T-Splines for Rhino end of life

Eliminate polysurfaces.

If polysurfaces are eliminated, all solids will be easy to edit.

1 Like


1 Like

Sorry to lose tSplines for Rhino. I started using tSplines in 2010, took the course in it, and watched all their webinars with a sense of complete wonderment. But I confess I never quite “got it.” I use it in a patient and deliberate way for two things.

For medical illustration work it is incomparable. For vascular, nerve or other biological systems the tSplines images look perfectly natural and right. Note that these objects, e.g. an aorta, do not have any fixed or machined dimensions, only known and approximate dimensions.

For my hobby, model aircraft, tSplines is very helpful in designing fuselages, including wing roots and empennages. The fillets just happen. Sky Greenawalt has made a science of using tSplines for aerodynamic shapes, and I strongly recommend his seminar on the subject.

I just slog through it, using the most utterly basic approach, based on the command: tSplines from lines. For precisely dimensioned objects, like the wings, I stick with NURBs.

I visualize the tSplines surface as a bubble in a cage, and I just keep hammering together, editing, and then discarding and recreating putative cages until I arrive at a bubble that works, that is, looks like an airplane. Box mode editing accomplishes the same thing. It is a fast way to build and revise cages and I use it, but I prefer to rely on tSpllines from lines.

The fluid, supposedly easy and intuitive realtime surface editing techniques you see in videos never really worked for me in aircraft modeling.

However, to judge from the webinars, some users have wonderful success using these freeform techniques. I think they are artistic. I am not.

In my experience I am able to do subtle, limited surface edits in realtime in smooth mode, but to keep the surface under good control, I must sooner or later extract the cage, simplify it, simplify it some more, and (often) realign it – and then recreate the bubble using tSplines from lines. I learned this approach from an aeronautical engineer. With patient work it produces good results. It is remarkable how few control points tSplines requires to build an airplane fuselage. This is a reproduction of a free flight model constructed in 1937. The cutaway surface was formed in tSplines, the rest is Rhino3D. Point is, tSplines and Rhino complement each other.

Not sure if the problems are with me or with the program, but I accept them. I would not try to get along without tSplines.

If tSplines will not run under Rhino 6, I will keep a copy of Rhino 5 on tap, simply in order to retain access to tSplines.

you believe tsplines is the only answer for creating a “quick” hull? i dont know in terms of precision if thats going to do the trick from what i read. even though it all can be scaleable. but in matters of a fast build up, lofts can work easily equally sufficient and might be probably more precise. and talking about aircrafts… pls dont do quick builds… thats a security and in many matters also an economical issue.

if you work on small hobby production of aircrafts or representational displays you can use whatever you want i think… but since there is an option which can do a fast and precise job, why choosing a ghetto version of a process :smiley: ok ok i get it is nice to work with bubbles… :stuck_out_tongue:

1 Like

While the underlying tech was always special, the original interface was tough to get. Autodesk completely reworked its tsplines interface in Fusion360.

This may have irked those with hard earned time investment profiancy in the original implementation, though my observation has been that newbies, as a group, picked up the new interface much faster, with greater ease, improved efficacy.

The AD user interface implementation is a significant improvement IMO. I’d speculate you’d ‘get it’ with much less effort.

Hopefully, the Rhino sub-d interface will focus on ease of: learning curve, use, numerically accurate form definition. (Regional accuracy capability similar to nurbs.) One can imagine a slick implementation negating, or significantly reducing, need for VSR type functionality.

Success melding already stellar surfacing, with product design oriented sub-d, combined with the power of grasshopper, will maintain Rhino as a conceptual design development fav into the future.


i never understand this t-splime yes i know what you can do with it but … its just so stupid i buy rhino
some years ago the reseller recommend it …oke i have it and in a few month later i discover t-spline…
i was thinking i buy a car (rhino) with out an engine …now i have to buy the engine at €800,-
its just so stupid
greetings Peter

From my point of view, VSR approach delivers a quality impossible to match by any sub-d method I know at the moment. I use T-Splines a lot to explore form quickly, like a sketch but in 3D, or for products of small scale that don’t need a simple and controled structure of the surfaces (besier patches). Both methods ar good but for different things, no one can replace the other.

