Nurbs theory and practice

Is there any discussion or essay that reviews the many tools/methods available in Rhino and the best use thereof in various surface blending problems?

I recently googled for something like ‘single-span surfaces’ and found ‘Alias Golden Rules’. Those are definitely a good introduction…

Hi Marcus,

I actually found that one yesterday. It’s very useful. I was hoping McNeel might have such a discussion with more task-specific examples

Buy yourself a radius gauge set and use it to measure the blends you have been asking about for the last several months. You will discover that they are all fillets and they all can be easily made in Rhino with the FilletSrf command.

The above is just an example. If you hunt around you may find better deals.

Thank you. That set would be very useful to get better measurements, but if every surface transition could be made with the fillet surface command I wouldn’t have been asking so many questions. Likewise, if everything could be accomplished with a single command, there would be no need for all the other tools that were developed.

I didn’t say every transition surface can be made as a fillet, but all the manufactured parts that i have seen you showing pictures of and asking questions about were fillets. As I said a radius gauge would verify that fact.

At the time the parts were manufactured there was a blueprint for the part and the blueprint defined the transitions as arcs.

I don’t think it has anything to do with Rhino or McNeeI, but to me it is absolutely a surprise that there is not a single book on Amazon called something like ‘Everything you need to know about NURBS modelling - how to plan your models’.

Hi Marcus,

That’s because computer math is way ahead of the populace many math teachers haven’t heard of non uniform rational bsplines. If rhino was better at things we wouldn’t need to know about it’s mathematical limitations and how to work around them. Most people are just too lazy to go this deeply, most likely a specialist book like this would not be a big seller, though I would gladly buy one.

I get quite a few references to nurbs but you won’t find much on how to do that surface blend because that’s program specific, but the general stuff is out there. Google nurbs I got these interesting links on the first page.

The Dirty Little Secrets of NURBS - Pilot3D

NURBs are the best techniques available for smooth curve and surface design, but you have to understand their capabilities to get the most out of your 3D …

What does NURBS mean and why should I care? [McNeel Wiki]

Sep 14, 2015 - There are several industry standard ways to exchange NURBS geometry. This means that customers can and should expect to be able to move …


Article about Nonuniform Rational B-Splines - NURBS a summary by Markus Altmann.


NURBS algorithm is becoming obsolete because it cannot handle sharp cones and sharp triangles. T-splines algorithm does not have this flaw, but it has other, less crippling flaws. Autodesk owns T-splines and it is not going to share it with other CAD vendors. The only way for other CAD vendors to survive is to develop better algorithm than T-splines. McNeel is too small to develop such algorithm. Big CAD vendors need about 20 years to invent and develop an algorithm which can beat T-splines. I believe that in a few years they will go out of business. This video explains main flaws of NURBS and T-splines:

Fusion 360 is notable for its unique and useful feature called healing. When you edit Fusion 360 model, you do not worry about fillets and polysurface seams because they heal automatically.

T-Splines originated from a very small group of very smart people, so tough to predict exactly where genius comes from. Typically, large organizations commoditize genius.

[quote=“ec2638, post:10, topic:35471, full:true”]
T-Splines originated from a very small group of very smart people, so tough to predict exactly where genius comes from. Typically, large organizations commoditize genius.[/quote]

Yes, the group was very small (Thomas Sederberg, Jianmin Zheng, Almaz Bakenov, and Ahmad Nasri). They invented basic idea but neither they, nor McNeel had the resources to develop this idea into mature algorithm. This is why T-Splines was bought by the Autodesk.

We are perceiving the crux of this differently and are going to have to agree to disagree on your fundamental point. Water under the bridge semantics anyway. It’s all good…

Looking back over the (increasing) speed of technological developments these last 20 years makes me think such a declaration is bordering on irrational. Of course some of my attitude is akin to whistling past a graveyard. The movement of the Big App corporations to a rent collecting - cloud-based model is troubling. I’ve pretty much written off Adobe and I get the same impression from Autodesk. Everything moves towards more control, curation and profit.

About a decade ago Autodesk became greedy monopoly.