[gmx-users] minimum image convention

[pratibha kapoor]

Hi,

I believe that subsequent analysis after simulation can be carried out on
any xtc file created using trjconv -pbc nojump or -pbc no or -pbc whole,
except for some cases like g_rms or g_msd. Please correct me if I am wrong.
Now, if I want to write my own code for cluster analysis can I again make
use of any xtc file (say that created with no pbc option)? I understand
that only thing I would have to incorporate is minimum image convention
condition  which I am not able to understand. Say for difference in x
coordinates of two atoms (dx) in a box of length l we should correct:
           dx=dx-l*anint(dx/l)
For me, dx=dx-l is understandable but multiplication with anint(dx/l) part
is not clear.
Please help me understand this.