Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!lll-crg!nike!oliveb!glacier!Shasta!simoni
From: simoni@Shasta.STANFORD.EDU (Richard Simoni)
Newsgroups: net.micro.mac
Subject: Re: Apple SCSI HD20
Message-ID: <847@Shasta.STANFORD.EDU>
Date: Wed, 17-Sep-86 00:17:15 EDT
Article-I.D.: Shasta.847
Posted: Wed Sep 17 00:17:15 1986
Date-Received: Fri, 19-Sep-86 22:09:57 EDT
References: <491@randvax.UUCP> <154@apple.UUCP> <1791@well.UUCP>
Reply-To: simoni@Shasta.UUCP (Richard Simoni)
Distribution: net
Organization: Stanford University
Lines: 30

In article <1791@well.UUCP> das@well.UUCP (David Shayer) writes:
>

>[apple introduces the IIGS]
>
>The Apple IIGS runs at 3 MEGAHERTZ!  Thats slower than an IBM PC.
>My Bicycle can go faster than that.  It comes with 256K, expandable
>to 8 meg.  How can Apple sell a computer with one third the speed and
>one quarter the memory of an Atari 1040ST for 150% of the price?
>Is there some vital point I have missed?

Well, there is one point which should be made, although it may or may not
be relevant to the IIGS vs. 1040ST vs. IBM PC debate.  A general statement
can be made about computer systems: clock speed <> system performance.
Why?  Because the number of clock cycles per instruction varies widely
across processors.  For example (this is off the top of my head and may
not be exactly right), most 6502 instructions take 2-6 cycles to complete,
while many Z80 instructions take 15-20 cycles to complete.  Although I'm
not up on my 16- and 32-bit processors, the point is that comparing clock
frequencies on machines with different processors is meaningless.  (In
addition, the number of instructions required to perform a particular
task can also vary greatly across processors.)

Just thought I'd try to clear up this common misconception.

Rich Simoni
Center for Integrated Systems
Stanford University
simoni@sonoma.stanford.edu
...!decwrl!glacier!shasta!simoni