Closed plexoos closed 1 month ago
8 out of 100 photons have significantly different histories cc @ggalgoczi
I found errors in the raindrop model. I am not sure this is because of that. When I tried to use CaTS for raindrop on GPU it did not work until I corrected the optical surface. Will propose a fix to the model today.
Thank you Gabor! It would be nice to see if the issue can be reproduced with our tools, feel free to create an issue
Using esi-shell container
Added a basic test comparing photon histories for GPU and CPU based simulations.