<html>
  <head>
    <meta content="text/html; charset=windows-1252"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <p>Dear Vahid,</p>
    <p>It seems to me the mp_grid is not actually used when running with
      -pp, but I'm not sure about that. You can try just setting mp_grid
      to 'num_kpoints 1 1', and see what happens, but for now I'd be
      careful with the results. As a quick check maybe you can set
      mp_grid to some ridiculous value and check if the results are
      still the same.<br>
    </p>
    <p>If it's actually the case that the mp_grid variable is not used
      in postproc_setup mode, then we should change the code such that
      it doesn't raise an error in this case.</p>
    <p>Best,</p>
    <p>Dominik<br>
    </p>
    <br>
    <div class="moz-cite-prefix">On 17.03.2017 15:29, Vahid Askarpour
      wrote:<br>
    </div>
    <blockquote cite="mid:3D773D14-C76C-4EBE-B8F0-EFE500EF726D@dal.ca"
      type="cite">
      <meta http-equiv="Content-Type" content="text/html;
        charset=windows-1252">
      <div class="">Dear Dominik,</div>
      <div class=""><br class="">
      </div>
      <div class="">I installed qe-6.1, which uses wannier90 version
        2.1. I ran a self-consistent run followed by a non
        self-consistent run on the following k-points:</div>
      <div class=""><br class="">
      </div>
      <div class="">   0.0000000   0.0000000   0.0000000  0.0312500<br
          class="">
           0.0000000   0.0000000   0.2500000  0.2500000<br class="">
           0.0000000   0.0000000  -0.5000000  0.1250000<br class="">
           0.0000000   0.2500000   0.2500000  0.1875000<br class="">
           0.0000000   0.2500000  -0.5000000  0.7500000<br class="">
           0.0000000   0.2500000  -0.2500000  0.3750000<br class="">
           0.0000000  -0.5000000  -0.5000000  0.0937500<br class="">
           0.2500000  -0.5000000  -0.2500000  0.1875000</div>
      <div class=""><br class="">
      </div>
      <div class="">Then I used the following si.sa.win file:</div>
      <div class=""><br class="">
      </div>
      <div class="">begin projections<br class="">
         Si:sp3<br class="">
        end projections<br class="">
        num_bands 12<br class="">
        num_wann   8<br class="">
        iprint   2<br class="">
        dis_win_min -1000.000<br class="">
        dis_win_max    17.000<br class="">
        dis_froz_min -1000.000<br class="">
        dis_froz_max     6.500<br class="">
        num_iter     500<br class="">
        dis_num_iter 1000<br class="">
        <br class="">
      </div>
      <div class="">nnkpts 8<br class="">
        begin nnkpts<br class="">
        1 1 0 0 0<br class="">
        1 2 0 0 0<br class="">
        1 3 0 0 0<br class="">
        1 4 0 0 0<br class="">
        1 5 0 0 0<br class="">
        1 6 0 0 0<br class="">
        1 7 0 0 0<br class="">
        1 8 0 0 0<br class="">
        end nnkpts</div>
      <div class=""><br class="">
      </div>
      <div class="">begin unit_cell_cart<br class="">
        bohr<br class="">
        -5.084373 0 5.084373<br class="">
        0 5.084373 5.084373<br class="">
        -5.084373 5.084373 0              <br class="">
        end_unit_cell_cart<br class="">
        <br class="">
        begin atoms_frac<br class="">
        Si       0.000000000   0.000000000   0.000000000<br class="">
        Si       0.250000000   0.250000000   0.250000000<br class="">
        end atoms_frac<br class="">
        <br class="">
        begin kpoints<br class="">
           0.0000000   0.0000000   0.0000000  <br class="">
           0.0000000   0.0000000   0.2500000  <br class="">
           0.0000000   0.0000000  -0.5000000  <br class="">
           0.0000000   0.2500000   0.2500000  <br class="">
           0.0000000   0.2500000  -0.5000000  <br class="">
           0.0000000   0.2500000  -0.2500000 <br class="">
           0.0000000  -0.5000000  -0.5000000  <br class="">
           0.2500000  -0.5000000  -0.2500000<br class="">
        end kpoints<br class="">
        <br class="">
      </div>
      <div class="">Here I used the same k-points as before and asked
        for nearest neighbors for k-point=1</div>
      <div class=""><br class="">
      </div>
      <div class="">I ran wannier90.x -pp si.sa in parallel and get the
        following error:</div>
      <div class=""><br class="">
      </div>
      <div class="">Error: You must specify dimensions of the
        Monkhorst-Pack grid by setting mp_grid</div>
      <div class=""><br class="">
      </div>
      <div class="">How do you specify an mp_grid if the k-points are
        chosen randomly? Or am I missing something in the input?</div>
      <div class=""><br class="">
      </div>
      <div class="">Thanks,</div>
      <div class=""><br class="">
      </div>
      <div class="">Vahid</div>
      <div class=""><br class="">
      </div>
      <br class="">
      <div>
        <blockquote type="cite" class="">
          <div class="">On Mar 16, 2017, at 8:46 PM, Vahid Askarpour
            <<a moz-do-not-send="true" href="mailto:vh261281@dal.ca"
              class="">vh261281@dal.ca</a>> wrote:</div>
          <br class="Apple-interchange-newline">
          <div class="">
            <div style="word-wrap: break-word; -webkit-nbsp-mode: space;
              -webkit-line-break: after-white-space;" class="">
              <div class="">Dear Dominik,</div>
              <div class=""><br class="">
              </div>
              <div class="">Thank you so much for your quick response. I
                will follow your suggestions.</div>
              <div class=""><br class="">
              </div>
              <div class="">Best wishes,</div>
              <div class=""><br class="">
              </div>
              <div class="">Vahid</div>
              <br class="">
              <div class="">
                <blockquote type="cite" class="">
                  <div class="">On Mar 16, 2017, at 8:35 PM, Dominik
                    Gresch <<a moz-do-not-send="true"
                      href="mailto:greschd@phys.ethz.ch" class="">greschd@phys.ethz.ch</a>>
                    wrote:</div>
                  <br class="Apple-interchange-newline">
                  <div class="">
                    <div bgcolor="#FFFFFF" text="#000000" class="">
                      <p class="">Dear Vahid,</p>
                      <p class="">You can use the 'nnkpts' parameter
                        (new in version 2.1) to explicitly specify for
                        which k-point pairs the overlap should be
                        calculated. This just side-steps the way that
                        Wannier90 tries to calculate the nearest
                        neighbours, and prints the neighbours you supply
                        directly to the .nnkp file. To get the .mmn
