<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<p><font face="serif">Using a boxsize of 167, the final runtimes for
the different groups differed significantly from each other. I
ran these on a similar processor to hammer (2.2 GHz Xeon class)
with roughly half of the L3 cache and 1/4 of the L2 cache. <br>
</font></p>
<p><font face="serif"><br>
</font></p>
<p><font face="serif"><b>First Place:</b> Power Rangers 41 minutes
on 9 threads<br>
</font></p>
<p><font face="serif"><b>Second Place:</b> Drunk@C 65 minutes on
9 threads<br>
</font></p>
<p><font face="serif"><b>Third Place:</b> GnuForce 113 minutes
on 12 threads<br>
</font></p>
<p><br>
</p>
<p>For all groups the times with partitions were roughly double the
diffusion times without partitions.</p>
<p>With respect to C and Fortran usage there were also some
disparities. In the case of the PowerRanges, the fortran times
and C times were very close to each other. Sometimes the fortran
was a few seconds faster. For the other groups differences in the
Fortran and C runtimes were huge. In the worst case the fortran
was three times slower than the C and in the best case it took
twice as long as the C.</p>
<p><br>
</p>
<p><br>
</p>
<pre class="moz-signature" cols="72">--
Andrew J. Pounds, Ph.D. (<a class="moz-txt-link-abbreviated" href="mailto:pounds_aj@mercer.edu">pounds_aj@mercer.edu</a>)
Professor of Chemistry and Computer Science
Mercer University, Macon, GA 31207 (478) 301-5627
<a class="moz-txt-link-freetext" href="http://faculty.mercer.edu/pounds_aj">http://faculty.mercer.edu/pounds_aj</a>
</pre>
</body>
</html>