Full Version: A free subd to NURBS solution

Show messages:  1-17  18-37  38-48

I was under the impression that there still some remaining issues about surface continuity at extraordinary vertices. Is that not the case anymore?

- Michael

---

@You ".. remaining issues about surface continuity at extraordinary vertices. Is that not the case anymore?"

Well...Maybe you're right but for what I've tested so far I didn't see any issues lately. But I'm not the best person to check this kind of "math" things.

Maybe it's a better idea to ask Max or other "math" guys here :)

Thanks.

---

Hi Marco,

> Maybe you're right but for what I've tested so far I didn't see any issues lately.

Could you post a couple of the tests?

- Michael

---

@You : "...Could you post a couple of the tests?..."

Hmm...nothing serious model right now....only some old practicing but I have to search and I will post some 3dm file.

Maybe if you ask PaQ you will be provided with much more "heavy-duty" and reliable stuff :)

---

Michael please try this and see if you see something wrong with the "subDiv" elaboration.

I've made with V3. It's a very simple use-case.

The red mark id where I think there's some strange "flow" but i don't know. Please check the entire model if you can.

Hope this can help :) This is the best I can do right now.

This is the 3dm file : http://take.ms/rhLTD



Ciao!

---

Hi Marco, well it looks pretty good. You have to do a pretty detailed reflection analysis to see any problems. (screenshots from Rhino Zebra analysis)

Some things are pieces that are tangent continuous but not curvature continuous like this:





So the question is - is that good enough for manufacturing? Maybe yes for some things and maybe no for others.

There are also corner spots where it's not quite G1 though:





Those are probably bigger issues.

So no doubt it's useful for a whole lot of cases but finicky industrial designers are going to be upset about those little details.

- Michael

Image Attachments:
subd_continuity1.jpg  subd_continuity2.jpg  subd_continuity3.jpg  subd_continuity4.jpg 

---

Thank you very much Michael for this detailed analysis. Much appreciated!

Well...I was already aware of "continuity issues" with this kind of stuff but as I said today I don't have enough spare time to dig into it :)

But...Do you think that the G2 continuity offered by the Moi's "Blend" tool is enough to deal with this kind of things ?
Or we (and Max) should need a more extended set of tools to generate this kind of "continuity" SubDiv ?


@You : "...So no doubt it's useful for a whole lot of cases but finicky industrial designers are going to be upset about those little details..."

I understand.

So...this mean the end of our (or my) dream to see one day the "subDiv" script fused into the Moi's core engine ?

Thanks.

Ciao!

---

Hi Marco,

> So...this mean the end of our (or my) dream to see one day the "subDiv" script fused into the Moi's core engine ?

With all due respect to Max's work, i would like to see a standalone version of a subd to NURBS solution.

Why, you never know what will happen in the future with Software solutions who work only for app x,x, or z.
Best example is T-Splines for Rhino, which Autodesk killed and then included in Fusion360. And if done right,
i would also assume that the price of MoI would increase drastically. See ViaCAD Pro when you have to buy
their subd to NURBS solution as an extra plug-in.

The person(s) who would take care of this problem *as a standalone product*, with crease edges functionality would
IMHO win the jackpot, in the 3D Industry.

Best regards
Stefan

---

Hi Stefan.

I understand your point of view...

But...I was thinking only to a more "simple and limited" tool.
I mean the porting of the current Max's subDiv scripts, with a some degree of improvements, into the Moi's core tool set.

I was not thinking to a totally brand new "area" of Moi :)

Thanks for sharing.

Ciao!

---

Patents might be a consideration?
- Brian

---

Well, the T-Splines folks are not the only ones who created such technology.
The IntegrityWare folks created a subd to nurbs solution (patend pending) too.
ViaCAD has subd to nurbs tech also. And if you look at Github there are
also things called T-Splines.

Regards
Stefan

---

I like Marco's idea of "Porting of the current Max's subDiv scripts, with a some degree of improvements, into the MoI's core tool set."

