Hi Danila,

René was a bit quicker then me due to a conference I am on. Thank you :)

Please take care about the order in the legacy "hidden" attribute
"_global_start" in 2D3V simulations - it's still a 3D attribute and one
entry inside is just useless. Also its order is confusing, see this
comment in S. Kuschel's postpic (before the openPMD update):

https://github.com/skuschel/postpic/blob/picongpureader/postpic/datareader/picongpuh5.py#L74-L75

It is great to hear openPMD works great for you!
We will write a change log for the upcoming 0.2.0 release (branch:
release-0.2.0) in the next weeks. If you already give it a good test
feel free to report any problems you might observe so we can fix them
before the release.

For easy analysis of the 0.2.0 PIConGPU data in openPMD markup, feel
free to give openPMD-viewer a try:
  https://github.com/openPMD/openPMD-viewer

We would love to hear your feedback!


Best,
Axel

On 05.08.2016 04:15, René Widera wrote:
> Dear Danila,
>
> in PIConGPU before 0.2.0 it is possible to get a `gridGlobalOffset`
> equivalent attributes from each dataset.
> Each dataset contains an attribute named `_global_start`. It defines the
> distance in cells from the current data to the origin (point in space at
> time step zero) of the simulation.
>
> Best regards,
> René
>
> On 02.08.2016 09:46, Khikhlukha Danila wrote:
>> Dear all,
>>
>> I have a questions regarding data save and post-processing  in picongpu.
>> As far as I got it openPMD is currently supported by a dev version
>> (looks like 0.2.0 also has it). The standard defines an attribute
>> gridGlobalOffset which is very helpful for simulation with moving window.
>>
>> I would like to ask you if there is a way how to define a
>>   gridGlobalOffset in 0.1.2 version? Indeed, the way how HDF5 dump is
>> described here
>> <https://github.com/ComputationalRadiationPhysics/picongpu/wiki/HDF5>
>> doesn’t assume such kind of an attribute.
>>
>> Maybe there is another way how one can deduct the coordinate of the
>> moving frame at given timestep?
>>
>> Best regards,
>>
>> Danila.
>>
>

--

Axel Huebl
Phone +49 351 260 3582
https://www.hzdr.de/crp
Computational Radiation Physics
Laser Particle Acceleration Division
Helmholtz-Zentrum Dresden - Rossendorf e.V.

Bautzner Landstrasse 400, 01328 Dresden
POB 510119, D-01314 Dresden
Vorstand: Prof. Dr.Dr.h.c. R. Sauerbrey
          Prof. Dr.Dr.h.c. P. Joehnk
VR 1693 beim Amtsgericht Dresden