In beersheba, the basic process for the deconvolution is as follows:
- select z-slice for deconvolution
- interpolate input for the deconvolution
- apply the deconvolution
- apply the e_cut to the output
- repeat previous steps for all z-slices, then
- redistributes the energies
- returns the deconvolved event
The way this is done in Beersheba is a bit obtuse, but I'm happy to write out a more explicit explanation of how this is done.
I believe for the energy cut to have any meaning in the absolute case, the cut should be applied after the energies are redistributed, or e_cut should be renamed. This is because the cut you're applying is to the interpolated and then deconvolved data which is no longer cutting on energy explicitly. This is highlighted by the fact that you redistribute the energies after this cut is applied.
I'd be happy to discuss this issue further with anyone interested/more knowledgable with the code.