[Wien] Athlon XP optimized BLAS and Lapack libraries available

[Kirk Feindel]

I used the pgf90 and pgcc compilers with acml on an AMD Duron platform.  I
receive an error when running the TiC example, at lapw1.  See a previous
posting.


----- Original Message -----
From: "Jeff Spirko" <spirko at lehigh.edu>
To: <wien at zeus.theochem.tuwien.ac.at>
Sent: Monday, July 21, 2003 10:04 AM
Subject: [Wien] Athlon XP optimized BLAS and Lapack libraries available


> AMD has released a beta version of the AMD Core Mathematical
> Libraries, which includes BLAS, LAPACK, and FFT routines optimized
> for Athlon XP processors.  They are available from www.amd.com under
> a fairly restrictive agreement, so I have not tested them yet.
> There is a Linux version.  It is said to be compatible with the GNU
> g77 fortran compiler or the PGF90 Fortran compiler.
>
> With the number of WIEN users using AMD processors, it would be
> interesting to find out if:
>  - does ACML work with WIEN2k?
>  - Is it faster than using Atlas BLAS and compiled LAPACK?
>  - Can it be used with the Intel Fortran compiler?
>
> Of course, with this clause in the license agreement:
>   5. Confidentiality.
>   a. AMD and USER agree that the Licensed Materials constitute beta
>      versions of products developed by or for AMD. USER understands
>      that the testing and quality assurance of the Licensed
>      Materials has not been completed, nor are the Licensed
>      Materials authorized for general release by AMD. Therefore,
>      USER agrees to treat the Licensed Materials, any feedback and
>      the existence of this Agreement as confidential to AMD. USER
>      will not divulge the information contained therein to any third
>      party, and USER also agrees not to disclose to any party that
>      it has received or is testing or has tested the Licensed
>      Materials except at the written request of or with the written
>      approval of AMD. USER further agrees not to disclose to any
>      third party any test results, measurements, metrics, data or
>      other information obtained by use or possession of this
>      Software.
> It might be a breach of contract to discuss the answers to the
> important questions above.
>
> --
> Jeff Spirko   spirko at lehigh.edu   spirko at yahoo.com   WD3V   |=>
>
> The study of non-linear physics is like the study of non-elephant biology.
>
> All theoretical chemistry is really physics;
> and all theoretical chemists know it. -- Richard P. Feynman
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien