XNurbs releases a ground-breaking NURBS software




Hi Toshiaki_Takano,

“rhino users, they’ll probably spend less on plugins than solidworks users.”

Thanks for your suggestion. Please refer to my reply to @markintheozarks. The Rhino plugin’s capacity is pretty much the same as our SolidWorks one, so we may need to keep the pricing the same. We haven’t made any announcement to SolidWorks community, and are considering a promotion price before announcing it to SolidWorks community. (Yes, SoidWorks users are normally willing to spend more than Rhino users, but our price should be the same or it will upset some users.)



Hi jim and ec2638,

Yes, we do have a tolerance setting (“PropertyManager.png”. One simple UI to solve all kinds of problems for NURBS creations – unlimited capacity for solving NURBS and generating high-quality surfaces.). The “Relax Precision” slider enables you to apply a different precision, i.e., 0.001 mm, 0.01 mm or 0.1 mm.

“Rhino has a global tolerance (for NURBS) that the user can set and also many of the NURBS geometry creation tools have local tolerance settings so that the User can temporally overide the global setting.”
What is the NURBS tolerance setting in Rhino? We may match them easily.


Whatever the user decides it should be. There are default settings if the user doesn’t care to decide.



If a user sets the tolerance as 0.00001 mm (assume that the model is 1X1X1 m), then all NURBS algorithm won’t work. This is why most CAD systems hide the setting from users.



“If it truly is revolutionary, …”

XNurbs uses one simple UI to solve all kinds of problems for NURBS creations – unlimited capacity for solving NURBS and generating high-quality surfaces based on energy-minimization method. After trying XNurbs, maybe some designers would tell others this fact. Then we will know if our price is acceptable for Rhino users.


True, but Rhino still lets you do it and suffer the consequences. Since it is used for everything from jewelry to shipbuilding and architecture that flexibility is necessary, unlike many other CAD systems that have or had a fixed unit system “under the hood.”

Anyway, you should just get a Rhino demo out instead of trying to argue its virtues on the Internet with a combined hundreds of years of surfacing experience here.


That just isn’t true. What the user discovers very quickly when they try that is the algorithms produce very dense geometry which in most situations will bog the whole modeling process down.

Most CAD programs hide lots of stuff from the users, Your SWX users won’t know anything about the internal structure of the surfaces that your software produces, but Rhino users will be able to see the choices you make in knot spacing, 2d parameters and control point placement of the surfaces that you create. And don’t be surprised if you get critical comments in regards to those choices.


In Rhino, ICEM or Alias, you can set a tolerance to which several tools refer. Industrial designers work in a huge variety of industries so that consequently, there is no “best tolerance”. A tolerance is always in the eye of the beholder. This is why (currently lacking in Rhino) one needs interactive and persistent locators that - while one modifies surfaces - tell the user how far she/he is from the desired condition. Then, surfaces always must be checked visually with ISO-angle tools or Isophotes, interior curvature combs, etc.

A tool may achieve G2 or G3 to tight tolerances, but what if behind that boundary condition, the surface looks awful? Software cannot know what looks good. The user has to intervene. Which is why interactive persistent tools are a prerequisite. There is no “one-click-surface” solution, like the “app generation” often desires. Software is clueless ; )


Spot on.

@XNurbs Part of a high-end faucet. On the left (single span surfaces, degree 5 and 7). On the right (a terrible mess). To iron out the internal shape of the surface on the right would be an exercise in futility.

(Marc Gibeault) #50

That’s not true.
Most hide the controls very well but you can change the setting if you know enough.
It’s true that most CATIA users don’t know about this, and it’s one of the reasons I see such awful surfacing from them! It’s a very powerful software but, as with any other, understanding what is going on under the hood really helps not being stuck in bad geometry.


Lagom makes a key point to highlight!

We’ve all asked software to provide continuity, and get it, only to find a resultant form that does not pass the sniff test for one reason or another.

We ‘pain’ to evaluate form in the screen before development progress further, is 3D printed, or god forbid, a tool gets cut with unforeseen and unresolved issues.


XNurbs technology seems promising, but serve practical examples, real modeling test. Currently Rhino is very defective, in multiblend, for example.
Fixed this and the best fillets would become almost perfect software! A dream…


Really? Wanting some advanced functionality like multiblend doesn’t mean the product is “defective”. That’s like saying my new Ferrari is defective because its not as fast as a Bugatti Chiron.


I meant this: ‘defective’ in managing multiblend through “Patch” (a raw tool, useless, in my opinion).
Let me say that Rhino is far behind other competitors, he still does not even have the SubD tools that have been present in other CAD for several years.


Again, aaand? A fancy “multi-blend” is nice to have, but “High end” surfacing isn’t using fancy tools that magically produce technically smooth surfaces from whatever garbage you feed it,


We do not add anything else. Multiblend is not useful, the fillets can not be handled with a nurbs modeler, multisweep does not make sense … etc. etc.
Content you!

I do not think there can be magic tools, I’m not so stupid.
Can I express my opinion? Except for adding some options, I have not seen any enhancement to surface creation tools, all commands have remained basically the same as those of Rhino 5: patch, sweep, network, loft, blend surface etc.
One of the few commands that surprised me was “Extend surface”: more intuitive, more powerful and very different from Rhino 5.
Significant improvements have been made on other fronts, such as display, rendering, annotations, meshing and various ‘cosmetic’ finishes.
I am wrong?


As far as I’m concerned, I would like to see some well-known cases where Rhino has difficulties … more than multiblend, multisweep and beautiful company!
Do not you agree with me? (as usual, no!).

(Marc Gibeault) #58

If you continue to repeat the same thing ten times a day, people will get more and more aggressive with you.
I come here to get information, answers and to make suggestions. Reading your posts all over the place just isn’t productive and turn me down.


Do not talk anymore, for your good! sorry…


Hello all,

The following are some surfaces generated by XNurbs (XXXXXXFilled.igs). The position tolerance setting is 0.001 mm. Additional requirements for each model:

Blending.IGS: Curvature continuity
Cylinder.IGS: Curvature continuity
patchhole.IGS: Tangent continuity

As you all can try and see that, while Rhino can show the knots, 2d parameters and control points of the surfaces, no human-being can generate a better result by modifying them. The surface generated by XNurbs is the one among all possible solutions that has the minimum energy and satisfies all input constraints, i.e., the smoothest surface.

We could complete the development of the Rhino plugin and release it within a couple of months if Rhino users like XNurbs product and their acceptable price is reasonable. McNeel is a responsible partner and they also want us to be financially successful for the plugin. After some guys in Rhino Forum try XNurbs, perhaps we could get more info about Rhino users’ thoughts on this, which will benefit Rhino community.

Blending.IGS (21.9 KB)
BlendingFilled.IGS (107.1 KB)
Cylinder.IGS (42.1 KB)
CylinderFilled.IGS (61.5 KB)
patchhole.IGS (80.2 KB)
BlendingFilled1.IGS (145.3 KB)
patchholeFilled.IGS (187.2 KB)