From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00 autolearn=ham
	autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,8c8550b9f2cf7d40
X-Google-Attributes: gid103376,public
X-Google-ArrivalTime: 2003-06-10 12:47:37 PST
Path: 
 archiver1.google.com!news1.google.com!newsfeed.stanford.edu!logbridge.uoregon.edu!news.tele.dk!news.tele.dk!small.news.tele.dk!newsfeed1.e.nsc.no!nsc.no!nextra.com!news2.e.nsc.no.POSTED!53ab2750!not-for-mail
Newsgroups: comp.lang.ada
Subject: Re: Is ther any sense in *= and matrices?
References: <mVpFa.110557$DV.125036@rwcrnsc52.ops.asp.att.net>
From: Ole Kristensen <ole@linux.local>
Message-ID: <m3adcp3e8w.fsf@linux.local>
Organization: Linux Private Site
User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.2
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
NNTP-Posting-Host: 148.122.128.188
X-Complaints-To: news-abuse@telenor.net
NNTP-Posting-Date: Tue, 10 Jun 2003 21:47:37 MEST
X-Trace: news2.ulv.nextra.no 1055274457 148.122.128.188
Date: 10 Jun 2003 21:48:47 +0200
Xref: archiver1.google.com comp.lang.ada:38932
Date: 2003-06-10T21:48:47+02:00
List-Id: <comp.lang.ada>

tmoran@acm.org writes:

> To summarize some of the benchmarks of A=A+B vs A+=B, 3x3 matrices, 10**7 iterations
> A=A+B  A+=B  ratio
> 55      19    2.9     Russ's special code on Sunblade, slightly faster than
>                       an 866MHz Pentium III according to
>                       http://www.phys.ocean.dal.ca/~kelley/linux/benchmark/benchmark_POM.html
> 4.4      3.3  1.3     MSVC++5.0 simple code on Tom's 866MHz Pentium III
> 1.4      0.8  1.6     Gnat 3.15p on Tom's 866MHz Pentium III
> 1.1      0.7  1.6     g++.5.8 (GCC) 3.2.2 using C templates on Hyman's "Sun workstation", speed unspecified
> 0.45     0.18 2.5     Ole-Hjalmar's machine ? gcc Ada or C the same?

Another data point: My old home PC, a 266MHz pentium II, compiler g++
3.2, Tom's simple C++ code:

ole@linux:~> time ./tmoran
Go 1000000 4460000 6320000
real    0m13.626s
user    0m10.800s
sys     0m0.000s

If I read the results correctly, it actually beats the SunBlade 2000,
which it definitely should not do. But there is much less of a
difference between + and += on this machine than on the fast machine
at work.  Maybe I should check the g++ version on my machine at work
once more.

Turning optimization off produces even less difference:

ole@linux:~> g++ -o tmoran tmoran.cc
ole@linux:~> !time
time ./tmoran
Go 1000000 12850000 14120000
real    0m34.219s
user    0m26.950s
sys     0m0.040s

Ole-Hj. Kristensen