[gmx-users] "trjconv -pbc whole" and SPC water

[Erik Lindahl]

Berk Hess wrote:

> Water molecules with only settles and no bonds are not made whole.
> Mainly for efficiency reasons. But trajectories from mdrun always
> contain whole molecules.
> How did you end up with broken waters? trjconv -pbc inbox?
>
> Berk.

Hi,

Since this doesn't affect the simulation performance (right?) we should
definitely make ALL molecules whole even if it takes slightly longer, at least
if you explicitly select "-pbc whole".

If there's a need for a faster version that disregards the waters we can of
course introduce that too, but the tools should do what they say :-)

Cheers,

Erik