Path: utzoo!attcan!ncrcan!scocan!seanf
From: seanf@sco.COM (Sean Fagan)
Newsgroups: comp.sys.m68k
Subject: Re: Moto's data predicts 68040 performance well below 20 MIPS
Message-ID: <1990Jul24.123842.19663@sco.COM>
Date: 24 Jul 90 16:38:42 GMT
References: <1990Jul13.163849.4282@msuinfo.cl.msu.edu> <6535.26a0e67f@vax1.tcd.ie> <713@obs.unige.ch> <382@zds-ux.UUCP>
Reply-To: seanf@sco.COM (Sean Fagan)
Organization: The Santa Cruz Operation, Inc.
Lines: 31

In article <382@zds-ux.UUCP> gerry@zds-ux.UUCP (Gerry Gleason) writes:
>MIPS is a measure of integer CPU performance, so I wonder why people are
>being so careful to tell us about the floating point hardware.  If you want
>to talk floating point, the metric is MFLOPS, not MIPS.

MIPS :== Millions of Instructions Per Second.

I don't see the word "integer" in any form anywhere in there, nor is it
implied.

There are some machines that can do some fp stuff faster than integer
instructions; a marketing-type person might want to use those.  (However, I
know of no machine where a NOP is slower than an FP add 8-).)

MFLOPS (Millions of FLoating-point Operations Per Second) is used because
there is a large class of people who don't really *care* how many
instructions the machine can execute per second, just how many fp adds and
multiplies it can do per second.  In a rather major class of computers
(Cray, for example), this is not completely related to MIPS (well, it is,
but it can be as much as 64 times as much as MIPS).

As to why telling about the floating-point hardware:  if I remember
correctly, there were some instructions on a VAX that would work faster if
you had the FPA, since the microcode would use it for some "integer"
instructions.

-- 
Sean Eric Fagan  | "let's face it, finding yourself dead is one 
seanf@sco.COM    |   of life's more difficult moments."
uunet!sco!seanf  |   -- Mark Leeper, reviewing _Ghost_
(408) 458-1422   | Any opinions expressed are my own, not my employers'.