[gmx-users] Basic Query

[Mark Abraham]

priyanka srivastava wrote:

> I have used PME and have mentioned rcoulomb in the
> .mdp file too, which as is pointed out by you, has no
> meaning!! But does this mean that my mdp file is
> wrong? Because if PME does not use rcoulomb it shud
> simply ignore the value.

As you'd see by reading sections 4.9 and 7.3.9 of the manual, Martin's 
statement isn't right. The grid dimensions are controlled by the two 
parameters he mentioned, but there are plenty of other parameters that 
influence PME.

You're right, in that if mdp file parameters will be ignored then the 
value you use isn't relevant.

> Actually I do not want to use twin range cutoff. So, I
> have given rvdw = rlist. Is this alright?

Yes, see section 4.6.3. This is only relevant for vdw since coulombtype 
= PME

> Portion of my mdp file looks like this:
> 
> ; Electrostatics
> coulombtype             =  PME
> fourier_nx              =  6.4
> fourier_ny              =  5.0
> fourier_nz              =  8.1
> pme_order               =  4
> rcoulomb                =  1.5
> rvdw                    =  1.5

Assuming you got these values from somewhere sensible and that they're 
consistent with the force field you're using, then this will give you 
some numbers that have at least as good a chance of corresponding to 
reality as anything :-)

> Also since I have not specified "vdwtype" in the mdp
> file, so it automatically takes Cut-off.

Set it specifically, so that you don't have to remember what the default 
is in a year's time.

Mark