This is the mail archive of the
mailing list for the GCC project.
Re: Benchmarking theory
David Konerding wrote:
> I certainly
> wouldn't trust the hardware vendors to do real statistics either--
This is getting off-topic fast, so just one reply from me on this:
I do not trust the vendor on their numbers - they have to repeat them
on-site for their system to be accepted (plus they have to tell me how
they arrived at said numbers, i.e. compiler options, etc).
Toon Moene - mailto:email@example.com - phoneto: +31 346 214290
Saturnushof 14, 3738 XG Maartensdijk, The Netherlands
Maintainer, GNU Fortran 77: http://gcc.gnu.org/onlinedocs/g77_news.html
Join GNU Fortran 95: http://g95.sourceforge.net/ (under construction)