                        output for all pairs of k-points, you could
                        generate the nnkpts block such that it includes
                        all the overlaps you need.<br class="">
                      </p>
                      <p class="">For the interface to QE, you need to
                        use the 'regular_mesh = .false.' when using
                        pw2wannier90 to allow for non-regular k-grids to
                        be used (new in QE v.6).</p>
                      <p class="">Best regards,</p>
                      <p class="">Dominik Gresch</p>
                      <br class="">
                      <div class="moz-cite-prefix">On 17.03.2017 00:13,
                        Vahid Askarpour wrote:<br class="">
                      </div>
                      <blockquote
                        cite="mid:49C80929-A6AF-407E-86B1-3044FD69D7C0@dal.ca"
                        type="cite" class="">
                        <font class="" size="4">Dear Wannier Community,</font>
                        <div class=""><font class="" size="4"><br
                              class="">
                          </font></div>
                        <div class=""><font class="" size="4">I am
                            attempting to include impurity
                            scattering rate calculation in the EPW
                            calculation. At present, EPW determines the
                            scattering rates only due to electron-phonon
                            scattering.</font></div>
                        <div class=""><font class="" size="4"><br
                              class="">
                          </font></div>
                        <div class=""><font class="" size="4">To include
                            impurity scattering, one needs the overlap
                            intergal for the lattice-periodic part of
                            the wave function |u_nk> between any two
                            k-points. The .mmn files creates such a file
                            but it is only for the nearest neighbor
                            k-points. Before attempting to redo a
                            separate code, I have two questions:</font></div>
                        <div class=""><font class="" size="4"><br
                              class="">
                          </font></div>
                        <div class=""><font class="" size="4">1. Is it
                            possible to output the .mmn file for all
                            pairs of k-points?</font></div>
                        <div class=""><font class="" size="4"><br
                              class="">
                          </font></div>
                        <div class=""><font class="" size="4">2. Is it
                            possible to do 1 not for a k-grid but for a
                            list of random k-points? I suspect this can
                            be done using </font></div>
                        <div class=""><font class="" size="4"><br
                              class="">
                          </font></div>
                        <div class=""><font class="" size="4"> 
                            <div class="">begin kpoints</div>
                            <div class=""># of k points</div>
                            <div class="">…</div>
                            <div class="">…</div>
                            <div class="">end kpoints</div>
                            <div class=""><br class="">
                            </div>
                            <div class="">and the same k-points must be
                              used in the preliminary self-consistent
                              and non self-consistent calculations with
                              QE. </div>
                          </font></div>
                        <div class=""><font class="" size="4"><br
                              class="">
                          </font></div>
                        <div class=""><font class="" size="4">I ask
                            question 2 because for
                            thermoelectric properties, one needs to
                            select k-points with energies  ~0.5 eV above
                            the CBM. So it is possible to generate a
                            very dense k-grid (100^3) and only include
                            the selected k-points in the impurity
                            scattering rate calculation.</font></div>
                        <div class=""><font class="" size="4"><br
                              class="">
                          </font></div>
                        <div class=""><font class="" size="4">I would
                            appreciate your help with the above
                            questions.</font></div>
                        <div class=""><font class="" size="4"><br
                              class="">
                          </font></div>
                        <div class=""><font class="" size="4">Vahid</font></div>
                        <div class=""><font class="" size="4"><br
                              class="">
                          </font></div>
                        <div class=""><font class="" size="4">Vahid
                            Askarpour<br class="">
                            Department of Physics and Atmospheric
                            Science<br class="">
                            Dalhousie University,<br class="">
                            Halifax, NS, Canada</font></div>
                        <br class="">
                        <fieldset class="mimeAttachmentHeader"></fieldset>
                        <br class="">
                        <pre class="" wrap="">_______________________________________________
Wannier mailing list
<a moz-do-not-send="true" class="moz-txt-link-abbreviated" href="mailto:Wannier@quantum-espresso.org">Wannier@quantum-espresso.org</a>
<a moz-do-not-send="true" class="moz-txt-link-freetext" href="http://mailman.qe-forge.org/cgi-bin/mailman/listinfo/wannier">http://mailman.qe-forge.org/cgi-bin/mailman/listinfo/wannier</a>
</pre>
                      </blockquote>
                      <br class="">
                    </div>
                  </div>
                </blockquote>
              </div>
              <br class="">
            </div>
          </div>
        </blockquote>
      </div>
      <br class="">
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
Wannier mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Wannier@quantum-espresso.org">Wannier@quantum-espresso.org</a>
<a class="moz-txt-link-freetext" href="http://mailman.qe-forge.org/cgi-bin/mailman/listinfo/wannier">http://mailman.qe-forge.org/cgi-bin/mailman/listinfo/wannier</a>
</pre>
    </blockquote>
    <br>
  </body>
</html>