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=ham autolearn_force=no version=3.4.4 X-Google-Thread: a07f3367d7,deb179aa052b8e20 X-Google-Attributes: gida07f3367d7,public,usenet X-Google-NewGroupId: yes X-Google-Language: ENGLISH,ASCII-7-bit Path: g2news1.google.com!postnews.google.com!c3g2000yqd.googlegroups.com!not-for-mail From: jonathan Newsgroups: comp.lang.ada Subject: Re: Shootout News: 64 Bits of Sweet Pancake Date: Tue, 15 Dec 2009 03:25:42 -0800 (PST) Organization: http://groups.google.com Message-ID: <8517623a-9092-488d-835c-f375320a40df@c3g2000yqd.googlegroups.com> References: <4b2670db$0$6549$9b4e6d93@newsspool4.arcor-online.net> NNTP-Posting-Host: 143.117.23.238 Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 X-Trace: posting.google.com 1260876342 19843 127.0.0.1 (15 Dec 2009 11:25:42 GMT) X-Complaints-To: groups-abuse@google.com NNTP-Posting-Date: Tue, 15 Dec 2009 11:25:42 +0000 (UTC) Complaints-To: groups-abuse@google.com Injection-Info: c3g2000yqd.googlegroups.com; posting-host=143.117.23.238; posting-account=Jzt5lQoAAAB4PhTgRLOPGuTLd_K1LY-C User-Agent: G2/1.0 X-HTTP-UserAgent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.0.15) Gecko/2009102814 Iceweasel/3.0.6 (Debian-3.0.6-3),gzip(gfe),gzip(gfe) Xref: g2news1.google.com comp.lang.ada:8392 Date: 2009-12-15T03:25:42-08:00 List-Id: A few comments on what we've been doing and why. Rankings are fun (well, at the moment anyway, when all goes well) but they sure change fast. I'ld expect someone to submit something faster in the next few weeks. The important thing in my view was getting some Ada multi-core programs into the public view. Here, by the way, are the present overall rankings of the Ada programs in the 64-bit tests: http://shootout.alioth.debian.org/u64/benchmark.php?test=all&lang=all&box=1 http://shootout.alioth.debian.org/u64q/benchmark.php?test=all&lang=all&box=1 I remember that after the 6 months we recently spent updating the Ada suite, the Ada median ranking was at about 1.3 on the above scale (64-bit tests). Three days ago it at about 1.7. Now (Dec 15) its about 1.25. I won't dare predict Dec 16. Here BTW are the number of page-views to the shootout site over the last 5 or so years (number 2 below, just after Debian Linux): https://alioth.debian.org/top/toplist.php?type=pageviews_proj 1 Debian Port to AMD64 3938690 2 The Computer Language Benchmarks Game 2179690 3 debian-installer 1402511 That's a lot of hits. I was getting tired of hearing this cited as definitive evidence that Ada was 4 times slower than C, or X times slower or whatever. The test machine has 4 cores available for processing, and the contest has become a test of language/programmer ability to exploit these 4 cores in parallel. Most of the Ada programs weren't multi-tasking/multi-core when they were originally written, and they were a factor of 4 slower than the competition for that reason alone. For typical students or programmers, browsing the shootout site may be the only exposure they'll ever get to Ada code, Ada runtime performance, or the Ada approach to exploiting multi-core processors, so these updates were long overdue. Finally, the fannkuch numbers - more fun than fractals. I just invested about 3.5 million processor-seconds in getting the 16th fannkuch number (139). We had an idle 16-core machine sitting around. I don't suppose there's a table of these numbers sitting around somewhere. The 17th will take about 70 million processor-seconds to calculate if I have to do it myself. Is it a prime number like the 14th, 15th, and 16th? The more interesting thing is long term trend. To find out what that is, plot the cubed root of the fannkuch numbers (the 2nd column) as a function of the 1st column: 2 1 3 2 4 4 5 7 6 10 7 16 8 22 9 30 10 38 11 51 12 65 13 80 14 101 15 113 16 139 Jonathan