I have a need to simulate coherent irradiance (power) on the detector in non-sequential mode. First I have tried to proceed the data directly from ZRD file. Based on equations on page 790-791 in Zemax manual I summed all Re and Im values of every ray hiting particulate pixel (based on x,y coordinate on the detector) in all three directions (X,Y,Z) and then calculate RMS value (TOTAL = sqrt((Re,x^2 + Im,x^2)^2+(Re,y^2 + Im,y^2)^2+(Re,z^2 + Im,z^2)^2). When I use the DDR output from detector ( no smoothing, incoherent output) I get similar result but only similar (see figures). The amplitude for both outputs could be relative however, the result from DDR is smoother that ZRD. Since I am new to Zemax I do not want to continue with more complex simulations unless I fully interpret those results. My question is shall both outputs (DDR and ZRD) yield the same results? If so what may be wrong with my approach?

Thank you very much for any clue.