<div dir="ltr"><div>Hi,</div><div><br></div>I know that this one is a bit too old, but I have a similar doubt.<br><br>I think he argues about the possibility of high r_cutoff values to violate the minimum-image convention. Is that simply not true?<br><br>And if it does violate, have anyone some answer about what would be worse: to violate the mic or to not compute the dispersion correction in an appropriate range?<br><br>Regards,<br><br>Em terça-feira, 17 de dezembro de 2013 09:35:30 UTC-2, JanLos escreveu:<blockquote class="gmail_quote" style="margin: 0;margin-left: 0.8ex;border-left: 1px #ccc solid;padding-left: 1ex;">
<div bgcolor="#FFFFFF" text="#000000">
Dear CP2K developpers,<br>
I was wondering about the comment in the section <br>
"FORCE_EVAL/DFT/XC/VDW_<wbr>POTENTIAL/PAIR_POTENTIAL/R_<wbr>CUTOFF",<br>
saying "<em>The cutoff will be 2 times this value". </em><br>
If I stick to the default value of 10.06 Angstrom for this
parameter, <br>
the cut-off will be 20.12 Angstrom, which is pretty large, but OK. <br>
If this is really what is applied, then it would imply that for a
cubic simulation <br>
box of 20 A cube with periodic boundary conditions, R_CUTOFF should
not <br>
be set larger than 5.0 Angstrom in order to NOT exceed recommended <br>
maximal value for this box size, namely half the box size. Do I see
this correctly ?<br>
It's just that the comment " <em>The cutoff will be 2 times this
value</em>" confuses me, <br>
asking myself why this input parameter been defined in this way.<br>
By the way, is the minimal image convention applied for the
calculation<br>
of these interactions ?<br>
<wbr> <wbr>
Regards, Jan Los<br>
<br>
</div>
</blockquote></div>