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

[Olaf Lenz]

Hi Konrad!

2014/1/10 Konrad Hinsen <address@hidden>

My understanding of the original text is that there is a distinction

between the position in the primary box and some "absolute"
position. My point is that this distinction need not exist.

​Yes, there is a distinction, and I think it needs to exist: the absolute position is the unfolded position, which might lie far outside of the primary box. The reason to distinguish between the folded and unfolded position is that if you just have the folded position, you cannot track a particle easily, e.g. to measure the MSD.

In a first approximation, one could assume that this case is handled by the "image", however, some simulation programs do not provide the image. And even if the image is used, the particle "position" field might still lie outside the primary box, e.g. when using a skin. And that *does* make a difference.

1) Having both "minimum" and "maximum" is redundant, as the difference

   must be "edge".

​No, it is not​! "edge" defines the size of the boundaries, it is a physical property of the system.

"minimum" and "maximum" define the bounding box of "position". It has no physical meaning, but is a hint for the person reading the file. However, I'm not really convinced that it is actually very useful: even when it is given, a reader should still always check whether the positions really are within the bounding box, otherwise he might get into trouble with badly crafted files. Furthermore, we would have to define whether the minimum and maximum refer to the "position" field, or the absolute position.

So, is there anybody who knows of a concrete case where one could profit from the fields minimum and maximum?

2) Any such specification should be optional, for trajectories that
   don't make any promises.

​Yes!​

Olaf

--

Dr. rer. nat. Olaf Lenz

Institut für Computerphysik, Allmandring 3, D-70569 Stuttgart

Phone: +49-711-685-63607