Aucbarpa.1208
net.works
utzoo!decvax!ucbvax!ARPAVAX:hickman
Thu Apr 22 18:30:27 1982
zeus benchmarks
At the January usenix, the zilog people who were showing off their Zeus
machine, were giving out free "Benchmark" listings which contained the
timing information for their machine.
Well, it turns out that their machine Was faster than ours (a 4Mhz z8000)...
But...
Just to show how valid benchmarks were, I and an associate from Microsoft
wrote a "Benchmark" that would specificly make our system look good.
Just a long way of saying, Let the buyer beware.
I have a somewhat strange way of defining the "Speed" of a system.
"The snappy-ness" factor. Basically, it is a gut feeling for how responsive
a system is under various activities. How long does it take to log in?
Is running vi on a big file slow? Things of this nature are very subjective,
but it turns out it is easy to compare via this method. One can very
quickly gauge the responsive-ness (?) of a system via the snappy-ness index.
Of course, running a program that does something is always useful. But one
MUST understand that any benchmark, at best (!), should be considered as
~v
~v
(sorry about that)
anyway, take such measurements with a ton of salt...
kipp
-----------------------------------------------------------------
gopher://quux.org/ conversion by John Goerzen <
[email protected]>
of
http://communication.ucsd.edu/A-News/
This Usenet Oldnews Archive
article may be copied and distributed freely, provided:
1. There is no money collected for the text(s) of the articles.
2. The following notice remains appended to each copy:
The Usenet Oldnews Archive: Compilation Copyright (C) 1981, 1996
Bruce Jones, Henry Spencer, David Wiseman.
