Why FilletEdge fails here

Why Rhino can’t deal with this simple filleting task? 2018, Rhino 6

Fillet a corner with 30 mm radius of a box 30 mm thick:

thanks, Tobias

Because the rolling ball has the same radius as the surface it’s trying to follow.
It completely deletes that surface.
If the radius is less than 30, or you model it another way, you can do it.

Ok… but shouldn’t Rhino be smart enough to deal with this?

Rhino uses the forum to tell you what it can and can’t do, and why. :wink:

// Rolf

1 Like

Rhino users use this forum to tell McNeel what Rhino can’t and should do and why. :wink:

It remembers me of the good ol’ Rhino 4 times where the FilletCorners command with radius 50 would fail on a rectangle of 100 x 200. By that time I got the same kind of answer: The command would delete 2 segments of the rectangle and therefore fails…

Time to move on and make simple things like this work.

There’s a little more to it than that I’m afraid. Besides Rhino not giving very explanatory responses to “expected functionality” it’s allowing you to do interesting and even spooky things has its benefits too. It allows you to do things which other CAD tools doesn’t allow you to do.

This aspect is sometimes (often) overlooked, but it would be good if dialog responses to “expected behaviour” would be more helpful. Sometimes a little hint about what to do differently would solve problems like this.

// Rolf

There is way enough spooky stuff created by Rhino. Let FilletEdge fail on complex geometry… fine. Then I’m all with you. But in this extremely simple cases it’s a shame that it fails and the user has to do it manually.

1 Like

Since the command is doing exactly what it was designed and reported to do, I’ll take issue with your characterization that is is a “shame”.
That said, I’ll fully admit it is an inconvenience.

@John, FilletEdge is obviously a powerful command, but actually I don’t know what its designed for and reported to do… (I never checked the help files to see if my complaint is legitimate). I just use it completely ignoring its theoretical function. :sunglasses: and the way I use it it’s a shame that it fails here.

When I presented my Rhino projects to Bob in 2008 he said that Rhino was never meant to be used like that. I usually don’t care about such things :wink:

So just see it as a wish that FilletEdge can handle this situations. It really should

Hi Tobias - it should. The broader requirement, I think, that would help solve this and many, many other cases, is for Rhino to be able to stick the fillet to a face on the polysurface that does not have an edge that was selected for FilletEdge. i.e. the fillet has to be able to jump across faces and ignore,/remove them and consider some interior face that is away from the edge. This is one the Great List but I gather, is a significant project competing with others…


1 Like

Thanks Pascal. And yes, of course it’s competing with other things on the ever growing list… but it’s a thing which especially novice users would probably consider as a bug. Or at least they don’t understand why it doesn’t work.

For me I do understand the limitations and I know many ways to get my fillet done (especially with a dead simple thing like this) but even then it would save me a lot of time to have a more intelligent FilletEdge tool

gr, Tobias

Hi Pascal

Another thing (maybe a simpler project) that may help in several cases if this, IMO.

mergetrimmed.3dm (221.5 KB)

Being able to merge (back) trimmed surfaces.

This is a little piece of a polysurf I’m working on right now.
I had to split the tapered surface to be able to sub-select and edit it.
But now I am no more able to use FilletEdge on it.
( Surfaces “A” )

If I could merge the surface back into a single polysurf ( Surface “B” ),
then FilletEdge would work.

… Makes any sense ? :slight_smile:



If doable, the ability to merge trimmed surfaces would represent a huge step forward in general, in addition to a contribution towards more robust filleting.

Inherent filleting improvements, obviously, a big plus, as indicated ad nauseam.

As best I can ascertain from arms length, V6 brought great enhancements with much underlying work underneath the hood - including v5/6 Mac integration. V7 direction…?

A laser focus on modeling enhancements for v7 has been my mantra. The timing is right IMO. Evidence indicates such is under consideration/active.

1 Like

Hi Emilio - it makes sense as far as it goes but I guess that is fundamentally the same problem for the filleter as I described above. - merging as you suggest might be a good thing to be able to do regardless and I see how it helps in this case, but that is just the circumstance. (that said, in general things behave better split at tangents)


Sure !

Yes, exactly.
I think that merging surfaces would be mostly a workaround to be able to use other commands.
But somehow, we are used to look for workarounds … :slight_smile: