Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!cmcl2!beta!hc!ames!amdahl!chongo
From: chongo@amdahl.amdahl.com (Landon Curt Noll)
Newsgroups: comp.arch
Subject: Re: 09/20/87 Dhrystone Results (1 of 2) [really 2.0]
Message-ID: <15055@amdahl.amdahl.com>
Date: Fri, 25-Sep-87 17:52:29 EDT
Article-I.D.: amdahl.15055
Posted: Fri Sep 25 17:52:29 1987
Date-Received: Sun, 27-Sep-87 02:47:50 EDT
References: <380@pcrat.UUCP> <14997@amdahl.amdahl.com> <707@winchester.UUCP>
Reply-To: chongo@amdahl.amdahl.com (Landon Curt Noll)
Organization: Amdahl Coup, UTS Products Hen house
Lines: 32
Keywords: benchmark, performance

In article <707@winchester.UUCP> mash@winchester.UUCP (John Mashey) writes:
 >In article <14997@amdahl.amdahl.com> chongo@amdahl.amdahl.com writes:
 >> >XAmdahl     5890/300E            0.00 -21932 -21868 UTS 1.2 (SVR2),cc  -O,
 >>
 >>The Amdahl 5890/300E performs the dhrystones before you even compile the 
 >>program!  :-)
 >
 >"I work on an X (fast computer): it can do an infinite loop in 6 seconds."
 >"Well, I work on Y (multiprocessor): it takes a little longer to do an
 >infinite loop, but it can do N of them at once."
 >
 >							... it will be
 >wonderful to stop ahving to write reams of words to give reasonable
 >explanations of optimization effects.)

I received a number of questions asking that the -O flag did to the results.
I ran the 1.1 Dhrystones on our local 5890/300E without the -O flag and got:

    Dhrystone(1.1) time for 500000 passes = 11
    This machine benchmarks at 42253 dhrystones/second

This was ran a loaded system in mid-day, though it does suggest that the posted
results of:

     Amdahl     5890/300E            0.00 43604 43668 UTS 1.2 (SVR2),cc  -O,

doesn't repend too much on the -O flag.

chongo <> /\oo/\
-- 
[views above shouldn't be viewed as Amdahl views, or as views from Amdahl, or
 as Amdahl views views, or as views by Mr. Amdahl, or as views from his house]