Re: [ESPResSo-devel] adress sample and code base

[Olaf Lenz]

Hi Pierre!

The ADResS implementation in ESPResSo is merely a proof-of-concept implementation. It does *not* give any performance advantage over a full simulation in high resolution, and it contains only the most basic ADResS functionality. If you want to use ADResS, you should probably go for ESPResso++. 

Have a look at 

  http://lists.nongnu.org/archive/html/espressomd-users/2013-11/msg00007.html

Otherwise, here you'll finde ES++

  http://www.espresso-pp.de/

We were already thinking whether we remove it from the ESPResSo code entirely.

Olaf

2013/11/21 Pierre de Buyl <address@hidden>

Dear Espresso developers,

I am interested in AdResS simulations (ideally of polymeric systems) and have
started to check the sample that comes with espresso.

The sample script is broken and I have started to put a missing piece back into
the system, the computation of the number of particles in ex,cg and hy
(exact/all atom, coarse grained and hybrid) in the system. To achieve that, I
have:
1. Added a global variable "int adress_regime[3]" to store the result.
2. Added a tcl routine "tclcommand_nregime".

The global may have been local instead but having a global means that one could
compute adress_regime within the force or integration loop instead (saving an
iteration).

I mimicked the routine "energy_kinetic" to write "nregime". To get the "status"
of a particle, I compare its weigth w to 0, 1 and 0<w<1 after having found that
in other parts of the code.

By the way, I tried to find a previous implementation of the routine via github,
google and "git grep n_regime $(git rev-list --all)" with no success.

If anyone knowledgeable with Espresso/AdResS would be willing to have a look and
comment on https://github.com/pdebuyl/espresso_work/commits/add_nregime I'd be
happy, if appropriate, to file a PR.

The branch add_nregime works with the scripts attached, Adress_t.tcl and
auxiliary.tcl. These are not clean enough to commit, however.

Regards,

Pierre

PS: a short intro is due on the list, I guess. I am a physicist working
(currently) at the dept of chemistry, KU Leuven. I have experience in writing
simulation code but polymer modeling is a recent topic. I have already discussed
with some developers of Espresso about http://nongnu.org/h5md/

--

Dr. rer. nat. Olaf Lenz

Institut für Computerphysik, Allmandring 3, D-70569 Stuttgart

Phone: +49-711-685-63607