----- Original Message -----
From: David Brian Serafini <[log in to unmask]>
To: <[log in to unmask]>; <[log in to unmask]>
Cc: <[log in to unmask]>; <[log in to unmask]>
Sent: Wednesday, September 29, 1999 10:01 AM
Subject: Re: ANNOUNCE: Lahey/Fujitsu Fortran95 for Linux
> I ran the Portland Group f90 compiler (v3.0) on a PIII/450 and got
> 90MFlops (+-1.5) for double precision and 117MFlops (+-2) for single
precision.
>
> The results you quote for the Fujitsu and NAG compilers (i.e. NAG much
faster
> than Fujitsu) doesn't fit with what we've seen on the NPB-serial
benchmakrs,
> where the Fujitsu compiler is about the same as Portland Group and Absoft
> and NAG and GNU g77 are significantly slower. See
http://www.nersc.gov/research/FTG/pcp/performance.html.
>
> -david
Where do the parentheses go on that second-to-the-last sentence?
(Fujitsu compiler is about the same as Portland Group) and (Absoft and NAG
and GNU g77 are significantly slower).
or
(Fujitsu compiler is about the same as Portland Group and Absoft) and (NAG
and GNU g77 are significantly slower).
or if you're British maybe even
(Fujitsu compiler is about the same as Portland Group and Absoft and NAG)
and (GNU g77 are significantly slower). ;-)
--
Bob Runyan
Lahey Computer Systems
http://www.lahey.com
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|