Roland Plüss wrote:
> d - e1 - e2. The result is then dynamic with the position and still more
> or less accurate.
>
>
Your proposal fails when the objects undergo rotation during the step
(unless they both are spheres).
Cheers,
/ h+