Hello,
Would someone have an explanation to the fact that the LAPACK CGEMV routine is faster when using and compiling the Fortran code (downloaded from netlib), rather than using the SGI vendor version in complib.sgimath ?
Not a lot faster, of course (approx 10%), but at least not slower. The size of the matrixes is 20x20.
(same observation with CPOTRF/CPOTRS)
Best regards,
Pierre
--
-----------------------------------------------------------------------
Pierre Hugonnet R&D Data Processing
COMPAGNIE GENERALE DE GEOPHYSIQUE - Paris Processing Centre
1, rue Leon Migaux / 91341 MASSY cedex / FRANCE
phone:(33) 164 47 45 59 fax:(33) 164 47 32 49
email:phugonnet(at)cgg.com (replace (at) by @)
-----------------------------------------------------------------------
The opinions expressed in this message are not necessarily
those of CGG or of Jacques Chirac
-----------------------------------------------------------------------
|