[gmx-developers] Velocities to start leapfrog

[David van der Spoel]

On 1/17/10 11:28 PM, Pablo Campomanes wrote:
> Dear gmx-developers,
>
> After having a look to gromacs source code for quite a long time, I am
> still wondering in where initial velocities generated using the grompp
> utility (I assume v(t0)) are shifted to v(t0 - dt/2) in order to start
> the leap-frog algorithm in "update.c"
>
> Could anybody shed some light in this issue?


src/gmxlib/random.c, routine:
void maxwell_speed(real tempi,int seed,gmx_mtop_t *mtop, rvec v[])

>
> Thanks in advance for your help,
>
> Pablo
>
> _____________________________________________
> Pablo Campomanes
> Laboratory of Computational Chemistry and Biochemistry
> BCH 4118 EPF Lausanne
>


-- 
David van der Spoel, Ph.D., Professor of Biology
Dept. of Cell & Molec. Biol., Uppsala University.
Box 596, 75124 Uppsala, Sweden. Phone:	+46184714205. Fax: +4618511755.
spoel at xray.bmc.uu.se	spoel at gromacs.org   http://folding.bmc.uu.se