Is There a Written Description of the Calculation Method?
FlucsPro or FlucsDL breaks each surface into small square cells determined by the user-editable grid sizes (in FlucsPro, light fittings may also be broken into small cells if they cannot be regarded as point sources). And each 6x6 degree patch of sky is also treated as a cell. Then light is transferred from light source cells to light receiving cells using fundamental physical principles. No form factors are used - just the cosines of the angles between the rays and the cells' normal vector - it is assumed that the cells are small enough or separated far enough to be able to treat as point sources - i.e. the solid angle subtended by the source cell at a point on the receiving cell is small.
Internal windows and holes are treated slightly differently; this is done solely to improve calculation speeds. The ray from each cell on the window (or hole) to each light-receiving cell is considered, and this is back-projected to find the patch of sky where the light comes from. This is obviously faster because we don't have to consider the rays from all the sky cells to all the light-receiving cells - effectively we are filtering out all such rays which would not pass through windows or holes.
Finally the inter-reflections are calculated between all the light receiving cells - there are three methods:
* None - don't calculate any inter-reflections.
* Average - assume all cells have the average reflectance R (unlikely) and calculate the total contribution of reflections as 1/(1-R) (which is the same as 1 + R + R squared + R cubed + ...). Then assume this contribution is reflected equally to all cells (again unlikely).
* Full - progressive radiosity calculation, again treating each cell as a point source. This is described in the manuals. So you can see from the above that it is important to use reasonably small grid sizes and use the appropriate inter-reflection method.
