[gmx-users] minimum periodic distance

[Wang Zhun]

Hi, all,
In a previous post
http://www.gromacs.org/pipermail/gmx-users/2004-January/008561.html
"use g_mindist -pi to chech the 'minimum periodic distance'  to its periodic images during the simulation. That should be at least 1 nm, but for PME maybe one would want that larger."
I wonder how 1nm is set. Is this value based on rvdw, rlist, etc? If  the minimum periodic distance is far less than 1nm, what does it imply? Is a  larger box needed to start MD?
Thanks.
BTW, in my former post there is a typo: editconf command should generate a "tric" box, not a "cubic" one, if box type is not set.