I have posted this before.but just to give it a bump:
FilletSrf often fails miserably when the chain of tangent fillets encounters a same size fillet.
The fillet radius in this example is 1mm.
Same_size_fillet_R1.3dm (265.5 KB)
Also I noticed in the latest version of the WIP FilletSrf now has a "Join"option. That is a good idea but the Join option should work the same as the Join and Boolean commands where the end result inherits the properties of the first selected object to be joined.
The current method that FilletSrf uses to create same size fillets often makes BadObjects (when it does anything at all). Rhino may not identify them as bad objects but subsequent downstream operations will reveal the bad objects. For instance, creating a mesh will produce holes in the mesh in places where the NURBS version shows none. That’s because the NURBS object is just hiding the holes (i.e. hiding the bad trims and joins). You can see an example of this in the above posted example. A corner fillet that looks OK from a distance has holes when you create a mesh or when you zoom in close in shaded mode
Also, FilletSrf still fails often when the Fillet comes to a point (at points where the two base surfaces become tangent). This failure will also result in bad trims and joins. There have been lots of examples posted, but here is another.
Pointy_FilletSrf2_R1.3dm (139.2 KB)
This file shows that clean result can be had if the pointy end is fixed before trimming and joining.