++ Plus ++

Stefan's idea of "Crease edges functionality."

Sell it as a MoI option. I'd get my wallet out for that! Then I could try designing more cars :)

Minor surface continuity issues are acceptably I think for software designed for quick & easy visualization. Professional designers are likely using something much more expensive than MoI.

Ed Ferguson

---

My patent comment was merely cautionary.
- B

---

PaQ
>>I really hope/pray Max will find the time and will to work on the surface simplification from his amazing subdiv script.
As I wrote earlier, I have a plans to implement this feature to Subdiv script. But at the moment it's difficult to do using current MoI Javascript API.
The current version of the Subdiv script saves an output data as a text STP file and then import it back to MoI. I can write a more complex stp export tool for stitched surfaces, but in my opinion it's waste of time.
Lets wait for a new js api functions in MoI v4.

Stefan
>>The person(s) who would take care of this problem *as a standalone product*, with crease edges functionality would IMHO win the jackpot, in the 3D Industry.
This script doesn't use any specific MoI functions. It can be easily ported to node.js project which can convert obj to stp. My script is free, so everyone can do it.
I'll think about this idea. But anyway it will be an opensource project.

Marco, Ed
>>I mean the porting of the current Max's subDiv scripts, with a some degree of improvements, into the Moi's core tool set.
Do you really need it as an internal tool? :) Everyone can use it right now for free.

Michael
Could you add this Zebra analysis to MoI? It will be very handy tool.

---

Hi Marco,

> But...Do you think that the G2 continuity offered by the Moi's "Blend" tool is
> enough to deal with this kind of things ?

Well it's kind of the main point for sub-d in a CAD program to have the continuity automatically taken care of so you don't need to do blend style surface modeling.


> Or we (and Max) should need a more extended set of tools to generate this
> kind of "continuity" SubDiv ?

I don't know, I think it should be possible to have G2 continuity but maybe it needs a different approach using numerical analysis methods.


> So...this mean the end of our (or my) dream to see one day the "subDiv" script fused
> into the Moi's core engine ?

No not necessarily. But there will be plenty of work involved because sub-d modeling needs its own distinct workflow.

- Michael

---

Hi Ed,

> Minor surface continuity issues are acceptably I think for software designed for quick & easy
> visualization. Professional designers are likely using something much more expensive than MoI.

This is true, but at the same time I am very frequently getting feedback from various kinds of professional designers who like to use MoI as well.

If it's possible for something to cover both the "quick & easy" _and_ "professional designer" cases that makes it more of a slam dunk to work on it.

- Michael

---

Hi Max,

re:
> Could you add this Zebra analysis to MoI? It will be very handy tool.

It is something I want to add, it's a little finicky that you have to generate a very dense display mesh for it to work very well though. So it's not just a pure display mode addition it needs to involve mesh management in some way as well.

- Michael

---

Hi Max.

Thank you very much for your detailed reply!

@You : "...Lets wait for a new js api functions in MoI v4..."

This is exactly what I'm waiting for :)

@You : "...Do you really need it as an internal tool? :) Everyone can use it right now for free..."

Oh yeah! It's amazing that all your scripts are free, especially SubDiv, but when I say "internal tool" I'm referring to something
that should leverage more heavily on some new V4 Api and that should be written in the same language Moi was written to.

Ciao! We'll catch up.

---

Ciao Michael!

Thanks a lot for sharing your thoughts.

@You : "...I think it should be possible to have G2 continuity but maybe it needs a different approach using numerical analysis methods..."

Very interesting...but...I'm out of my league here :)

Anyway, I think that what you're talking here about is far beyond the things that we actually can do with the current Moi's Api.

Thanks again and have a nice day.

We'll catch up.

---

P.S.

@You : "...I don't know, I think it should be possible to have G2 continuity but maybe it needs a different approach using numerical analysis methods..."

...unless you're talking about some sort of external standalone tool that should analyze for example the data of the same STEP file
that Max is currently using to perform his SubDiv calculation...

---

Show messages:  1-17  18-37  38-48