<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
  <meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
  <title></title>
</head>
<body bgcolor="#ffffff" text="#000000">
    Thanks much for the help on my other questions this week.  As a
token of my esteem, here's another question.  -grin-<br>
<br>
    We have a problem that does the following:<br>
<br>
    Pick random number between 2.1 and 5.9 in .1 increments.<br>
    Multiply result by 10^9.<br>
    Divide result by 4.7*10^-23.  Save as answer<br>
<br>
    The settings, held over from this problem's original CAPA
incarnation, were 2% tolerance, 2 significant figures.<br>
<br>
    One line from a particular student submission report:<br>
<br>
<table
 _base_href="http://loncapa.gwu.edu/res/gwu/kmurphy5/Astronomy 2 - GWU-Produced/Homework/">
  <tbody
 _base_href="http://loncapa.gwu.edu/res/gwu/kmurphy5/Astronomy 2 - GWU-Produced/Homework/">
    <tr class="even"
 _base_href="http://loncapa.gwu.edu/res/gwu/kmurphy5/Astronomy 2 - GWU-Produced/Homework/">
      <td
 _base_href="http://loncapa.gwu.edu/res/gwu/kmurphy5/Astronomy 2 - GWU-Produced/Homework/"
 valign="top">ascheire</td>
      <td
 _base_href="http://loncapa.gwu.edu/res/gwu/kmurphy5/Astronomy 2 - GWU-Produced/Homework/"
 valign="top">gwu</td>
      <td
 _base_href="http://loncapa.gwu.edu/res/gwu/kmurphy5/Astronomy 2 - GWU-Produced/Homework/"
 valign="top">147820964</td>
      <td
 _base_href="http://loncapa.gwu.edu/res/gwu/kmurphy5/Astronomy 2 - GWU-Produced/Homework/"
 valign="top">none</td>
      <td
 _base_href="http://loncapa.gwu.edu/res/gwu/kmurphy5/Astronomy 2 - GWU-Produced/Homework/"
 valign="top">1.0e38</td>
      <td
 _base_href="http://loncapa.gwu.edu/res/gwu/kmurphy5/Astronomy 2 - GWU-Produced/Homework/"
 valign="top">1.0E+38</td>
      <td
 _base_href="http://loncapa.gwu.edu/res/gwu/kmurphy5/Astronomy 2 - GWU-Produced/Homework/"
 valign="top">INCORRECT</td>
      <td
 _base_href="http://loncapa.gwu.edu/res/gwu/kmurphy5/Astronomy 2 - GWU-Produced/Homework/"
 valign="top">1126752146</td>
      <td
 _base_href="http://loncapa.gwu.edu/res/gwu/kmurphy5/Astronomy 2 - GWU-Produced/Homework/"
 valign="top">1</td>
      <td
 _base_href="http://loncapa.gwu.edu/res/gwu/kmurphy5/Astronomy 2 - GWU-Produced/Homework/"
 valign="top"><br>
      </td>
    </tr>
  </tbody>
</table>
<br>
    Any thoughts?  One of our profs recalls that CAPA used to have a
similar problem where if the tolerance range did not cover the rounding
for sig figs, one could not possibly enter a correct answer.  We then
had to either increase sig figs or change the tolerance.<br>
<br>
    Any thoughts?  In this case the random number was 4.8 so the
unrounded result is 1.021e36 so 1.0e36 is certainly outside the 2%
tolerance.<br>
<br>
    I'm inclined to believe that we've run into this same problem.  Is
this expected behavior?  Should our perl script do the rounding for
us?  Should I put this in as a bug?<br>
<br>
    -P<br>
<pre class="moz-signature" cols="72">-- 
"Wait!  I'm getting one of those things.  You know, a headache with pictures." --Fry
--
Peter Kovac
Systems Coordinator

e: <a class="moz-txt-link-abbreviated" href="mailto:kovac@gwu.edu">kovac@gwu.edu</a>
o: (202) 994-3811
f: (202) 994-3001

725 21st St. NW
Physics Department, GWU
Washington, DC 20052


Public key for <a class="moz-txt-link-abbreviated" href="mailto:kovac@gwu.edu">kovac@gwu.edu</a>
-----BEGIN PGP PUBLIC KEY BLOCK-----
Version: GnuPG v1.2.2 (GNU/Linux)
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=5cYt
-----END PGP PUBLIC KEY BLOCK-----
</pre>
</body>
</html>