Sort:  

Actually I remember reading your post back then. If I am not mistaken the shape of a Sigmoid_function is similar to the one on the EIP.

Yes, and I am in favour of the convergent rewards curve.