Agreed, and the assumption is that you’ll loose VSR access in future, therefore some advance to surfacing and product design optimized sub-d, is the path to sustainability.

You won’t if you stick with Rhino V5…

I wonder… http://www.carbodydesign.com/2014/03/shape-modeling-plug-in-for-rhino-is-72-percent-off/

I’ve never used the VSR but it seems to get a lot of good reviews. I think I’d still buy T-Splines for V5 knowing what I know now.

I needed to model some coral for a job where it was going to be 3D printed, moulded and cast and then plated in gold and platinum - there’s no way I could have done these in Rhino easily.

This my very first project in T-Splines and my first sub d adventure:

My old MacBook Pro 2013 with bootcamp:

A 1:1 123D Catch - part of a wax positive which was cast in bronze with one of the coral pieces in situ:


1 Like

Yep, best way to do this - make it cheap so people buy it and get used to using it, then pull the plug, while at the same time saying “its over here now, come and get it…”. --Mitch

1 Like

In the next decade NURBS will be replaced with three mutually incompatible geometric modeling kernels: Autodesk kernel (T-splines), Dassault kernel (SolidWorks and CATIA), and Google kernel. Small CAD vendors will be either sold to Google, or they will go out of business.

Don’t think so…

ACIS - used by Autodesk and a great many other MCAD systems, will likely be around for a long time to come. Catia uses its own kernel, while Solidworks uses Parasolid. And, according to the article above, the Russians are perhaps to be reckoned with…


1 Like

This has been coming for a long time so whilst not what I want to hear… inevitable. I knew it was coming before I made the change from Clayoo to T-Splines about a year ago. I have no regrets, I love t-splines.

For my jewellery where I design more on aesthetic than measurement, where filleted edges are desired more than sharp corners…t-splines is heaven. Quick edit on a design …t-splines make it so easy for a lot of things.

Sure its a little buggy like the ghost mesh it leaves after a cage edit or flow but you just delete it, all good and move on. Mixed up normals when you convert to Nurbs…explode and join with a macro and done.

Rhino is like Photoshop to me…I know I am using a very small percentage of its power for what I do and my level of proficiency. I am reminded of this every time I read posts from the likes of Pascal and many others.

I still use CS5 and really have no need for anymore… and in the immediate future will likewise stay on Rhino5 with t-splines. I am pretty sure that will either do me out or until something else comes along, hopefully from McNeel.

Seems like natural evolution to me…McNeel will develop Rhino in a way that they think will allow them to sell the most number of licenses…and only they really know what their largest target audience is. I doubt it is what I do.

Relatively short term solution…and not good for ensuring future development and longevity.

Thde only catch (apart from all the other obvious ones): it’s not really “over there” at all… if you don’t count renting Alias Surface/Autostudio for 8000+$ dollars a year.

From where is this info? I can’t found any official statements - just a rumors. Does anybody knows some more?

Here is an answer from Autodesk support:

Krittika, Nov 1, 01:56 PDT:
Hello from Autodesk!

Currently we do not have infromation regarding this. If we get any update about this information then we will surely update you accordingly.

Thank you!

Should you need additional assistance, feel free to reply directly to this email.

Best Regards,

Autodesk Education Community Support

1 Like

Its at the very top of this topic!
Probably a matter of miscommunication at Autodesk.

Regards, Robert

Hi Sochin,
You are the only one I know who has moved from Clayoo to TS, can you tell us the reasons you prefer it?
I was considering migrating in the opposite direction since TS will be dropped, and I doubt McNeel will have their SubD solution ready for V6 (please someone tell me different). Does anyone know if Clayoo will run in V6?

It was stated in the Clayoo webinar that as soon as Mcneel is ready, then they will do there part to make it work on their part for V6 .(All in laymans terms on my part). If you go to the webinar towards the q&a near end. I think that is where I heard that. Fwiw