Re: [h5md-user] fix remaining imprecisions - particle position

[Pierre de Buyl]

On Fri, Jan 10, 2014 at 03:36:41PM +0100, Olaf Lenz wrote:
> Lets adapt the proposal from my first mailing:
> 
> For those components where `boundary` is set to "none", any value of
> `edges` or `image` must be silently ignored.

Any value of `image` then. Edges only play a role if 'image' is different from
zero.

Your proposal
http://thread.gmane.org/gmane.science.simulation.h5md.user/500/focus=516
is fine to me, feel free to make the change :-)

P


> 2014/1/10 Pierre de Buyl <address@hidden>
> 
> > On Fri, Jan 10, 2014 at 03:18:42PM +0100, Felix Höfling wrote:
> > > Am 10.01.2014, 15:01 Uhr, schrieb Pierre de Buyl
> > > <address@hidden>:
> > >
> > > >read position
> > > >if coordinate i is of type 'periodic' and 'image' is present
> > > >then
> > > >  compute position_i += image_i*edges_i
> > > >else
> > > >  position is fine the way it is
> > > >end if
> > > >
> > > >The modification would be to remove "Otherwise, the data indicate
> > > >absolute
> > > >positions in space." Then, the following cases are ok:
> > > >- periodic position, with no 'image', such as the Monte Carlo
> > > >mentioned above.
> > > >- periodic positions, with 'image'
> > > >- absolute position: as long as there is no 'image' element, you
> > > >can store
> > > >  absolute positions even if the boundary is periodic.
> > > >
> > > >Olaf, does that reply to your query about the MSD?
> > > >
> > >
> > > Basically, what your scheme above states is "Otherwise, the position data
> > > are fine the way they are." How shall we phrase it in a concise and less
> > > familiar way?
> > >
> > > To me, "absolute position" means that the data do not need further
> > > adjustment, which is precisely what you're saying. I wouldn't drop the
> > > sentence to also specify what is meant if image is absent.
> >
> > Ok, it does look equivalent actually :-)
> >
> > Then, maybe one would need to clarify that 'image' is only to be used in
> > those
> > directions that specify 'periodic' for the boundary? (i.e. a wall in z but
> > PBC
> > in x and y)? 'image' would be zero any in direction z then so the
> > interpretation
> > of the file does not change.
> >
> >
> >
> >
> >
> 
> 
> -- 
> --
> Dr. rer. nat. Olaf Lenz
> Institut für Computerphysik, Allmandring 3, D-70569 Stuttgart
> Phone: +49-711-685-63607