VSR end of Life-

I use VSR professionally (I’m using it right now), I’m not here saying Alias is great and VSR isn’t. But that tool shows a way of creating ball corners that is FAST and well done. I make lots and lots of ball corners using VSR/RH5. It’s doable, I get the same result, but it’s time consuming and tedious to do them to that same quality. You need to do a lot of pipe trimming, blending and matching to get that result in VSR. I’m on your side man.

Rhino users upgrade to newer releases if they see important improvements and new features they need. Many Rhino users requested a bunch of improvements over the years, yet the major ones were not implemented at all. We talk about several improvements to the main NURBS modeling tools such like “Blend surface”, “Match surface”, adding “Explicit control” and convenient control point editing with handles along the control polygon. People wait for these since Rhino 4, Rhino 5, Rhino 6, Rhino 7, and most likely we won’t see them in Rhino 8 as well. Existing Rhino users decide whether they want to upgrade to the next Rhino based on the new features added compared to their current program. By adding functionality to the main NURBS modeling tools “McNeel” will be able to attract a greater user base, hence sales and profits. I also would love to buy Rhino 8 if it has at least a portion of the functionality that VRS offered for Rhino 5 nearly a decade ago. :slight_smile:

However, as I mentioned in an earlier post, if Rhino 8 stays nearly the same as Rhino 7 regarding “Blend surface” and “Match surface”, then I see no reason to upgrade to Rhino 8 due to the lack of important features that were requested by many Rhno users for years. Not to mention that currently I have Rhino 7 and Rhino 8 WIP, and in fact the latter renders the 3d objects worse than Rhino 7.

Rhino 7:

Rhino 8 WIP:


You are probably not a very bad cad modeller…although the neighbours have already warned me about some Tom_P guy, they said he uses 4-sided matching, and even worse MULTIBLEND… :stuck_out_tongue_winking_eye:

The issue is simply workflow related. ICEM basically has only one type of surfacing tool, a simple patch that you feed 2 or 4 points or curves into. Thats it! (There is tons of stuff under the hood, but we are talking Class-A surfacing here).

You dont set up any conditions like in Rhino’s “curve network” or constraints or anything.

Instead you just build a little patch however you like. I like to bulid from curves, so you can control the main countours. (Since we usually build to theoretical edges, I actually dont even use the Blend Srf tool :slightly_smiling_face:)

Due to the single span nature of Class-A geometry, single surfaces cant represent complex shapes like in Rhino, so tools that mash up 25 crosssections into some sort of rats nest network are of no use, since you end up with a simple patch of 4x6 or 5x6 anyway.

But here comes the main point: You dont set up conditions while creating a surface, which you forget about and cant even check properly after a surface is created. But instead you run EVERY single surface through the matching tool to create your topology, checking it right away.

And the matching tool is where the magic is happening. You can control everything relevant from within the tool. CP count, distribution, edge angle, blending along the patch, projection, etc.

The matching is not only creating continuity, it is also a shaping tool.

While you are doing that, you always keep one eye on the diagnose bar, that tells you right away what is happening between the 2 surfaces.

Deviation is instantly recognised. Thats why this workflow is basically self correcting.


With Nurbs modelling you build a topology that remains unchecked until youre almost done, and then notice some problems deep within your topology that might make 4-Sided matching neccesary. While in ICEM checking is done in every step that you do, without even asking for it.

And that workflow is super fast too. Class-A is not something that is painfully born over days and may never be touched again. Sometimes a whole front bumper gets rebuild twice a day until designers are happy. And each time in full Class-A quality.


Maybe you’re not familiar with TsPipe? My complaints are based on sitting in front of Rhino for 10 hours a day and repeatedly stubbing my toe on the same bugs for the past 10 years. Like I love VSR and am all for getting those features added, but some bugs threaten my sanity…

And no to Grasshopper. I’m not spending another half day getting something to work only to find it’s missing something like blocks or layers or groups (and don’t say Elefront because no it doesn’t).

1 Like

Ok, now we will take your word for it!

1 Like

This is how to make Rhino 30% more powerful for quality NURBS modeling:


Hello Bobi,

this is going in a good direction, Control Point modelling is a great weak point at the moment!

One question, why would you want a “Drag Strength” toggle for each mode?
As a preset? So you dont have to switch it on/off when changing modes?
That would be a nice idea, as Im a huge fan of “Drag Strength”, I actually prefer it over “Delta” in most cases.

What about toggling "Drag Strength with a hotkey? Because in your proposal it feels a bit visually overloaded.

And please dont forget “Direction Locking”, “Degree/Order Change” and “Delta” value!

I would also prefer to have Normal and Tangent as seperate modes. This way CPs can be touched and moved in one click with selection by proximity.

Visual indicators like a gumball could be solution. But that might require selecting CPs first, then either picking a tangent or a normal and modifing them, which is double the amount of clicks. What are your thoughts on that?


And make that a dockable panel or toolbar like selection filter!

