2D simulation of rigid bodies and images

[judevishnu]

Hi,
I am trying to simulate a 2D colloidal system .
I set the z component of positions and the 3rd dimension of box to zero while initialization. The data I get looks fine for positions and velocity (z components are not getting integrated and is zero), however I see that I get very high values along the 3rd column of the image array. Is this expected? Is there anything else I should be doing for this to be corrected? I am attaching the gsd file output here.
my_arch.zip

best,
Jude

[joaander]

No, this is not expected. What version of HOOMD-blue are you running? Can you provide a minimal script that reproduces this behavior? Hopefully 1 body and an integrator is sufficient. If not, keep adding portions of your model (one at a time) until you find which operation or force computation is causing the behavior.

[judevishnu]

Hi Joshua,
I am using v5.1.1
I tested this with the dimer example in hoomd tutorial by setting z box dimension to zero and z component of position to zero. I still see image counts in the 3rd column of image array.
I am attaching the script with one such dimers .
testRigid.zip

best,
Jude

[joaander]

Thanks! Fixed in #2071.

[judevishnu]

Hi ,
Thank you! Which version is this fixed for ? Is it fixed from v5.2?

[joaander]

#2071 will be included in the next release.