Aunc.3264
net.works
utzoo!decvax!harpo!duke!unc!wm
Tue Apr 13 15:29:05 1982
another view of mips
Why is everyone attacking using the term MIP to describe
computers? Why is this term any more or less meaningful
than using watts RMS (truly a meaningless term if there
ever was one) to describe a stereo system? Why one of the
very people who was attacking the use of MIPS even used
the term MHz to describe a processor chip. Most computers
have a minimum instruction time that applies mainly to
register to register operations. How many of these instructions
a certain processor can execute in a second is every bit as
meaningful as how fast it's clock runs, or even how many
Whetstones it can execute. How many people run Whetstones
in real life? Does a Whetstone bear any resemblance to the
computations that most people will run on their computers?
Things like text editors, compilers and such?
Certainly I/O bandwidth is important. But what I hear coming
over the net is that you should not even mention the term
mips unless you are a computer salesman trying to impress
someone who has more money than sense.
Many people have spent much time trying to come up with
more meaningful measures of performance, but they always
end up so application specific that they are no better to
the average buyer than before.
Yes, comparing computers is a risky business. But without
measures, however vague, we are left to compare computers
by the color of their cabinets.
-----------------------------------------------------------------
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.
