<html>
  <head>

    <meta http-equiv="content-type" content="text/html; charset=UTF-8">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    <br>
    <div class="moz-forward-container">I'll keep thinking about it, but
      in the meantime, I noticed that when displaying <br>
      <br>
      a = A*inv(A)<br>
      <br>
      we get the customary list of columns as the total width exceeds
      the available horizontal space. However, we get one "column 1"
      header and three "column 2" headers instead of "column 1", ...,
      "column 4". Seemingly it is not only the header which is repeated,
      but aldo the columns, i.e., column 2 appears three times.<br>
      <br>
      This seems to be a display bug.  <br>
      <br>
      Regards, <br>
      <br>
      Federico Miyara<br>
      <br>
       <br>
      <br>
      <blockquote type="cite"
cite="mid:170bffc2b46.d855246737750.4965631307156628552@bytecode-asia.com">
        <div style="font-family: Verdana, Arial, Helvetica, sans-serif;
          font-size: 10pt;">
          <div>Back to the possible reason of causing this issue when u
            substitute a value into the %s  into A*inv(A)<br>
          </div>
          <div><br>
          </div>
          <div>Assume a = A*inv(A), a(1,1) which suppose to be ~ 1 is : <br>
          </div>
          <div><br>
          </div>
          <div>--> a(1,1) <br>
          </div>
          <div>ans  =<br>
          </div>
          <div><br>
          </div>
          <div>                                                         
                      <br>
          </div>
          <div>   1.630D-38 + 1.586D-40s + 4.110D-43s^2 + 1.120D-46s^3 +
            1.998D-50s^4 + 3.586D-55s^5 <br>
          </div>
          <div>  
--------------------------------------------------------------------------- <br>
          </div>
          <div>                                                         
                            <br>
          </div>
          <div>                      1.630D-38 + 4.283D-42s +
            7.960D-46s^2<br>
          </div>
          <div><br>
          </div>
          <div>the coefficients in numerator and denominator are very
            small value and with complex number involve, there might be
            come kind of limitation in the numerical software. <br>
          </div>
          <div><br>
          </div>
          <div>this could be shown if you subs with smaller number: <br>
          </div>
          <div><br>
          </div>
          <div>--> horner(A*inv(A), %i*2*%pi*1) <br>
          </div>
          <div>ans  =<br>
          </div>
          <div><br>
          </div>
          <div>   0.9991047 + 0.0595012i   0.0000225 - 5.350D-08i  
            0.0000132 - 5.350D-08i   0.0000132 - 5.350D-08i<br>
          </div>
          <div>   0.0000552 - 0.1037974i   0.9990938 + 0.0595013i 
            -0.0008828 - 2.194D-09i  -0.0008828 - 2.194D-09i<br>
          </div>
          <div>   0.0000229 - 7.699D-10i   0.0008778 - 1.540D-09i  
            1.0008549 + 0.0595013i   0.0008789 - 1.549D-09i<br>
          </div>
          <div>  -0.0000645 + 0.1025666i  -0.0008778 + 1.540D-09i 
            -0.0008789 + 1.549D-09i   0.9990971 + 0.0595013i<br>
          </div>
          <div><br>
          </div>
          <div>you get the correct answer<br>
          </div>
          <div><br>
          </div>
          <div>This is so far I could go, further details such as how
            the a double precision real or complex number handled in
            numerical software are beyond my knowledge. <img
              ref="image"
              src="cid:part1.ED4381C2.42E874EC@fceia.unr.edu.ar"
              data-attr="smiley" title="Smile" style="vertical-align:
              middle; width: 24px; height: 24px;" class=""><br>
          </div>
          <div>thanks.<br>
          </div>
          <div><br>
          </div>
          <div>Regards,<br>
          </div>
          <div>Chin Luh</div>
          <div><br>
          </div>
          <div><br>
          </div>
          <div><br>
          </div>
          <div class="zmail_extra" style=""
            data-zbluepencil-ignore="true">
            <div><br>
            </div>
            <div id="Zm-_Id_-Sgn1">---- On Mon, 09 Mar 2020 22:20:38
              +0800 <b>Antoine Monmayrant <a
                  class="moz-txt-link-rfc2396E"
                  href="mailto:amonmayr@laas.fr" moz-do-not-send="true"><amonmayr@laas.fr></a></b>
              wrote ----<br>
            </div>
            <div><br>
            </div>
            <blockquote style="border-left: 1px solid rgb(204, 204,
              204); padding-left: 6px; margin: 0px 0px 0px 5px;">
              <div>
                <div>Hello,<br>
                </div>
                <div><br>
                </div>
                <div>It seems that this value is below %eps==2.2e-16, so
                  I though that once encoded in double the difference
                  between 2e-20 and 0 is too small to be properly taken
                  into account.<br>
                </div>
                <div>Am I missing something?<br>
                </div>
                <div><br>
                </div>
                <div>Antoine<br>
                </div>
                <div><br>
                </div>
                <div><br>
                </div>
                <div>Le Lundi, Mars 09, 2020 15:11 CET, Chin Luh Tan
                  <<a href="mailto:chinluh.tan@bytecode-asia.com"
                    target="_blank" moz-do-not-send="true">chinluh.tan@bytecode-asia.com</a>>
                  a écrit:<br>
                </div>
                <div> <br>
                </div>
                <blockquote> <br>
                </blockquote>
                <div style="font-family: Verdana, Arial, Helvetica,
                  sans-serif;font-size: 10.0pt;">
                  <div>initially I thought could it be possible the
                    number is too small (or too large) to be handled  by
                    double? <br>
                  </div>
                  <div> <br>
                  </div>
                  <div>after trying scaling down the problem, I notice
                    that in Scilab 6.0.2 onwards:<br>
                  </div>
                  <div> <br>
                  </div>
                  <div>--> 2e-20*%i<br>
