In the course of teaching myself Blender, I've always been led to believe that marking edges 'Sharp' had an effect only if used in concert with the Edge Split modifier. But then I saw a couple of tutorials which used the 'Sharp' Edge Data property without splitting edges.
The objects below are identical. Autosmooth is enabled for both at 60 degrees. Both have had a Bevel modifier applied to all edges but the ones in the curved section, controlled by weight. The only difference between them is that the one on the right has had its sharper edges marked 'Sharp'
And after the Bevel modifier is applied, there are split vertex-per-face normals on the marked 'Sharp' version, again on the right. The vertex count in both versions is the same.
The Blender manual goes no further than to say the 'Sharp' property works with the Edge Split modifier.. but no Edge Split modifier has been assigned. Can anyone explain what marking 'Sharp' actually does, under the bonnet, contrary to all the explanations I've heard?