<div>Hi,</div>
<div> </div>
<div>I have been trying to follow the tutorial <a href="http://www.dillgroup.ucsf.edu/group/wiki/index.php/Free_Energy:_Tutorial">http://www.dillgroup.ucsf.edu/group/wiki/index.php/Free_Energy:_Tutorial</a> by David Mobley to reproduce the solvation energy for methane in water. I followed the instructions and got the solvation energy in good agreement with the answer in the tutorial. But I noticed that the fluctuation for every dgdl is very large. The numbers are attatched below. I have read the thread in the mailling list about the error thing: <a href="http://www.gromacs.org/pipermail/gmx-users/2007-May/027333.html">http://www.gromacs.org/pipermail/gmx-users/2007-May/027333.html</a> and <a href="http://www.gromacs.org/pipermail/gmx-users/2007-May/027237.html">http://www.gromacs.org/pipermail/gmx-users/2007-May/027237.html</a>, but still I am very confused about the error and fluctuation. If the fluctuation is so large, how well can we trust the solvation energy we got? Thanks a lot in advance for your comment.</div>
<div> </div>
<div>lambda dgdl fluctuation</div>
<div>0.0 4.713 11.426</div>
<div>0.1 2.209 13.584</div>
<div>0.2 0.534 14.920</div>
<div>0.3 -1.333 16.274</div>
<div>0.4 -3.944 18.264</div>
<div>0.5 -7.992 21.191</div>
<div>0.6 -17.773 27.100</div>
<div>0.7 -28.915 25.747</div>
<div>0.8 -24.6079 14.5232</div>
<div>0.9 -11.135 7.6318</div>
<div>1.0 -0.0948 4.337</div>
<div> </div>
<div>Jianhui</div>
<div>Biocomputation and bioinformatics </div>
<div>Department of Physics</div>
<div>Rensselaer Polytechnic Institute</div>