Rhino 8 Development

Couldn’t this be added as a module, like GH?

for degree 1 and 2 objects our stuff, in my experience works pretty reliably. It’s when you get to deg 3 the wheels fall off. I’m sure there are folks who will scream at me, but I do a lot of simple block outs and sub object selection and gumball edits always get me home, you just have to not ask more than it can give. I always do this with iso’s on so I can immediately see if it rebuilds poorly.

IF… all the stuff stays deg 1 or 2 I have had a lot of success with it.

puts on asbestos suit and prepares for flames<

warming up here…


…I’ll be back soon.


hi @theoutside,

let’s start by revisiting this pone from April 2016:

still unaddressed by anyone at McNeel.

Same with this other one:


1 Like

I didn’t say YOU had success with gumball edits… I said I had success with gumball edits.


1 Like

one more solidediting error being completely ignored:

1 Like

I never go above deg 1 for any solid editing. Even then it can get messed up as pulling a linear edge that was part of a planar surface out of planar can wreak havoc instantly.

right… but that would violate the deg 1, 2 rule… but I get what you are saying… direct editing is fussy and needs to be better. We agree there.

Not at all… It’s perfectly possible to have a warped surface that is degree 1 x 1:

Editing that one is fine. However, if the planar side has more than 4 linear edges - watch out…

right, not to be fussy but in the lower example, the side surface is no longer deg 1… nor could it be in that circumstance

Yes, that’s exactly my point. It’s easily messed up by one simple edit.

right but again… if you edit knowing that you can’t violate the deg 1 rule, then you should make that edit like this instead…

I don’t want to come out on the side of arguing that it’s great as is, it’s not…

I’m simply clarifying the rules of where we are. Violate the rules and the edits will very clearly suck… however…color within the lines and they work quite well.

Naah, that’s not what I wanted… This is:

The problem is you really need to be an expert to understand what traps not to fall into. The average user - nope.


To be fair, a lot of mcad programs can easily do this stuff.
(And a lot of them are much cheaper than NX)
Ironcad, fusion 360 just to name two.
And @theoutside : yes, some of this works for degree 3 surfaces too.

I’m certain there are, they likely use a completely different kernel than we do for better or worse depending on your opinion.

. again… we need to do better here with in the constraints of what and how we do things.

My point to all of this (yet again) is folks are trying to break the rules and then are upset when it fails. I’m simply trying to clarify the rules so we are being honest about what is happening.

I will point out, that this tool, and a several tools in rhino allow you to break the rules so as not to stop you from working. We don’t judge or call out if your work is good or bad, we simply allow you to make your own choice assuming you are capable of knowing for yourself if what you did is good or not.

The shell command is a good example of this… instead of saying “nope, that wont shell, forget it” we give you everything we can even if the result is incomplete, which allows you to finish the job by hand if needed, where other programs just go “nope, forget it”…

this workflow is similar, does it “break”? objectively IMO yes…

should we allow you to break it? IMO also yes… because you can then decide if you can work with the “broken bits” or not. In a lot of cases a rebuild or refit of a “broken” surface can return it to usefulness and we are not going to take that option away from you.

does that make sense?


That’s a great explanation Kyle, and a very good philosophy: give us some result instead of no result.

The shelling and fillet examples are good ones, but in those case we have a very obvious: “we could only do so much, here you go” …because we see the missing surfaces, or I trimmed fillets.

In the case of solid editing (especially if you work with isocurves off) the weasel doesn’t tell you it made a complete mess. I think before even fixing the solvers having a way to confess making messes would go a long way. Even the lowly humble Boolean tells you it had to double tolerances to work. Even puts dots where the trouble happened.

Solid editing starts with solid and honest communication.



I disagree - most people are not deliberately TRYING to break the rules. They simply are not aware of them. And the rules are actually very complex for the average user to understand.

Of course - I was never advocating anything else. I was simply arguing that you cannot make a blanket rule like “stay with degree 1 or 2 and you’ll be fine”. Degree 1 can be problematic under certain situations; and for me anyway, degree 2 is already completely out of the question.

(note the lower right result is actually a single kinked surface still degree 2 in around direction and degree 1 in the along direction.)

I use this stuff every day and some days very intensively, and for me it’s a godsend - I don’t know what I would do without the solid editing tools. But I know exactly what I can do safely and what I can’t. Even then I screw up once a in awhile, Rhino usually tells me by throwing up that spinning disk while trying to figure out how to fit a surface into something that’s no longer fit-able.

Yes, this might be a cool idea. A first step might be a message like

“Solid editing operation changed the degree of one or more surfaces in order to work”

That way you would know about some questionable stuff right away. However, it wouldn’t even cover my example above, as the degrees did not change. So a second message might be

“Solid editing operation added ‘n’ knots to some surfaces in order to fit them to the edges”
– or –
“Solid editing operation resulted in kinked surface((s) being added to model”

Anyway that’s my 2¢ for this morning. On to other stuff.


Agreed. A little more verbosity if solids are ‘messed up’ during an operation would help. Those should be displayed in the command line, then, please, not in some popup. Maybe with some other warning colour.

Ever worked with Revit? It bombards you with such warning messages all day long - the nasty, modal type where you need to read a lot, then make educated decisions what to do with that information. I regularly feel the urge to run away very far then… (gladly I don’t touch Revit much lately, and I will definitely use RhinoInside - but let’s stay on topic).

1 Like

Yeah, I understand that. It’s always a toss-up how ‘in-your-face’ this stuff should be or not… Many people don’t read the command line feedback though.

well, maybe it could start with a pop up and one of the options would be “transfer warnings to command line mode”

I don’t know if it is possible, but something that could help would be having the warning/error messages on another colour in the command line, like the yellow and red that usually programmin IDEs use.

That way, it could catch your eye for beeing something that doesn’t happen often on the command line

1 Like