Inconsistent filleting behavior between tools

[eobet]

Here is the result from VariableBlendSrf:

And here is the result from BlendEdge:

The green circle is the natural result that would be desired in all cases, and I don’t understand why it only happens in 1 of these 4 instances. The fact that there’s no explicit option to not extend the blends when you aren’t trimming and joining adds so much work to an already high filleting workload…

(Rhino wasn’t able to fillet past the “tight” corner you see to the left in the screenshots, so that’s why I’m in this mess in the first place.)

[Fred_C]

Moved to Rhino for Windows.

[pascal (Pascal Golay)]

Hello - the difference is, one is a surface tool, and one a ‘solid’ tool. The first does not care about the overall topology of the input.

-Pascal

[eobet]

Imo, neither seems to care.

But to try to understand what to me at glance seems random, can you tell me why only 1 in the 4 given examples choose not to extend the fillet/blend past the shortest surface?