Follow-up Comment #1, bug #22557 (project relax): This should only happen for experimentally measured R1 values and errors. If R1 is fit, then it should not be randomised as base Monte Carlo simulation data but rather treated as a normal parameter which is fit in each simulation and then errors calculated for it. _______________________________________________________ Reply to this item at: <http://gna.org/bugs/?22557> _______________________________________________ Message sent via/by Gna! http://gna.org/