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: 103376,38fc011071df5a27
X-Google-Attributes: gid103376,public
X-Google-ArrivalTime: 2003-06-05 11:02:38 PST
Path: 
 archiver1.google.com!news1.google.com!newsfeed.stanford.edu!logbridge.uoregon.edu!arclight.uoregon.edu!wn13feed!wn12feed!worldnet.att.net!204.127.198.203!attbi_feed3!attbi.com!rwcrnsc53.POSTED!not-for-mail
Message-ID: <3EDF8594.3000901@attbi.com>
From: "Robert I. Eachus" <rieachus@attbi.com>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US;
 rv:1.0.2) Gecko/20021120 Netscape/7.01
X-Accept-Language: en-us, en
MIME-Version: 1.0
Newsgroups: comp.lang.ada
Subject: Re: Ideas for Ada 200X
References: <6a90b886.0305262344.1d558079@posting.google.com>
 <3ED41344.7090105@spam.com> <3ED46D81.FF62C34F@0.0> <3ED46E07.4340CABC@0.0>
 <3ED4F3FD.A0EF7079@alfred-hilscher.de> <xIeBa.1017346$3D1.597085@sccrnsc01>
 <bebbba07.0305292326.75f5fbef@posting.google.com>
 <6vWcnTWjF83bD0qjXTWcpA@gbronline.com>
 <bebbba07.0305311049.753bb8f0@posting.google.com>
 <KvydnZ2dnunHfUej4p2dnA@gbronline.com>
 <bebbba07.0306022127.46f6c998@posting.google.com>
 <3EDCBDF4.1050900@attbi.com>
 <bebbba07.0306031420.69c20f71@posting.google.com>
 <3EDEC9A7.9050602@attbi.com> <7vy90h2akm.fsf@vlinux.voxelvision.no>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
NNTP-Posting-Host: 24.62.164.137
X-Complaints-To: abuse@attbi.com
X-Trace: rwcrnsc53 1054836157 24.62.164.137 (Thu, 05 Jun 2003 18:02:37 GMT)
NNTP-Posting-Date: Thu, 05 Jun 2003 18:02:37 GMT
Organization: AT&T Broadband
Date: Thu, 05 Jun 2003 18:02:37 GMT
Xref: archiver1.google.com comp.lang.ada:38720
Date: 2003-06-05T18:02:37+00:00
List-Id: <comp.lang.ada>

Ole-Hjalmar Kristensen wrote:

> As you say, it depends on the system whether or not it pays to turn
> off the checks. Here are the results from my system. For Integer
> turning off the checks is a win, for Float it does not seem to matter.
> One other difference: with checks on, the integer "+" is actually
> slightly faster than add.  With checks off, it's 3 times slower.

And you can see why I provided all eight cases.  (I probably should have 
add four Long_Float cases as well.)  What you will see depends more on 
the actual hardware than aything else.  You can be compute bound, and 
the checks will be a significant penalty, or you can be memory access 
bound, and the checks will add nothing.  Also how much of the implicit 
copy is from cache affects the timing.  (Incidently, I didn't do this 
but if reverse the outside loop in the "+" (for I in reverse...) this 
will speed things up if the whole array doesn't fit in cache.