<html>
<head>
</head>
<body>
Dear Luciano,<br>
<br>
When a message appears during an attempt to minimize a system using SD:<br>
<pre wrap="">Stepsize too small... Converged to machine precision..."<br><br>it essentially means that the SD algorithm has problems and that it is time to switch to the CG.<br><br>It may also be helpful to delete the SOL molecules that experience excessive forces during both SD and CG and reminimize. Monitor forces in the standard (screen) output during EM.<br><br>You may also find that EM performs better with double precision fftw libraries. To recompile gmx with double precision: <br><br>make clean<br>./configure --disable-float<br>make<br>make install<br><br>For the fftw with double precision it is just:<br><br>./configure<br>etc.<br><br>Finally, it may be necessary to submit the system to multiple iterations of SD and CG, gradually decreasing emtol, followed by deletion of unwanted atoms with large forces after each iteration.<br><br>Regards,<br>Martina Bertsch, Ph.D.
Northwestern University
Feinberg School of Medicine
Molecular Pharmacology and Biol
ogical Chemistry
303 East Chicago Avenue
Chicago, IL 60611<br></pre>
<br>
<br>
Luciano wrote:<br>
<blockquote type="cite" cite="mid:1ed1b81ed09c.1ed09c1ed1b8@universiabrasil.net">
<pre wrap="">Hi gmx-users and specially David and Anton very very thanks for advise.<br><br>I have been read the last mails about EM (SD and CG) and doing EM to my<br>system (<br>protein + water) without pbc it have showed to emtol = 10000:<br><br>Conjugate Gradients:<br> Tolerance = 1.00000e+04<br> Number of steps = 1000<br> F-Norm = 1.21548e+05<br><br>Step 1, E-Pot = 7.2018726562e+04, Fmax = 8.01764e+03, atom = 1155<br><br>writing lowest energy coordinates.<br><br>Back Off! I just backed up octa20orig.trr to ./#octa20orig.trr.15#<br><br>Back Off! I just backed up after_em_pr.gro to ./#after_em_pr.gro.15#<br><br><br>But when I set emtol = 7000 to next step the result was:<br><br>Conjugate Gradients converged to 10000 in 2 steps<br> Potential Energy = 7.20187e+04<br>Maximum force: 8.01764e+03<br><br><br>Negative w: -5.389869190490e+21<br>z= -2.909245276160e+11<br>gpa= 3.167263553258e+11, gpb= 2.842420549736e+11<br>a= 0.0000
00000000e+00, b= 1.130758611745e-10<br>EpotA= 7.016328906250e+04, EpotB= 7.019690625000e+04<br>Negative number for sqrt encountered (-5389869190490223542272.000000)<br>Terminating minimization<br><br>writing lowest energy coordinates.<br><br>Back Off! I just backed up octa20orig.trr to ./#octa20orig.trr.16#<br><br>Back Off! I just backed up after_em_pr.gro to ./#after_em_pr.gro.16#<br><br>Conjugate Gradients did not converge in 3 steps<br> Potential Energy = 7.01633e+04<br>Maximum force: 3.67083e+05<br>What are the terms z, gpa, gpb, a and b? Why the negative number for sqrt?!<br>What I should do?! About the message of Martina Bert, Sun, 16 Nov 2003<br>01:53:35 -0600, what the matter when the message comes up?<br><br>"Stepsize too small (... nm) Converged to machine precision"<br><br>Thanks in advance...<br><br>Luciano<br><br><br>######################################################<br> Luciano Tavares da Costa<br> Mestre em Engenharia Qu&
iacute;mica<br> icq #232852036 ; <a class="moz-txt-link-abbreviated" href="mailto:ltcosta@universiabrasil.net">ltcosta@universiabrasil.net</a> <br> Doutorando Laboratório de Biofisicoquímica<br> Instituto de Química - USP<br> <br>######################################################<br><br><br>------------------------------------------------------------------------<br>Vote no Universia para o Prêmio IBest 2004 e concorra a uma<br>viagem para Trancoso.<br><a class="moz-txt-link-abbreviated" href="http://www.universiabrasil.net">www.universiabrasil.net</a><br><br></pre>
<center>
<table border="1">
<tbody>
<tr>
<td>
<div align="Right" class="headerdisplayname" style="display: inline; ">
Part 1.1</div>
</td>
<td>
<table border="0">
<tbody>
<tr>
<td>
<div align="Right" class="headerdisplayname" style="display: inline; ">
Content-Type:</div>
</td>
<td>text/plain</td>
</tr>
<tr>
<td>
<div align="Right" class="headerdisplayname" style="display: inline; ">
Content-Encoding:</div>
</td>
<td>quoted-printable</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
</center>
<br>
<br>
</blockquote>
<br>
</body>
</html>