>> I prepared a very large system using SIRAH; when I try to run a
>> minimization, pmemd stops after reading the input and complains:
>> ! ERROR: nfft1 must be in the range of 6 to 512!
>> ! ERROR: nfft2 must be in the range of 6 to 512!
>> ! ERROR: nfft3 must be in the range of 6 to 512!
>>
>> In an earlier post it was suggested to change the parameter 'gridhi' in
>> mdin_ewald_dat.F90 to a higher value (e.g. from 512 to 1024) and
>> recompile.
>>
>> My question is:
>> What is a reasonable value for gridhi?
>>
>> I assume (from AmberManual section 20.7.2) that it should be >= the box
>> dimension for all systems I want to simulate, is this correct?
>
> For all atom simulations, we generally end up with grid spacings of about 1
> Ang. (This is determined by PME error estimates.) Hence, gridhi should be
> big enough to accommodate the box size (expressed in Ang.).
>
> The SIRAH people might have experience about whether one can have larger
> grid spacings (and hence smaller grid dimensions) for coarse-grained
> simulations. (Maybe I should know this, but I don't.)
>
> .....dac
thanks for the answer!
So I'll try out an see what happens.
Best regards,
Anselm
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Apr 21 2023 - 09:00:02 PDT
