Re: [h5md-user] Offset

[Pierre de Buyl]

On Mon, Sep 23, 2013 at 03:27:01PM +0200, Felix Höfling wrote:
> Am 11.09.2013, 13:13 Uhr, schrieb Konrad Hinsen
> <address@hidden>:
> 
> >Pierre de Buyl writes:
> >
> > > Some people work with [-L/2:L/2], others with [0:L].
> > >
> > > offset is there so that users may dump their position data without
> > > operation between memory and file, whatever box convention they use.
> >
> >Do I conclude then that "offset" is supposed to be -L/2 in the first
> >case, and 0 in the second?
> >
> >
> >Olaf Lenz writes:
> >
> > > I was always wondering about the offset, but I didn't want to
> >start that
> > > discussion. Thanks, Konrad. :-/
> > >
> > > The offset is mathematically irrelevant.
> >
> >I agree.
> >
> > > I also do not see what the offset is needed for in the file.
> >
> >I'd look at this from a different point of view: if it's there, what
> >information does it convey? What conclusion can a trajectory reader
> >draw from it?
> >
> > > The only case that I can think of where the offset might be useful is,
> > > when it is somehow guaranteed that the positions are all within the
> > > primary image that is defined by the offset and the edges.
> >
> >Exactly.
> >
> > > Therefore, from my point of view, the offset is useless and can be
> > > removed from the specs.
> >
> >Or add a guarantee about the positions being in the primary image to
> >the specs. From my experience, that can be useful information to have.
> >
> >Konrad.
> 
> The purpose of the offset was precisely this guarantee, although it
> got lost in the course of time. Originally, we had minimum and
> maximum coordinates for the particle positions which where then
> replaced by the offset (plus edge lengths).
> 
> The current draft says that particle positions are stored "as
> periodically wrapped or unwrapped coordinate vectors". What decides
> whether they are wrapped or not? (BTW, if the word "folded" is more
> common the spec should use it.) If the positions are wrapped, then
> it is easy to add the guarantee that they are within the primary
> box.

Even if there is no guarantee of folding, offset still serves a purpose to
define the box. At the very least, as a convenience for checkpointing.

P