[Beowulf] Intel 64bit (emt) Fortran code and AMD Opteron
Many of your questions may have already been answered in earlier discussions or in the FAQ. The search results page will indicate current discussions as well as past list serves, articles, and papers.
Serguei.Patchkovskii at sympatico.ca Serguei.Patchkovskii at sympatico.caThu Oct 28 15:56:48 PDT 2004
- Previous message: [Beowulf] Intel 64bit (emt) Fortran code and AMD Opteron
- Next message: [Beowulf] Intel 64bit (emt) Fortran code and AMD Opteron
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
On 28 Oct 2004 at 12:06, Roland Krause wrote: > does anybody here have positive or negative experiences with using the > Intel EMT-64 Fortran compiler on AMD Opteron systems? It works fine, as long as you do not use Prescott new instructions (which are in any event of less importance on Opterons - they are not as decode-crippled as P4s are), I've built a few moderately complex quantum chemistry codes with EM64 ifort, and they run OK on Opterons. > I am at this point not so much interested in speed issues but more > stability and correctness especially with respect to OpenMP. Or in > other words: Is it worth trying yet? I can't comment on OpenMP, but for serial code it is a nice, fast and reasonably stable compiler. Not as stable or as fast as Pathscale's, but far superior to PGI's. As usual, YMMV - and probably will. Serguei
- Previous message: [Beowulf] Intel 64bit (emt) Fortran code and AMD Opteron
- Next message: [Beowulf] Intel 64bit (emt) Fortran code and AMD Opteron
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
