[CP2K-user] Why might CP2K lead to a very different static relative permittivity value compared to another code?

[Ant]

I am quite new to molecular dynamics, so apologies in advance if this turns 
out to be a silly or poorly formulated question.

As part of an effort to become more familiar with CP2K/Quickstep, I have 
tried to reproduce a predecessor's NPT simulations based on the Qbox code (
http://eslab.ucdavis.edu/software/qbox/) to calculate the static 
permittivity of water.  I get the same results at lower pressure, but not 
as pressure increases and would be interested in whether anyone can point 
me to any information on why this might be.  

I can't provide any more specifics, but can say that I'm using the same 
time step, functional, kinetic energy cutoff, and periodic boundary 
conditions as my predecessor.  I have tried varying the cutoff and my 
results don't change much.  He used twice the number of molecules that I'm 
using and he used "heavy" water.  I'm not clear on the thermostat he used, 
but my temperature values average around my input value.  I don't have any 
other information from him.  Maybe there's another constraint/issue that 
hasn't occurred to me that might lead to differences? 

Thank you for any guidance.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20190831/c1854e3c/attachment.htm>