<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix">Jack,<br>
      <br>
      Use an editor to modify ClpMain.cpp.<br>
      <br>
      Change all&nbsp; %15.8g to %17.10g (or use %..f) - that should do it.&nbsp;
      The 8 accuracy is why you get the ,80.<br>
      <br>
      If that is OK then I can modify to allow user to use something
      like -DOUTPUT_FORMAT=17.10g in configure to override default.<br>
      <br>
      John Forrest<br>
      <br>
      On 15/07/13 05:19, Jack Hagart wrote:<br>
    </div>
    <blockquote
cite="mid:CANztv9oznaRUr8TL9uue_D_-LzvYdOGhB2yeBS1XM4cDDiXrCg@mail.gmail.com"
      type="cite">
      <meta http-equiv="Context-Type" content="text/html;
        charset=ISO-8859-1">
      <div dir="ltr">
        <div>Hello everyone from CLP mail list. I need a help.</div>
        <div>&nbsp;</div>
        <div>I&acute;m using the CLP to a software I developed. CLP works very
          well to me. The problem is, when I started use some real data,
          with numbers greater, the CLP output file is using round to
          the values of the variables in the solution.</div>
        <div>&nbsp;</div>
        <div>I implemented the same mathematical model in GAMS/CPLEX,
          with exactly same input data, and see the result:</div>
        <div>&nbsp;</div>
        <div>
          <table rules="NONE" frame="VOID" cols="3">
            <colgroup><col width="26"><col width="121"><col width="125"></colgroup>
            <tbody>
              <tr>
                <td height="17" width="26"><br>
                </td>
                <td width="121">CLP</td>
                <td width="125">GAMS/CPLEX</td>
              </tr>
              <tr>
                <td height="17">1</td>
                <td>$43.597.637,00</td>
                <td>$43.597.636,76</td>
              </tr>
              <tr>
                <td height="17">2</td>
                <td>$23.108.017,00</td>
                <td>$23.108.017,02</td>
              </tr>
              <tr>
                <td height="17">3</td>
                <td>$44.053.085,00</td>
                <td>$44.053.085,30</td>
              </tr>
              <tr>
                <td height="17">4</td>
                <td>$7.624.834,80</td>
                <td>$7.624.834,80</td>
              </tr>
              <tr>
                <td height="17">5</td>
                <td>$17.312.266,00</td>
                <td>$17.312.265,81</td>
              </tr>
              <tr>
                <td height="17">6</td>
                <td>$11.199.682,00</td>
                <td>$11.199.682,39</td>
              </tr>
              <tr>
                <td height="17">7</td>
                <td>$16.631.305,00</td>
                <td>$16.631.304,60</td>
              </tr>
              <tr>
                <td height="17">8</td>
                <td>$9.903.795,80</td>
                <td>$9.903.795,82</td>
              </tr>
              <tr>
                <td height="17">9</td>
                <td>$10.465.548,00</td>
                <td>$10.465.548,33</td>
              </tr>
              <tr>
                <td height="17">10</td>
                <td>$12.723.051,00</td>
                <td>$12.723.051,42</td>
              </tr>
              <tr>
                <td height="17">11</td>
                <td>$8.210.366,70</td>
                <td>$8.210.366,68</td>
              </tr>
            </tbody>
          </table>
        </div>
        <div>&nbsp;</div>
        <div>&nbsp;For example, the line 1. The right solution, that one that
          respects all the constraints, is $43.597.636,76, or the
          GAMS/CPLEX solution. The CLP plot $43.597.637,00. Its the same
          solution, but CLP round the decimals, and put 637 in the place
          of 636,76.</div>
        <div>&nbsp;</div>
        <div>The same is happening with others values of the solution.
          But one thing is strange. Take a look in the lines 4 and 8. In
          this cases, CLP didn&acute;t round the values!</div>
        <div>&nbsp;</div>
        <div>This is happening just to numbers greater than 10 million.
          I tried change the option output to 4 and 6 to improve
          precision but it didn&acute;t worked to CLP 1.14.4.</div>
        <div>&nbsp;</div>
        <div>Can someone help me with new ideas?</div>
        <div>&nbsp;</div>
        <div>A small piece of the original CLP solution report is
          bellow. The variable f(1,13) is that one in line 1 in the
          table above.</div>
        <div>&nbsp;</div>
        <div>
          optimal<br>
          Objective value -1.6527462e+008<br>
          &nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 1 f(24,27)&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 1.6526604e+008&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
          0<br>
          &nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 5 f(1,13)&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
          43597637&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; &nbsp;0<br>
          &nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 7 f(13,14)&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
          43597637&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0<br>
          &nbsp;&nbsp;&nbsp;&nbsp; 10 f(2,14)&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
          23108017&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0<br>
          &nbsp;&nbsp;&nbsp;&nbsp; 14 f(3,15)&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
          44053085&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0<br>
          &nbsp;&nbsp;&nbsp;&nbsp; 18 f(4,16)&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
          7624834.8&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0</div>
        <div>
          &nbsp;</div>
        <div>Best Regards,</div>
        <div>Jack</div>
        <div>&nbsp;</div>
        <div>
          <table rules="NONE" frame="VOID" cols="3">
            <colgroup><col width="26"><col width="121"><col width="125"></colgroup>
            <tbody>
              <tr>
              </tr>
              <tr>
                <td>
                  <p>&nbsp;</p>
                  <p>&nbsp;</p>
                  <p>&nbsp;</p>
                </td>
              </tr>
            </tbody>
          </table>
        </div>
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
Clp mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Clp@list.coin-or.org">Clp@list.coin-or.org</a>
<a class="moz-txt-link-freetext" href="http://list.coin-or.org/mailman/listinfo/clp">http://list.coin-or.org/mailman/listinfo/clp</a>
</pre>
    </blockquote>
    <br>
  </body>
</html>