[gmx-users] Re: gromacs energies very different from other MD engine for the very same system and condition

[chris.neale at utoronto.ca]

Alan: you should try Ilja's suggestion, it was a good one. The idea is  
that during pdb2gmx you may have obtained a .gro file that was rounded  
and then you may have used this .gro as input to grompp and so your  
run did not start with the exact same coordinates in gromacs and namd.  
A similar test would be to be sure that you run grompp with the same  
.pdb file that you used in namd, etc. (or perhaps you already did that?)

Actually, from what you post below I think that you have your answer.  
Use the orig .pdb into grompp and things should be ok, probably even  
in single precision, although the output .gro will be rounded as you  
see.

PS: let's keep a positive mood.

Chris.

On Tue, Jul 27, 2010 at 16:36, <gmx-users-request at gromacs.org> wrote:

> I think the bond terms differ simply due to the round off error. The gro
> file is worse than pdb when it comes to round off in Cartesian coordinates.
> You can confirm this is the case by taking your gro file and converting it
> back to pdb using gromacs tools and then use the resulting pdb (with the
> same round of as the gro file now) file from gromacs in NAMD. You should
> see
> the same numbers for bonded terms then.
>

I doubt that would do any difference.

Anyway, I tested with double precision and found out that my gro file is
still rounded off like in single precision, i.e., my gro in double is the
same in single and I was not expecting that.

Example:

pdb: 13  N   ALA A   2       1.927   1.789   1.165  1.00  0.00

gro (single or double): 2ALA      N   13   0.193   0.179   0.117


I was expecting gro when using pdb2gmx_d to be like:

gro (single or double): 2ALA      N   13   0.192700   0.178900   0.165000

Hummm.... I don't like this.


Alan
--