[gmx-developers] WARNING: Found more than 12 triclinic correction vectors, ignoring some.

[Berk Hess]

Hi,

In most cases there warnings are harmless (but annoying).
If you results look ok, it is fine.
It has been fixed in CVS.
If you want a fix, you can try to implement the diff between
revision 1.79 and 1.80 of src/gmxlib/pbc.c

Berk

Justin A. Lemkul wrote:
>
> Hi all,
>
> I am simulating a simple protein/ligand in water using an octahedral 
> box.  For visualization purposes, I am rendering the box in its true 
> octahedral shape using the following command:
>
> trjconv -s md.tpr -f 20ns.xtc -o compact.xtc -ur compact -pbc mol
>
> When I do this, I receive a series of warnings like the following:
>
> WARNING: Found more than 12 triclinic correction vectors, ignoring some.
>   There is probably something wrong with your box.
>          Box (3x3):
>             Box[    0]={ 6.87729e+00,  0.00000e+00,  0.00000e+00}
>             Box[    1]={ 3.23209e+00,  6.46199e+00,  0.00000e+00}
>             Box[    2]={-2.97018e-01,  1.10216e-03,  5.59953e+00}
>
> I saw a previous discussion across the list here:
>
> http://www.gromacs.org/pipermail/gmx-developers/2006-November/001953.html
>
> The result was a request for a bugzilla and a potential fix.  I am 
> using GMX 3.3.3, and the warnings only seem to come up when I use -pbc 
> mol.  Other PBC corrections (like nojump) don't seem to result in 
> these warnings.
>
> When I visualize the trajectory, the box seems to smoothly morph 
> between an octahedron and a rectangular shape.  I don't know if that's 
> relevant or just a consequence of visualization.
>
> Any ideas as to what's going on?  Is there a problem with this type of 
> manipulation?
>
> Thanks in advance for any comments.
>
> -Justin
>