cp2k9 under cygwin: bomd at graphite: slow?

[Victor Volkov]

Good day/good evening everybody:

recently I compiled cp2k version 9 under Cygwin,
and I am using it with BOMD to try predicting possible

favourable orientations of small aromatic molecules next to graphite.

I attach here one of trial examples, where a small aromatic molecule
is overwhelmed by the original slightly unfavourable initial geometry

that it flies away rather than it would stay next to the surface.

I think, this is reasonable - the interactions with graphate are not strong,

(I included dftd3)

and a better care may be taken for the initial geometry.

However, what I have noticed that for the size of the system,

to have 1-2 steps per day,

the simulation seems to be quite slow.

On one hand, this can be due to the compiled version under Cygwin - I doubt.

On the other hand, because criteria and reasons for setting "environment"
(type of ensemble, thermostat, barostat, Poisson periodicity, stress tensor etc)
are not described for such simulations well anywhere,
there is a bit of intuitive guessing.

If you would find a moment to express your opinion about the input
to point weak spots if you fine, I would be thankful.

Also,  ... since I present the files

perhaps, your comments could be didactic-helpful for other.

Best for you,

Victor