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=-0.3 required=5.0 tests=BAYES_00,
	REPLYTO_WITHOUT_TO_CC autolearn=no autolearn_force=no version=3.4.4
X-Google-Thread: 103376,c1bdceb867926fdb
X-Google-NewGroupId: yes
X-Google-Attributes: gida07f3367d7,domainid0,public,usenet
X-Google-Language: ENGLISH,ASCII
Path: 
 g2news1.google.com!news2.google.com!news1.google.com!border1.nntp.dca.giganews.com!nntp.giganews.com!feedme.ziplink.net!news.swapon.de!newsfeed.straub-nv.de!open-news-network.org!newsfeed01.sul.t-online.de!t-online.de!newsfeed.arcor.de!newsspool1.arcor-online.net!news.arcor.de.POSTED!not-for-mail
From: "Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de>
Subject: Re: Interfacing Ada with C
Newsgroups: comp.lang.ada
User-Agent: 40tude_Dialog/2.0.15.1
MIME-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
Reply-To: mailbox@dmitry-kazakov.de
Organization: cbb software GmbH
References: 
 <b9fdff2c-7d8d-492c-af8d-313fdb1568d5@k19g2000yqc.googlegroups.com>
 <m24ofou965.fsf@pushface.org>
 <0ee9eec7-6024-4fb8-8df0-f65c146e4b84@i28g2000yqa.googlegroups.com>
 <m2zkxgrmkc.fsf@pushface.org>
 <143ef70b-7e74-426b-a621-a5fd157849be@x21g2000yqa.googlegroups.com>
 <m2iq43rvv1.fsf@pushface.org>
 <da987804-3948-4871-ab52-4a8e95f06d44@k39g2000yqb.googlegroups.com>
 <m2eieqrpq5.fsf@pushface.org>
Date: Mon, 26 Jul 2010 22:39:37 +0200
Message-ID: <13wnhsc8an2ne.135d54f2p02nn$.dlg@40tude.net>
NNTP-Posting-Date: 26 Jul 2010 22:39:36 CEST
NNTP-Posting-Host: 5f22aed7.newsspool3.arcor-online.net
X-Trace: 
 DXC=hAia:JcPYLGn`gW2MTm]<CMcF=Q^Z^V3H4Fo<]lROoRA8kF<OcfhCOKa]LNHib7JTK[6LHn;2LCVN[<mhadbfdUKM7I50]8=2XIl\K]W2IfJCG
X-Complaints-To: usenet-abuse@arcor.de
Xref: g2news1.google.com comp.lang.ada:12592
Date: 2010-07-26T22:39:36+02:00
List-Id: <comp.lang.ada>

On Mon, 26 Jul 2010 20:46:10 +0100, Simon Wright wrote:

> Ada novice <posts@gmx.us> writes:
>> On Jul 26, 1:21�am, Simon Wright <si...@pushface.org> wrote:

>>>(b) the results
>>> are the same with GNAT GPL 2010 and GCC 4.5.0, (c) the output looks good
>>> (from your inputs) ..
>>>
>>> � �$ ./test_zgeev
>>> � � 2.00000000000000E+00 �4.00000000000000E+00
>>> � � 2.00000000000000E+00 -4.00000000000000E+00
>>> � � 9.99999999999996E-01 �2.07319734774360E-16
>>
>>
>> I have Win XP and gcc (GCC) 4.3.6 20100603 for GNAT GPL 2010
>> (20100603). I get slightly different output:
>>
>>  1.99999999999999E+00  3.99999999999999E+00
>>  2.00000000000000E+00 -4.00000000000000E+00
>>  1.00000000000000E+00 -3.45920620709768E-16
>>
>> Your GCC version is more recent than mine. Maybe this is causing the
>> discrepancies or maybe it's because of your different OS. Did you
>> install your GCC separately? I got mine bundled with the GNAT GPL
>> 2010.
> 
> Mine was part of GNAT GPL 2010, but on Mac OS X LAPACK and BLAS come
> with the OS; I suspect that on Windows they don't.

It makes little sense to compare decimal representations, which aren't
exact on most machines. Then before comparing any results, they should be
rounded to the accuracy of the given method.

-- 
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de