n Friday 09 February 2001 11:46 pm, Chris Gilbert wrote:
> On Friday 09 February 2001 11:39 pm, Jason R Thorpe wrote:
> > On Fri, Feb 09, 2001 at 11:12:48PM +0000, Chris Gilbert wrote:
> >  > DOH!  Sorry those figures are wrong.  I had a kernel with UVMHIST
> >  > enabled (from when playing the kernel vm stuff)  The real figures are:
> >  > /usr/bin/time ./a.out
> >  >         9.17 real         0.00 user         2.06 sys

Doh, I got those figures with a kernel with profiling (although without 
profiling it's not much different, however testing with normal memory shows:
17.86, 0.04, 4.42, which means that the kinetic ram is knocking around 50% 
off the times, note that the real difference might be bigger because I'm 
running with most of the normal memory in use (but there kernel is in it)

> Oh updated make configure time is, which is half that it is when UVMHIST is
> enabled (doh):
> 	106.59 real        20.35 user        62.93 sys
> I'm going to try it with the new bootloader in a mo.

Again that's with a profiling kernel, without:

97.18, 19.19, 55.75
however once more with normal memory its:
161.16,25.91, 103.73

Which really does show the kinetic ram is faster  (makes me feel that buying 
it wasn't a complete waste :)

Has anyone got one of the overclocked SA's in an RPC?  I believe that someone 
was selling them overclocked claiming that they were as fast as the kinetic.

Cheers,
Chris