Note: A workaround for Hydrology only models that send all subcatchment flow to the pervious area. Briefly, v16 had a total infiltration of 14 inches in the attached model but v21 has a total infiltration of 3.5 inches. The input file uses the attached rainfall file and it routes all of the flow to the pervious area of the subcatchment. If you plot the losses in v16 and v21 you will see that the losses stop after about 30 days in v21 and continue in v16. I found that if you put in a small evaporation rate (0.01 inches/day will do) then v21 will duplicate the answers of v16. Looking at the depths in the subcatchment it seemed in v21 without evaporation that the depth in pervious area was stuck at the depression storage value.
In summary – the problem seems to be that the losses stop after the depth in the pervious area is above the depression storage unless you have evaporation to sort of kick start the infiltration losses again.