This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: An unusual Performance approach using Synthetic registers
- From: Lars Segerlund <lars dot segerlund at comsys dot se>
- To: Robert Dewar <dewar at gnat dot com>
- Cc: gcc at gcc dot gnu dot org
- Date: Wed, 08 Jan 2003 12:32:09 +0100
- Subject: Re: An unusual Performance approach using Synthetic registers
- References: <20030108112036.43EC1F2D89@nile.gnat.com>
Now this stupid $500 motherboard/computer managed in excess of 90 % of
the performance of a dual xeon system at the time, if equipped with the
same ram and scsi drives :-) .. since a xeon was about $2000 at the time
this 'stupid' system was serious bang for the buck as long as you didn't
need floating point, ( compile farms and chip synthesis was what we used
it for, and simple math gives you 4 machines for the price of a cpu,
which is real value for money since it gives you 4 times the memory
bandwidth which was the limiting factor at the time ).
They were very cheap and fast machines.
i.m.h.o. / Lars Segerlund.
Robert Dewar wrote:
dumb statistic, fwiw: dual processor 500Mhz Celerons.
My Mandrake 8.2 distro calls it a pentiumpro.
I am surprised anyone would ever have built a dual processor with such
a strange processor choice. By the way the reason 8.2 calls it a
pentiumpro is probably just because it is so slow that it looks like
it is a pentiumpro :)