[gmx-developers] Calc_pot: usage and accuracy ?'s

[David]

On Sat, 2003-03-22 at 01:52, DaJustice1 at aol.com wrote:
> Dear Developers:
> 
> I am continuing to hack away at the calc_pot code.   There is very little 
> documentation of the code, so I am a little confused on the usage.   Up to 
> this point I have been just experimenting with it.   I have opened a .tpr 
> with the efTPX enum, and init_calc_pot accepted it, and initialized the 
> proper structures.   I then called calc_pot which gave a pot. energy that was 
> twice the pot. energy for the same frame of a simulation (using g_energy).   
> Is calc_pot accurate?   Am I using it incorrectly?   The bottom line is; I 
> need to take two separate sets of charges and one traj., and for the two sets 
> of charges find the pot. energy for each set of coordinates of the traj.   I 
> am open to suggestions and I love to code, so please take a moment to help... 
>   Thank you.

For your purpose you want to run a trajectory with the first set of
charges, then modify the tpr file and use the mdrun -rerun option to
calculate energies from all the coordinates in  the original trajectory,
but with your new tpr file (i.e. new charges etc.).

As to calc_pot, that was a bit of a hack, because... (blush, shame) it
was to complicated to call  our own energy calculation routines... This
will hopefully change in the next release. If your energies are
*exactly* a factor of two higher, that suggests they are computed
twice...


> 
> ~Justice~
-- 
Groeten, David.
________________________________________________________________________
Dr. David van der Spoel, 	Dept. of Cell and Molecular Biology
Husargatan 3, Box 596,  	75124 Uppsala, Sweden
phone:	46 18 471 4205		fax: 46 18 511 755
spoel at xray.bmc.uu.se	spoel at gromacs.org   http://xray.bmc.uu.se/~spoel
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++