Falloff to Zero for Bone Envelopes

As probably is known, when using bone envelopes directly (by checking only these in the armature modifier), the deformation will not be smooth at locations where an envelope ends (see images).

The manual says that envelopes create weights with quadratic falloff. I assume that means the weight will be non-zero at the end of the envelope from the inside, but exactly zero at the outside, thus being discontinuous. So it is no surprise that the deformation will be discontinous, too.

That should be possible to fix by changing the falloff function to one that smoothly goes to zero at the end of the envelope, e.g. 1+cos d.