Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!umich!yale!think!samsung!uunet!brunix!phg
From: phg@cs.brown.edu (Peter H. Golde)
Newsgroups: comp.arch
Subject: Re: Black magic, IBM RIOS.
Message-ID: <36107@brunix.UUCP>
Date: 12 Apr 90 00:17:14 GMT
References: <PCG.90Apr3204004@odin.cs.aber.ac.uk> <1990Apr4.140713.8996@specialix.co.uk> <PCG.90Apr9120504@odin.cs.aber.ac.uk> <18208@rpp386.cactus.org>
Sender: news@brunix.UUCP
Reply-To: phg@cs.brown.edu (Peter H. Golde)
Organization: Brown University Department of Computer Science
Lines: 16

In article <18208@rpp386.cactus.org> jfh@rpp386.cactus.org (John F. Haugh II) writes:
>I tested a 20MHz S/6000 with a backlevel compiler and managed to [ only ]
>produce 2.1 seconds for the register loop.  I don't know if it was 7
>instructions, but assuming it was yields
>
>	4096 x 4096 x 7 / 2.1 = 55,924,053 instructions per second
>
>which translates to 2.8 instructions per cycle.

I tested my 386 machine on the register loop and it took 0.0 seconds
(it doesn't even HAVE an inner loop).  Does this mean it's
faster than an S/6000? (and to think I never knew....)

I hereby name this benchmark "Dhumbstone".  :-) :-) :-)

--Peter Golde (phg@cs.brown.edu)