I'd be interested to see how you've arrived at a graph from the parameters. It seems to me it produces a graph that is reducing (if normalised, it would be reducing from 1.0 at zero load with the parameters I've seen) unless I'm doing something wrong. So it does very much the same thing as the old LoadSens.
The RubberPressureSensitivityPower is the sensitivity to contact pressure. Wit the new tyre model the contact pressure is not uniform and it varies across the contact patch. And the size of the contact patch varies with with load, air pressure and camber etc.. and the pressure distribution itself is also affected. To find the overall load-sensitivity you basically have to integrate the function ((P(x,y)+Offset)^n) over the entire contact patch. The easiest way to do this is to use real-time model in ttool and run through different loads. Overall though the friction decreases with increasing load, but the physics is much more detailed in the new model compared to rF1.
Yes, but comparing the new parameter to LoadSens: since the power is negative you always get diminishing returns for extra load. The offset allows you to flatten out the first part of the curve (by moving farther into it, actually). Loadsens let you define the starting gradient (always negative as far as I'm aware) and the end point at a given load; so these parameters pretty much do the same job. There's no doubt in rF2 it's part of a more complex model, but I don't think we were really going into that...
