I've only been able to do a few tests so far, and in fact I'm now running through analysis again after having changed the sampling to every frame rather than every 10 frames. I did this in the hopes I could observe the effects of changes in CPS, Twang, and Lag more effectively (i.e., I think some of the effect of tweaking these settings was being obscured by the 1/10 sampling).

Other tweaks I made to the script for testing purposes include changing the sign of Grav in the equation where it is applied. I'm guessing this might mess things up on lateral movements, but I was finding that applying a positive setting to Grav was creating an upward rather than downward bias when applied to a Mag ytran for which negative values pull the parts down and positive push them up. My test figure is only moving up and down at the moment so I don't know the effect of changing the sign of Grav on side-side movements. Another workaround might be to rotate the base of the magnet 180 degrees. Will play with that more in a bit.

Before changing the sign in the Grav equation, I changed the min-max of the Gravity slider so it could go negative. Also, I changed the sensitivity since even a very small value of Grav (e.g., 0.2) was having a huge effect when being applied to the BrstNatural MT. I did this because the Multiplier seemed to have no effect so I couldn't otherwise dampen the Gravity effect to see how it was working.

I also reduced the sensitivity of the Lag slider since small values of lag (again, like 0.2) were imposing lags of something like 10-15 frames.

Ah... the Analyze just finished for the every frame test so I'll report back in a bit.

This is really great stuff!