Dear Jonathan,<div><br></div><div>thank you for the reply.</div><div><br></div><div>Using:</div><div><br></div><div><div>guiding_centres = T</div><div><br></div><div>the situation improves:</div><div><br></div><div><div> Cycle: 1000</div>
<div> LINE --> Iteration : 1000</div><div> LINE --> Spread at initial point : 87.6286372239062</div><div> LINE --> Spread at trial step : 87.6286372239063</div><div>
LINE --> Slope along search direction : -1.541916642988277E-017</div><div> LINE --> ||SD gradient||^2 : 1.544096354738086E-015</div><div> LINE --> Trial step length : 2.00000000000000</div>
<div> LINE --> Optimal parabolic step length : 4.338226673863846E-004</div><div> LINE --> Spread at predicted minimum : 87.6286372239062</div><div> LINE --> CG coefficient : 0.999995534026802</div>
<div> WF centre and spread 1 ( 8.776812, 0.465560, 7.738687 ) 58.36280050 <<<-----------</div><div> WF centre and spread 2 ( 5.485742, 9.409140, 8.702739 ) 9.81781675</div><div> WF centre and spread 3 ( 7.665904, 4.358372, 8.556469 ) 9.82773595</div>
<div> WF centre and spread 4 ( 3.155283, 4.303197, 8.509082 ) 9.62028402</div><div> Sum of centres and spreads ( 25.083740, 18.536269, 33.506978 ) 87.62863722</div><div><br></div><div> 1000 0.000E+00 0.0000000393 87.6286372239 35.11 <-- CONV</div>
<div> O_D= 48.5782103 O_OD= 0.0676344 O_TOT= 87.6286372 <-- SPRD</div><div> Delta: O_D= 0.0000000E+00 O_OD= -0.1554312E-14 O_TOT= 0.0000000E+00 <-- DLTA</div><div><br></div></div><div>the center of the first WF is now at the correct place (center of one of the molecular sites) but still the spread is too big. :-(</div>
<div><br></div><div>cheers,</div><div><br></div><div>Gianluca</div><br><div class="gmail_quote">On Mon, Dec 5, 2011 at 1:01 AM, Jonathan Yates <span dir="ltr"><<a href="mailto:jonathan.yates@materials.ox.ac.uk">jonathan.yates@materials.ox.ac.uk</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"><br>
On 3 Dec 2011, at 14:39, Gianluca Giovannetti wrote:<br>
<br>
> Dear All,<br>
<div class="im">><br>
><br>
> However in the wannier90.wout the spread of one of the Wannier functions is very large (----> 152.37216676):<br>
><br>
> Cycle: 1000<br>
</div><div class="im">> WF centre and spread 1 ( 6.466896, 3.625330, 2.430574 ) 152.37216676<br>
> WF centre and spread 2 ( 5.485734, 9.409137, 8.702740 ) 9.81928932<br>
> WF centre and spread 3 ( 7.666070, 4.358506, 8.556600 ) 9.83226929<br>
> WF centre and spread 4 ( 3.155228, 4.303197, 8.509269 ) 9.62144690<br>
> Sum of centres and spreads ( 22.773929, 21.696169, 28.199184 ) <a href="tel:181.64517227" value="+18164517227">181.64517227</a><br>
><br>
</div><div class="im">> 2) How can i get a small SPREAD?<br>
<br>
</div>Gianluca,<br>
<br>
Try re-running the calculation with<br>
guiding_centres = T<br>
<br>
I think you have a problem with the code finding the wrong branch cut for the complex log in the evaluation of Omega_D (ie your total Omega_D is huge, but Omega_OD is sensible). The guiding centres option should help take the correct branch cut.<br>
<br>
<br>
Jonathan<br>
<font color="#888888"><br>
<br>
<br>
<br>
--<br>
Department of Materials, University of Oxford, Parks Road, Oxford, OX1 3PH, UK<br>
tel: <a href="tel:%2B44%20%280%291865%20612797" value="+441865612797">+44 (0)1865 612797</a> <a href="http://users.ox.ac.uk/~oums0549/" target="_blank">http://users.ox.ac.uk/~oums0549/</a><br>
<br>
_______________________________________________<br>
Wannier mailing list<br>
<a href="mailto:Wannier@quantum-espresso.org">Wannier@quantum-espresso.org</a><br>
<a href="http://www.democritos.it/mailman/listinfo/wannier" target="_blank">http://www.democritos.it/mailman/listinfo/wannier</a><br>
</font></blockquote></div><br></div>