1 Like

My proposed “Control point modeling” is in a different direction, as my example above shows. :slight_smile: It’s strictly focused on manual control point manipulation rather than rebuilding the surface’s structure. The general idea is to be able to set custom Drag strength to certain Drag modes so that the user could benefit from using two different settings at once. For example, if Gumball uses the CPlane coordinates and the “CPlane” option in the “Control point modeling” window is not active, then the user will move the selected objects or control points with Gumball with 100% drag strength. However, if the Drag strength for the “Control polygon” setting is active, then moving the selected control points via the proposed tiny arrow handles next to each control point (similar to what Alias uses) will be affected by the Drag strength’s percentage, so that the user will be able to make extremely fine adjustments via those handles while he or she could still use the Gumball for moving with 100% drag strength.

I made my pop-up window based on copying some UI elements from existing Rhino tools, so it’s a bit larger than what I would call optimal utilization of space.

If I was in charge with decisions about the GUI of Rhino, I would definitely look into the possibility to implement a function to auto-hide certain windows that could be easily accessible via simple hovering with the mouse near the left or right edge of the screen. Solidworks uses those for many years and they are a great space saver. For example, its “Solidworks Resources” panel (similar to Rhino’s “Properties” panel) automatically hides when not in use, leaving only several small icons stacked vertically on the right side of the screen. In Rhino, those could be the main Object properties, Materials, Layers etc. Clicking one of those icons, for example, the layers, should open the properties panel with the Layers tab in focus.


The UV setting is not rebuilding the surface’s structure. Its an ICEM thing, where the Surface’s structure and shape remains the same, you can simply add oder remove CPs. Just like the VSR plugin.

I can see the benefit of setting a different Drag Strength for Gumball and for CP-Modeling, just maybe not for EVERY mode. But I really like your concept.

I love the little arrows that you are showing in your proposal!

1 Like

Bobi, you should get McNeel to hire you. In this way you propose and they execute. At least they don’t walk in the dark or take small steps. We need someone to steer development on the right path.
:wink: :+1:


Agreed. Before Autodesk grabs him.


As I mentioned above, a huge part of what makes Alias so powerful is the ability to do fine adjustments while dragging the tiny arrow handles next to the control points set at a very small drag strength such like 1-5%. Meantime, directly dragging a control point (or control points) or other objects will move them at a normal speed (full drag strength).

I’m a huge fan of the “Explicit control” in Alias and I would love to see it implemented in Rhino as an integrated option of the main surfacing tools such like: Blend surface, Match surface, Loft, Sweep 1 rail, Sweep 2 rails and Network surface. In many occasions Blend surface and Match surface make surfaces with very dense control points that anyway need rebuilding with the “Rebuild surface” or “Rebuild surface UV” tools, then applying “Match surface” again (this is like using 3 tools in a row, as well as some analysis tool, which forces the user to spend a lot of time and click 20-30 times in different panels). Everything could be made much simpler and user-friendly by adding “Explicit control” inside those tools. Also, Zebra, curvature analysis and others should be possible to be turned on or off via dedicated checkboxes inside those modeling tools.


I completely agree with you on having better surfacing tool would be perfect and really needed.
I agree that this would make Rhino more powerful but
Aren’t we sure this small arrows aren’t patented from Alias?
If so and AD catches Rhino breaking the patent what it would happen?

I remember Bob telling this to a User meeting years ago speaking about Tsplines…

Keep proposing things but don’t think mcneeliers are dumb :smile: they’re just really sloooooooow.


Other programs use those tiny arrow handles for years, so I guess it’s not a patent of Alias. Of course, every program made them look in a it different way depending on the need. VSR for Rhino 5 also had those arrows, even though they looked way too thick to my liking. :slight_smile: And I also don’t like the green colour they used and I’m colour blind and depending on the shading and camera angle sometimes they appear flashing/changing colour to my eyes. :slight_smile:


In no way I’m leaving Rhino. :smiley: I stick with it for 20 years now (started using Rhino 2 since 2002) despite the fact that I was offered to work with Alias in a few companies. Alias is more powerful due to its control point modeling and “Explicit control” that Rhino lacks at the moment, but my hopes are that the management at “McNeel” will see the huge benefit that those tools could bring for Rhino. :slight_smile:


you pretty much nailed it, explicit control along with cp massaging with drag strength, static zebra stripes along with an isophote shader, which I actually prefer in Alias. Query edit to maintain history across surface creation and matching is also pretty powerful and would be awesome if implemented in rhino.


That’s right. Other people also gave some great ideas in this topic, so I think that “McNeel” should have pretty good idea about the general improvements that Rhino users wish to see implemented in Rhino ASAP. :slight_smile:

Probably the easiest improvement that could be added even in Rhino 7 SR20 or SR21 is the ability to alight the direction of the end handles of “Blend surface” to be tangent to the target surface’s side edges. Currently, the lack of this option forces the user to extend the side edges of the target surface with a line, then use the latter to manually snap the Blend surface’s handles to them. The following images are old and made when Rhino 6 debuted, but they pretty much explain the general issue with “Blend surface” at the moment.

Preserve control point flow is yet another much needed enhancement that Rhino lacks:
Match surface G2 - Preserve flow.3dm (1.0 MB)

Another easy-to-implement option is the ability to untrim a portion of a trimmed surface which is normal to the selected edge(s).


We all hope this.
Thank you for sharing your invaluable knowledge and experience.

1 Like