[CP2K:1307] Re: multiple force_eva run with different colvar restraints

[Jun Cheng]

It is brilliant! well done.

I just realized another question about my calculations.
In the calculations of free energy change one atom is gradually mutating to
null (i.e. ghost), so in the force_eva section of the final state, GHOST is
turned on in the KIND section of that atom. If I understand the input
reference correctly
(%FORCE_EVAL<http://cp2k.berlios.de/input/InputReference%7E__ROOT__%7EFORCE_EVAL.html>
%SUBSYS<http://cp2k.berlios.de/input/InputReference%7E__ROOT__%7EFORCE_EVAL%7ESUBSYS.html>
%KIND<http://cp2k.berlios.de/input/InputReference%7E__ROOT__%7EFORCE_EVAL%7ESUBSYS%7EKIND.html>%GHOST),
*BASIS_SET NONE *should be set as well. Is that correct? I already did some
runnings without *BASIS_SET NONE *because of carelessness. After I noticed,
I did some testings and found that the potential energies with and
without *BASIS_SET
NONE *are slightly different. My question is whether the presence of *BASIS_SET
NONE* matters very much.
By the way, I also found that POTENTIAL is compulsory when GHOST T is turned
on, otherwise the job cannot get running. Since the atom is a ghost,
POTENTIAL is meaningless. Why POTENTIAL is needed? Or I misunderstand
something?
Many thanks.

Cheers, Jun

On Wed, Aug 27, 2008 at 4:07 PM, Laino Teodoro <teodor... at gmail.com>wrote:

> well  Axel is the guy to thank ;-)  he did most of the nice/useful things
> with the pre-processor ..Teo
>
> On 27 Aug 2008, at 17:02, Jun Cheng wrote:
>
> Thanks for the tips.
>
>
>
> >
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20080827/28c6cc9c/attachment.htm>