Currently, if a livecatch happens, angular momentum (only for x and y) is nulled, while livecatchbounceMultiplier is only used on velocity.
It could be tested, if the livecatches feel better, when the liveCatchBounceMultiplier (which is for example 0.2 for a slightly inaccurate livecatch) is not only applied to velocity, but also to the angular momentum.
possibly only angular momentum in the direction of the collision-normal could be reduced. (like for velocity, or the same amount of the overall velocity reduction, otherwise calculation would be quite hard.)
Currently, if a livecatch happens, angular momentum (only for x and y) is nulled, while livecatchbounceMultiplier is only used on velocity.
It could be tested, if the livecatches feel better, when the liveCatchBounceMultiplier (which is for example 0.2 for a slightly inaccurate livecatch) is not only applied to velocity, but also to the angular momentum.
-->
possibly only angular momentum in the direction of the collision-normal could be reduced. (like for velocity, or the same amount of the overall velocity reduction, otherwise calculation would be quite hard.)