                  </div>
                  <div>ans  =<br>
                  </div>
                  <div> <br>
                  </div>
                  <div>   0. <br>
                  </div>
                  <div> <br>
                  </div>
                  <div>in Scilab 5.5.2<br>
                  </div>
                  <div> <br>
                  </div>
                  <div>-->2e-20*%i<br>
                  </div>
                  <div>ans  =<br>
                  </div>
                  <div> <br>
                  </div>
                  <div>    2.000D-20i <br>
                  </div>
                  <div> <br>
                  </div>
                  <div> <br>
                  </div>
                  <div>Could this be the issue?<br>
                  </div>
                  <div> <br>
                  </div>
                  <div>thanks.<br>
                  </div>
                  <div> <br>
                  </div>
                  <div>rgds,<br>
                  </div>
                  <div>CL<br>
                  </div>
                  <div class="x_-1527920102zmail_extra">
                    <div> <br>
                    </div>
                    <div>---- On Mon, 09 Mar 2020 21:04:12 +0800 <b>Stéphane
                        Mottelet <<a
                          href="mailto:stephane.mottelet@utc.fr"
                          target="_blank" moz-do-not-send="true">stephane.mottelet@utc.fr</a>></b>
                      wrote ----<br>
                    </div>
                    <div> <br>
                    </div>
                    <blockquote style="border-left: 1.0px solid
                      rgb(204,204,204);padding-left: 6.0px;margin: 0.0px
                      0.0px 0.0px 5.0px;">
                      <div>Maybe an inversion made in state-space could
                        me more reliable ?<br>
                        <br>
                        S.<br>
                        <br>
                        Le 09/03/2020 à 13:36, <a target="_blank"
                          href="mailto:sgougeon@free.fr"
                          moz-do-not-send="true">sgougeon@free.fr</a> a
                        écrit :<br>
                        > Hello Federico,<br>
                        ><br>
                        > cond() and rcond() do not accept rationals,
                        but the determinant of A is very close to 0:<br>
                        > --> det(A)<br>
                        > ans =<br>
                        ><br>
                        > 3.573D-10<br>
                        > ------------------------<br>
                        > 0.0000027s² +7.433D-10s³<br>
                        ><br>
                        > This likely explains that the inversion is
                        not reliable.<br>
                        ><br>
                        > Regards<br>
                        > Samuel<br>
                        >
                        _______________________________________________<br>
                        > users mailing list<br>
                        > <a target="_blank"
                          href="mailto:users@lists.scilab.org"
                          moz-do-not-send="true">users@lists.scilab.org</a><br>
                        > <a target="_blank"
href="https://antispam.utc.fr/proxy/1/c3RlcGhhbmUubW90dGVsZXRAdXRjLmZy/lists.scilab.org/mailman/listinfo/users"
                          moz-do-not-send="true">https://antispam.utc.fr/proxy/1/c3RlcGhhbmUubW90dGVsZXRAdXRjLmZy/lists.scilab.org/mailman/listinfo/users</a><br>
                        <br>
                        --<br>
                        Stéphane Mottelet<br>
                        Ingénieur de recherche<br>
                        EA 4297 Transformations Intégrées de la Matière
                        Renouvelable<br>
                        Département Génie des Procédés Industriels<br>
                        Sorbonne Universités - Université de Technologie
                        de Compiègne<br>
                        CS 60319, 60203 Compiègne cedex<br>
                        Tel : +33(0)344234688<br>
                        <a target="_blank"
                          href="http://www.utc.fr/~mottelet"
                          moz-do-not-send="true">http://www.utc.fr/~mottelet</a><br>
                        <br>
                        _______________________________________________<br>
                        users mailing list<br>
                        <a target="_blank"
                          href="mailto:users@lists.scilab.org"
                          moz-do-not-send="true">users@lists.scilab.org</a><br>
                        <a target="_blank"
                          href="http://lists.scilab.org/mailman/listinfo/users"
                          moz-do-not-send="true">http://lists.scilab.org/mailman/listinfo/users</a></div>
                    </blockquote>
                  </div>
                  <div> <br>
                  </div>
                </div>
                <div><br>
                </div>
                <div>  _______________________________________________ <br>
                </div>
                <div>users mailing list <br>
                </div>
                <div><a href="mailto:users@lists.scilab.org"
                    target="_blank" moz-do-not-send="true">users@lists.scilab.org</a>
                  <br>
                </div>
                <div><a
                    href="http://lists.scilab.org/mailman/listinfo/users"
                    target="_blank" moz-do-not-send="true">http://lists.scilab.org/mailman/listinfo/users</a>
                  <br>
                </div>
              </div>
            </blockquote>
          </div>
          <div><br>
          </div>
        </div>
        <br>
        <br>
        <fieldset class="mimeAttachmentHeader"></fieldset>
        <pre class="moz-quote-pre" wrap="">_______________________________________________
users mailing list
<a class="moz-txt-link-abbreviated" href="mailto:users@lists.scilab.org" moz-do-not-send="true">users@lists.scilab.org</a>
<a class="moz-txt-link-freetext" href="http://lists.scilab.org/mailman/listinfo/users" moz-do-not-send="true">http://lists.scilab.org/mailman/listinfo/users</a>
</pre>
      </blockquote>
      <br>
    </div>
  </body>
</html>