Re: Alpha 500 vs Bi Ppro.
Keith Rohrer (kwrohrer@uiuc.edu)
Fri, 16 May 1997 02:05:27 -0500 (CDT)
> Elliot Lee wrote:
> > > you mean you actually have two pentium pro processors? I ask simply
> > > because I've never seen it written that way. That would make a big
> > > difference as well. Get two alpha chips and the alpha would probably win.
> > > Or try something besides compiling code.
> >
> > That doesn't make any sense. At some point you MUST execute compiled code,
> > whether you are executing the program itself or the interpreter... An
> > interpreter will almost always be slower, independant of platform.
>
> My favorite benchmark is "compile the kernel". As Gcc requires lots
> more CPU time to generate alpha-code, some people argue that this is
> not a valid test. When all I'm interested in is compile performance
> because I'm developing on the machine it may be a valid test after
> all, but if I'm going to run e.g. "latex" on a regular basis, then the
> results of the compile jobs matter.
If the compilation speed for that particular machine matters, kernel
compilation is a good test. If the compilation-like speed (e.g. latex
speed) is important, compile the kernel for a particular target with
a gcc which generates code for that target (i.e. build for alpha or
for intel on both platforms), to remove the problem of alpha-gcc
doing different work from i486-gcc.
Keith