The output for time-horizon 1.5 should be compatible with the output of the dcurves package: So-called prevalence of 0.22 for the dcurves package and 165 real-positives / 585 real-negatives for rtichoke output.