[CP2K:1312] Re: multiple force_eva run with different colvar restraints

[Laino Teodoro]

On 27 Aug 2008, at 18:47, Jun Cheng wrote:

> I have a better understanding now. Thanks.
> As I asked in the last post, if GHOST T and BASIS_SET NONE are   
> given, the atom will be a true ghost. This means POTENTIAL should  
> be no longer needed. But the job won't get running unless  
> POTENTIAL .... is given. How come that happens?

Well.. this is kind of "buglet".. I should clean it better.. anyway  
does not matter..  the pseudopotential is not used if you specify GHOST.
I will change it in the future in such a way to have the possibility  
to specify "NONE"..

> Could you please explain in a little more detail how the mapping  
> and free energy module (ALCHEMICAL CHANGE in my case, I think) work  
> in cp2k? Or give some references I can read. I noticed that there  
> are some input examples in /cp2k/tests/FE directory, but I am  
> afraid that I can't understand them.

The alchemical change module is still in a kind of experimental  
version and a publication will come in the future.
If you want to use it in the meanwhile, as in the vast majority of  
open source codes.. well.. put your hands on it!
the source is there.. you can read it.. study it.. check it.. ;-)

Teo

> Thanks.
>
> Jun
>
>
> On Wed, Aug 27, 2008 at 4:45 PM, Laino Teodoro  
> <teodor... at gmail.com> wrote:
> Yes unfortunately there's a misunderstanding.
> A GHOST atom is not what you think of.
> A ghost for CP2K is an atom with no nuclear charge but with a basis  
> set (which you can specify). This kind of GHOST atoms
> are used mostly for BSSE: read carefully the online reference  
> input: http://cp2k.berlios.de/input/ 
> InputReference~__ROOT__~FORCE_EVAL~SUBSYS~KIND.html#GHOST
>
> Of course you can put BASIS_SET NONE and in this case what you have  
> is essentially NOTHING (i.e. no electrons, no basis set and no  
> pseudo).
>
> What you want to do can be done easily or using BASIS_SET NONE (and  
> specifying it GHOST, see above) or just avoid the mapping of that atom
> in the free energy module. Both methods (if properly done) will  
> lead to the same numbers.
>
> Teo
>
> On 27 Aug 2008, at 17:37, Jun Cheng wrote:
>
>> POTENTIAL is needed? Or I misunderstand something?
>
>
>
>
>
> >

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20080827/6e64c8dc/attachment.htm>