missing hits from opticks

[plexoos]

@ggalgoczi Could you take a look at this code and let me know if you spot anything off?
https://github.com/BNLNPPS/eic-opticks/blob/main/src/g4app.h#L264-L294

For some reason, I’m not getting any hits when I run this test:
https://github.com/BNLNPPS/eic-opticks/blob/main/tests/test_simg4ox.sh

[plexoos]

I switched to opticks_raindrop.gdml and now I see opticks hits but no geant hits

USER=fakeuser GEOM=fakegeom simg4ox -g $OPTICKS_HOME/tests/geom/opticks_raindrop.gdml -m $OPTICKS_HOME/tests/run.mac

@ggalgoczi do you know why the Geant hits might be missing now?

[ggalgoczi]

Interesting, most likely it is due how the geometry is defined in the GDML. Opticks uses the EFFICIENCY property for the border while G4 uses the SensDet tag. Is that assigned to the correct volumes?

https://github.com/BNLNPPS/eic-opticks/blob/main/src/g4app.h#L141-L157

These are the lines where G4 hits are processed. Do these output 0 hits?

[plexoos]

I am using this geometry https://github.com/BNLNPPS/eic-opticks/blob/main/tests/geom/opticks_raindrop.gdml
It's identical to the one you originally created: https://github.com/BNLNPPS/esi-g4ox/blame/main/geom/opticks_raindrop.gdml

Here is the my test command:

docker run --rm ghcr.io/bnlnpps/eic-opticks:develop bash -c 'simg4ox -g tests/geom/opticks_raindrop.gdml -m tests/run.mac'

it produces 0 Geant4 hits according to the log (see PhotonSD lines):

### Run 0 starts.
PhotonSD::Initialize:  G4_WATER_lv_PhotonDetector   G4_WATER_lv_PhotonDetector_HC
PhotonSD::EndOfEvent Number of PhotonHits: 0
2025-07-28 14:33:22.952 INFO  [1] [QSim::simulate@395] sslice {    0,    1,      0,    100}
2025-07-28 14:33:23.249 INFO  [1] [QEvent::gatherPhoton@584] [ evt.num_photon 100 p.sstr (100, 4, 4, ) evt.photon 0x7f968c000000
Opticks: NumHits:  100
 Run terminated.
Run Summary
  Number of events processed : 1
  User=0.720000s Real=0.782765s Sys=0.010000s [Cpu=93.3%]
================== Deleting memory pools ===================
Number of memory pools allocated: 11; of which, static: 0
Dynamic pools deleted: 11 / Total memory freed: 0.086 MB
============================================================

[plexoos]

When I use another geometry basic_detector.gdml which I created some time ago for tests, I get 2 Geant4 hits and 14 Opticks hits:

docker run --rm ghcr.io/bnlnpps/eic-opticks:develop bash -c 'simg4ox -g tests/geom/basic_detector.gdml -m tests/run.mac'

### Run 0 starts.
PhotonSD::Initialize:  MirrorPyramid_PhotonDetector   MirrorPyramid_PhotonDetector_HC
PhotonSD::EndOfEvent Number of PhotonHits: 2
2025-07-28 15:25:49.048 INFO  [1] [QSim::simulate@395] sslice {    0,    1,      0,    100}
2025-07-28 15:25:49.339 INFO  [1] [QEvent::gatherPhoton@584] [ evt.num_photon 100 p.sstr (100, 4, 4, ) evt.photon 0x7fbe54000000
Opticks: NumHits:  14
 Run terminated.
Run Summary
  Number of events processed : 1
  User=0.730000s Real=0.764544s Sys=0.010000s [Cpu=96.8%]
================== Deleting memory pools ===================
Number of memory pools allocated: 13; of which, static: 0
Dynamic pools deleted: 13 / Total memory freed: 0.088 MB
============================================================

[ggalgoczi]

This should be solved by my latest fix for the opticks_raindrop in:
4101e08

The fix is only about the sensitiveness of the surface in EIC-Opticks. Before photons going inside were recorded. Now photons going out are recorded.

For the bordersurface used here by Geant4 directionality does matter unlike for skinsurface.