<div dir="ltr">While you're on to it.<div>I expect to read in a benchmark report, the number of time the software was executed, the mean running time, and the standard deviation. Running and timing it once can hide a pretty large error.<br>
<br><div class="gmail_quote">On Sun, Jul 18, 2010 at 11:37 AM, Shlomi Fish <span dir="ltr"><<a href="mailto:shlomif@iglu.org.il">shlomif@iglu.org.il</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
Hi Nadav,<br>
<div class="im"><br>
On Sunday 18 Jul 2010 10:03:32 Nadav Har'El wrote:<br>
> On Sat, Jul 17, 2010, Shlomi Fish wrote about "New Freecell Solver gcc-4.5.0<br>
vs. LLVM+clang Benchmark":<br>
> > On the other hand, with gcc-4.5.0 with "-flto" and "-fwhole-program"<br>
> > Freecell Solver ran at 85.1303749084473 seconds.<br>
> ><br>
> > I admit that I ran the gcc benchmark with a good renice and only in the<br>
> > virtual console, while running the LLVM/clang benchmark without a renice<br>
> > and in KDE and Compiz, but it still cannot explain the dramatic<br>
> > difference.<br>
><br>
> Two nitpicks:<br>
><br>
> 1. Instead of admitting to not running the two benchmarks in the same<br>
> conditions, can't you spare another 85 seconds (!) and run one of them<br>
> again?<br>
<br>
</div>Yes, I can. I'm on to it, which will take a little longer because I've deleted<br>
the svn checkout of LLVM today before I read your E-mail.<br>
<div class="im"><br>
><br>
> 2. Do you really think that your measurements are accurate down to the<br>
> individual picosecond? :-)<br>
<br>
</div>No, I don't. But that's what I copy and paste from my timing program which is<br>
using gettimeofday() and that's what I get after being processed with a Perl<br>
script.<br>
<div class="im"><br>
><br>
> Anyway, I guess that in any case it shows that gcc has nothing to be<br>
> ashamed of.<br>
<br>
</div>Well, at least not in comparison to clang and LLVM. Reportedly, gcc does not<br>
yield as good results as, say, Intel's icc. (But see:<br>
<a href="http://yro.slashdot.org/article.pl?sid=05/07/12/1320202&tid=142&tid=118&tid=123" target="_blank">http://yro.slashdot.org/article.pl?sid=05/07/12/1320202&tid=142&tid=118&tid=123</a><br>
<div class="im">).<br>
<br>
Regards,<br>
<br>
Shlomi Fish<br>
<br>
--<br>
-----------------------------------------------------------------<br>
Shlomi Fish <a href="http://www.shlomifish.org/" target="_blank">http://www.shlomifish.org/</a><br>
</div>Stop Using MSIE - <a href="http://www.shlomifish.org/no-ie/" target="_blank">http://www.shlomifish.org/no-ie/</a><br>
<div class="im"><br>
God considered inflicting XSLT as the tenth plague of Egypt, but then<br>
decided against it because he thought it would be too evil.<br>
<br>
Please reply to list if it's a mailing list post - <a href="http://shlom.in/reply" target="_blank">http://shlom.in/reply</a> .<br>
<br>
_______________________________________________<br>
</div><div><div></div><div class="h5">Linux-il mailing list<br>
<a href="mailto:Linux-il@cs.huji.ac.il">Linux-il@cs.huji.ac.il</a><br>
<a href="http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il" target="_blank">http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il</a><br>
</div></div></blockquote></div><br></div></div>