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,FREEMAIL_FROM
	autolearn=unavailable autolearn_force=no version=3.4.4
X-Received: by 2002:a24:db06:: with SMTP id c6mr2334560itg.47.1559862851587;
        Thu, 06 Jun 2019 16:14:11 -0700 (PDT)
X-Received: by 2002:aca:ea05:: with SMTP id i5mr1828958oih.51.1559862851284;
 Thu, 06 Jun 2019 16:14:11 -0700 (PDT)
Path: 
 eternal-september.org!reader01.eternal-september.org!feeder.eternal-september.org!news.dns-netz.com!news.freedyn.net!newsfeed.xs4all.nl!newsfeed9.news.xs4all.nl!85.12.16.69.MISMATCH!peer02.ams1!peer.ams1.xlned.com!news.xlned.com!peer02.am4!peer.am4.highwinds-media.com!peer03.iad!feed-me.highwinds-media.com!news.highwinds-media.com!s188no115808itb.0!news-out.google.com!l135ni226itc.0!nntp.google.com!s188no115806itb.0!postnews.google.com!glegroupsg2000goo.googlegroups.com!not-for-mail
Newsgroups: comp.lang.ada
Date: Thu, 6 Jun 2019 16:14:10 -0700 (PDT)
In-Reply-To: <01dfb261-56b9-4bbb-aad9-e0099071f559@googlegroups.com>
Complaints-To: groups-abuse@google.com
Injection-Info: glegroupsg2000goo.googlegroups.com;
 posting-host=2001:8004:1420:18a6:3c2c:6580:a925:cfe;
 posting-account=rfeywQoAAAC0TKn5ZjdVW0ytcQM1oMSv
NNTP-Posting-Host: 2001:8004:1420:18a6:3c2c:6580:a925:cfe
References: <55b14350-e255-406c-ab11-b824da77995b@googlegroups.com>
 <01dfb261-56b9-4bbb-aad9-e0099071f559@googlegroups.com>
User-Agent: G2/1.0
MIME-Version: 1.0
Message-ID: <431f7439-e540-4f8e-bb6d-8542cceab52f@googlegroups.com>
Subject: Re: Toy computational "benchmark" in Ada (new blog post)
From: David Trudgett <dktrudgett@gmail.com>
Injection-Date: Thu, 06 Jun 2019 23:14:11 +0000
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
X-Received-Bytes: 3316
X-Received-Body-CRC: 557082291
Xref: reader01.eternal-september.org comp.lang.ada:56513
Date: 2019-06-06T16:14:10-07:00
List-Id: <comp.lang.ada>

Il giorno venerd=C3=AC 7 giugno 2019 03:48:19 UTC+10, Olivier Henley ha scr=
itto:
>=20
> Interesting but did you run the C version on your machine alongside your =
Ada version?
>=20
> In the text, from what I understand, you are comparing the Ada computatio=
n time to the reported OP C performance (17ms) and I fail to understand how=
 the comparison can be valid if performed on different machines?  =20
>=20
> Note: I tried to find the original C code but in vain and from my rapid r=
eading of your blog post you do not mention having run that C code yourself=
.
>=20
> Thank you,
>=20
> Olivier

Good point, Olivier. It would have been ideal to run the same C code on my =
machine, but alas! there are difficulties:

o I would have to run the example on Windows (maybe a particular version?)

o I would have to run Visual Studio C compiler (maybe a particular version?=
)

o I would have to research how to enable AVX2 instructions in that environm=
ent.

o I would have to guess the exact configurations, etc., used originally.

None of those points is "fun", which was the only point of the exercise, re=
ally.

Having said that, I do believe that the machine I ran this on, being quite =
a few years old by now, is probably within the ballpark of the original har=
dware. But that is difficult to say, because it is not just (or even mostly=
) a processor and clock speed that will determine the result of this exerci=
se, but also RAM speed, cache speed and size, bus bandwidths, and lots of o=
ther things. (Which is why, as you suggest, it would have been ideal to run=
 the same code on my setup.)

If it makes any difference (you may wish to try it yourself), I have made t=
he Ada code available here:

https://gitlab.com/DataLinkDroid/ada-2012-procedural-map-reduce/tree/master=
/src

Regards,
David