Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uwm.edu!zaphod.mps.ohio-state.edu!think!snorkelwacker!spdcc!ima!haddock!news
From: news@haddock.ima.isc.com (overhead)
Newsgroups: comp.sys.mac
Subject: Re: MIP rating for the Mac II
Keywords: performance
Message-ID: <15663@haddock.ima.isc.com>
Date: 16 Jan 90 03:27:33 GMT
References: <2788@mtuno.ATT.COM>
Reply-To: suitti@anchovy.UUCP (Stephen Uitti)
Organization: Interactive Systems Co
Lines: 56

In article <2788@mtuno.ATT.COM> rjk@mtuno.ATT.COM (Roberto Kohler) writes:
>Thanks to all those that had something useful to say.
>The question was:
>>>Does anyone out there know how the Mac II, Max IIx, and Mac II CX compare
>>>in performance with the 80386 based machines?  I'm trying to find out the 
>>>MIP rating of the Macs and how they compare with 386 based machines.
>>>Would the Mac IIx be equivalent to a 20 MHz 386 machine, etc.
>
>Some Benchmarks:
>sieve (100 iterations) - 5 secs (386 20 Mhz) 4 secs (mac II and IIx)
>qsort (byte benchmark) - 18 secs (386 20 Mhz) 9 secs (mac II) 6 secs (mac iix)
>using Turbo C 2.0 for the 386 and MPW C for the Mac (using -mc68020 option)

Three representative 386 systems & a Mac II:
numbers are relative to a VAX 780 (whose speed is set at "1.00").

Machine	Sieve	Matmul
386/33	7.19	2.09
Mac II	3.71	0.51
386/20	1.31	0.91
386/20	0.83	0.008

The fourth 386/20 has poor RAM cache & no FPU (80387).  'Matmul'
is floating point.  My guess is that the 386/20's cost about what
the Mac II costs, with similar resources (disk, etc.).  The 386's
are running UNIX, no load.  The point is that 386's at the same
speed are not always comparable.  If you don't have an FPU, and
you are doing floating point (matmul) performance will be
minimal.  Incidently, the Mac II (16.6 MHz 68020) compared well
(pretty even) with a Sun III (20 MHz 68020).  Lightspeed C
produces fast code compared to the Sun's native compiler.

>I don't believe that any of the 3 machines you mentioned are in the same
>ballpark as the 386 machines (especially @ 25MHz).  They are all 16MHz
>machines, for one thing.

It depends on what you want:  If you have $20K and want the
fastest machine, you probably go with a high end machine.  If you
have $3K - $5K, it probably makes less difference.  A 20 MHz 386
is not necessarily faster than a Mac II (16 Mhz 68K).  It depends
on caching, etc.  System performance may be of interest.

>A straight MIPS comparison between Mac II machines and 80386 machines
>is somewhat problematic-- the Mac has practically zero overhead from
>its windowing environment where DOS & OS/2 machines lose a lot of
>performance to their windowing interfaces.

If windowing is an issue, then software quality should be too.
I don't have any benchmarks for software quality.

The bottom line is that the question needs to say "fast for what".
If you are doing low-end word processing, maybe a low end PC clone
will get you there for under $1K.  Go for it.

Stephen.
suitti@haddock.ima.isc.com