowledge, compiler settings are as equivalent as possible.
> For the TiC test case, I get total energies :ENE that differ by about 10^-4 Ry from one machine to the other.  (I compared the ifort+mkl to the precompiled executables on the xeon machine and found as good as identical results there.)
>  
> Has anyone compared these energies between different platforms and found agreement or disagreement?
>  
> And does anyone know what WIEN-version was used for the files in the TiC.tar testcase?
>  
> Thanks!
>  
>  
> Kevin Jorissen
> University of Washington
> Dept. of Physics
> Box 351560
> Seattle, WA-98195-1560
> U.S.A.
>  
> phone +1 206 543 3904 
> fax +1 206 685 0635 
> e-mail kevin.jorissen at ua.ac.be 
> web page http://fraangelico.phys.washington.edu/~jorissen
>  
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien