[gmx-users] Re: How does fourierspacing correspond to fourier_nx, fourier_ny, and fourier_nz?

[Mark Abraham]

On 21/04/2012 1:35 AM, Andrew DeYoung wrote:
> Hi Florian and Mark,
>
> Thank you for your time!
>
> I am actually using (fourier_nx, fourier_ny, fourier_nz) ~ (14, 14, 125).
> This is what a colleague who has recommended, although she is using a
> different (somewhat slower) MD package.  One question, if you have time.  In
> that other code, one specifies the maximum number of wave vectors along each
> (positive or negative) direction.  So specifying (14, 14, 125) in that codes
> means "use 14 vectors along +x, use 14 vectors along -x; use 14 vectors
> along +y, use 14 vectors along -y; use 125 vectors along +z, use 125 vectors
> along -z".  So that corresponds to something like (28, 28, 250) or (29, 29,
> 251) actual grid points, I think.  But in the Gromacs manual
> (http://manual.gromacs.org/current/online/mdp_opt.html#ewald), I am not sure
> if fourier_nx, fourier_ny, and fourier_nz specify "the number of grid points
> along x, y, and z" or "the number of vectors to use along +x and -x; +y and
> -y; and +z and -z".  I think that Gromacs uses the former, but I am not
> sure.

The former for PME. The latter for plain Ewald (which you should 
probably not want to use).

Mark