Re: Unwrapped coordinates in H5MD format

[Jean-Noël Grad]

Dear Xin Yuan,

Thank you for bringing our attention to this matter. The user guide 
currently doesn't explain how to recover the unfolded coordinates; I 
will try to fix that next week when going over the backlog of issues in 
the user guide.

To answer your question, the unfolded coordinates can be recovered with 
the formula `pos_unfolded = particle_pos + particle_image * box_length`. 
The image count should be zero for aperiodic systems. It is a 
requirement from the H5MD standard, which mandates that data written 
under 'particles/atoms/position/value' must be folded coordinates, with 
'particles/atoms/image/value' containing the image count (section 
'position' and 'image' in 
https://www.nongnu.org/h5md/h5md.html#particles-group).

Please find attached a MWE that applies this formula to a single frame 
of a trajectory.

Best,
JN

On 3/17/22 18:47, Xin Yuan wrote:
> Dear ESPResSo users,
>
>
> I noticed that the coordinates recorded in H5MD format are wrapped 
> inside the
> simulation box by default. Is there an option to record the unwrapped
> coordinates in H5MD instead? I didn’t find any relevant information in 
> the
> documentation. Thanks for any help!
>
> Regards,
>
> Xin Yuan

h5md.py
Description: Text document