IMC and ISMC both working well for grey case and for multi frequency, with several new benchmarks added. New geometry class to easily generate large grids. Redid all of the hybrid tracking bits and bobs that I presented at M&C2023, should now be a lot easier to follow and also more efficient. In a lot of cases surface tracking within the new geometry class will be fast enough that there isn't much time to be saved anyway, but definitely has some interesting possibilities especially if using on top of a CSG geometry. When merging with the main branch I discovered that you've created materialSource_class.f90... I had a similar source with the same name, for now I've renamed my original one to imcMaterialSource_class.f90. At some point it might be possible to combine these in a fairly general way, although my one is a bit more complex so you might want to keep it IMC-specific. The main differences between the two are that I am sourcing particles in each material with numbers proportional to the total energy emitted, and if the geometry is geometryGrid_class then this is done within the bounds of each material rather than rejection sampling within the whole geometry which would be too slow for some of the benchmarks.
IMC and ISMC both working well for grey case and for multi frequency, with several new benchmarks added. New geometry class to easily generate large grids. Redid all of the hybrid tracking bits and bobs that I presented at M&C2023, should now be a lot easier to follow and also more efficient. In a lot of cases surface tracking within the new geometry class will be fast enough that there isn't much time to be saved anyway, but definitely has some interesting possibilities especially if using on top of a CSG geometry. When merging with the main branch I discovered that you've created materialSource_class.f90... I had a similar source with the same name, for now I've renamed my original one to imcMaterialSource_class.f90. At some point it might be possible to combine these in a fairly general way, although my one is a bit more complex so you might want to keep it IMC-specific. The main differences between the two are that I am sourcing particles in each material with numbers proportional to the total energy emitted, and if the geometry is geometryGrid_class then this is done within the bounds of each material rather than rejection sampling within the whole geometry which would be too slow for some of the benchmarks.