I appreciate all the ongoing development work on Rhino. I have a specific question regarding constraint-based sketching:
Over the past few years, we have seen some experimental constraint tools in Rhino WIP, but there has not been a clear update on whether a fully integrated parametric sketch constraints system — similar to what is available in traditional mechanical CAD software — will be included in Rhino 9.
At present, Rhino’s constraint tools are limited and not as interactive or robust as the parametric sketch engines in other CAD platforms. The ability to apply constraints (length, angle, perpendicularity, coincident terms) and have the sketch dynamically update as part of the modeling process would be an extremely valuable addition to Rhino’s workflow, especially for users doing precision 2D/3D mechanical or fabrication work.
Could you please clarify:
Is there an official plan for a complete sketch constraints system in Rhino 9?
If so, what level of functionality can we expect — for example:
real-time display of constraint values during sketching
associative constraint editing
parametric updating of sketches with constraints applied
If not, is there a strategic reason Rhino is not pursuing this direction?
Especially designing parts for 3d printing etc. So much easier when you can make quick size changes with a sketch. Often have beginners asking which CAD software to use…be nice to recommend Rhino.
Anyway. Doesn’t seem like there will be constraints.
But I’d also like to hear why McNeel didn’t pursue this any further. Seems like they started then stopped.
Parametric sketch constraints would be gamechanger for our customers who use Rhino for CNC cutting. Most of their parts are simple 2d shapes - think circular gaskets, or rectangular parts with notches - which come in a variety of sizes. Being able to sketch such parts and change dimensions easily would be a huge boost and enable them to keep design work within Rhino.
Almost without exception our customers use Solidworks for design - even though its much more expensive - and Rhino can easily be used for drawing such 2d parts - the parametric sketch functionality is a major reason for that.
Of course us Rhino experts know all of these parametric parts can be trivially designed in Grasshopper - but the thing is that for most of our customers that’s a very different skillset and training cost and far from being as intuitive.
Grasshopper provides amazing expert level capability for parametric design - but it does complicate the entry level and simple parametric design problems - it is these where a parametric sketch system with constraints would shine - and be super valuable for the entry level user - of which there are many!
The project itself never got sufficiently constraint. “Constraints” means different things to different users. The developers asked for feedback that would allow them to define that project better but felt that this feedback never came.
As for the future, the project isn’t necessarily off the table for ever.
If you want to contribute, you can make a new thread with a very specific and detailed example of what you would like to see. Something like “I want Solidworks in Rhino” isn’t going to move that project in any way…
-wim
Thanks @wim I will take up the challenge to provide use cases and problems for constraints, it may take a week or two.
I wonder if the lack of feedback is because it was sought in the forums? Most forum users are more engaged and advanced than the average user and so are more likely to be competent grasshopper users - who would see little value in Sketch Constraints as they could acheive the same and better in Grasshopper?
At least in our field, of the several hundred Rhino users we’ve trained, I can think of only 2 who have ever been active in the forums, and only 3 who have learned Grasshopper (and yes the two groups are the same people). The rest are working quite productively without either. It is these users who I think would benefit most from Parametric Sketch Constraints.
Every time I have shown parametric design to such users they are immediately excited - and immediately overwhelmed by the “complexity”, the Constraints system we briefly saw in the WIP was well on the way to giving these users their cake without having to eat Grasshoppers (i personally love Grasshopper )
but this is just my experience maybe our field is exceptional and other fields have better proportional representation in the forums?
I agree, “don’t just tell us that we want Rhino to have a constraint system like Fusion of Solidworks” is not good enough is troubling, because that is exactly what is missing and what we would like to have in Rhino. Grasshopper is an entirely different paradigm for parametric modeling and I suspect many are intimidated with it. My mind simply doesn’t think the Grasshopper way.
Since the Constraints system won’t be in Rhino 9, could the team consider releasing what’s already been developed as an experimental, “as-is” plug-in? It seems like a huge waste to shelf such high-quality code and effort. Releasing it as an optional tool would be a perfect compromise: it gives the community a chance to provide feedback without forcing the feature into the core software before it’s “ready” by your standards.
Personally, I’m constantly forced to jump back and forth between Rhino and Fusion 360 just to handle simple mechanical parts. It’s not that Rhino can’t do it, but without a constrained sketcher, making iterative changes is incredibly tedious. You have to manually track every dimension and angle, which really kills the design flow. If this were available as a plug-in, it would save many of us from having to switch software for every minor adjustment.
Sketch constraints are fundamentally the same in every software. What you do with the sketch might change, but the foundation is pretty similar: create a sketch that’s controlled with dimensions and other constraints. Which is exactly what it looked like was being planned for RH9 before it was scrapped.
Here’s a basic example. I’ve sketched the following, maybe I had an image I was tracing to get the approximate shape of something.
How do I set the size of those areas now? I first need to measure them because I don’t know what they are, then I need to explode curves, lengthen curves, join curves…I mean there’s just no workflow in Rhino to make sketches that can be easily modified. Compare that to:
So again I’m not really sure what feedback you want when this is basically the same in all software.
As a furniture designer/manufacturer, I do highly endorse this feature as will speed-up tremendously my work.
There was some old plugins for Rhino5 like RhinoWorks (formerly RhinoParametrics) and “Driving Dimensions” (by LEDAS) that unfortunately it is not working anymore in the newest versions of Rhino.
for a sunglasses project, my students use this gauge/jig/guidework (german: Schablone) to find a nice fitting for their nose
(upload://aptZhQ61wLWHquW4YtFP4DNhQhm.png)
and for some students that claim that they know how to use constraints in solidworks or onshape i have this teeth-breaking challenge to set up this socket / leg: only by cones, sphere, planar surface (and one fillet).
.
. @wim - feel free to connect me with the developer(s) that are in charge, and they are welcome to join my classroom remote.
.
my main argument for constraints: students will need to understand the concept, also if they are hired and will change the cad.
i do not see much activity regarding developers / constraints on the linked topics or this one:
Tom, you know I’m your ally on this since the beginning.
My feeling is that constraint is a pure speculation and requies a lot of work to make it working.
Is a challenging project that can break multiple other commands due to the way history is handled and to give enough depth to it. I feel McNeel decided not to invest on this.
We gave plenty of examples where a parametric sketch could help but this is the result.
we have to hope in another Joshua, could we CNTRL+C/CNTRL+V him ?
If both (1) constraints are actually developed for Rhino 10, and (2) you are thinking about the final released first version, then yes, 4-5 years would be the earliest.
-wim
(i personally love Grasshopper 
)
