Re: speed of deschallmac on PMac 7300/200

Arvin Meyer (onsite@esinet.net)
Sun, 25 May 1997 14:07:57 -0400


Similar circumstances on an Intel 200 P5 under Win95 last night yielded
1,061,009 fastest and 1,010,105 slowest. More of them ran closer to the
lower mark. Using NT4-Server on the same machine the night before, I
averaged around the 1000k mark but it was running a few services in the
backround.

I see no major advantage to running at higher priority when the system is
basically idle. Lots of similar Intels are reporting 50 to 100k less
performance than mine, a few a little better, so I think that the
motherboard may have some effect as well as the processor chip. When my
NT-Server is heavily working, I have seen as little as 786k/sec being
processed.

So it would appear that machines and operating systems are relatively
equal, Chip MHz to Chip MHz and O/S to O/S. I have been checking out DES
web page statistics at sites all over the country, and this appears to be
the case. The highest statistics seem to have come from 500 MHz DEC Alphas,
but I haven't seen any statistics from similar Sun and Apple systems (Does
Apple even make a 500 MHz PowerPC?) If nothing else gets accomplished,
this knowledge is worth its weight in gold to MIS depts, deciding on major
equipment purchases.

CRACK DES NOW! http://www.frii.com/~rcv/deschall.htm

Arvin Meyer

On-Site Solutions

"Developing results-orientated databases for companies
that demand a tangible return on investment."

e-mail: onsite@esinet.net
phone: (804) 973-9140
http://www.esinet.net/cabg/consult/onsite.html
--------------------------------------------------------------------

----------
> From: Duane T Williams <duane@cmu.edu>
> To: deschall@gatekeeper.megasoft.com
> Subject: speed of deschallmac on PMac 7300/200
> Date: Sunday, May 25, 1997 12:26 PM
> As a point for comparison, the May 14 version of DESChallMac running on
my
> Power Macintosh 7300/200 typically reports 1048k keys/sec when running in
> the foreground with no other activity on the machine. (Note: k in the
> reports = 1000, not 1024).
>