Hi Michael,

>>The other kind of thing that tends to be difficult to apply a quad-only topology<<

So you want my example to be only quads,... OK, I will do that if you can output a mesh of this model from MOI with no tri`s


- Steve

Hi Steve,

> So you want my example to be only quads,...

Do whatever you wish with your example - your previous one that you posted here: http://moi3d.com/forum/index.php?webtag=MOI&msg=3869.48 seemed to be constructed with all quads, so I thought that was what you were going for.

But if you produce one with triangles in it I wonder if you will consider it to be a "bad topology" yourself though?

- Michael

Hi Michael,

To me, tris would only be classed as bad topology if they where isolated, surrounded by quads/terminating at edges forcing bad flow. In the model where you have cut partial spheres, then you have poles, yes, the poles can be terminated with quads, but the quads would need to be extremely small due to you wanting correct reflection rather than just stopping rendering artefacts.

As I do not haver a lot of spare time, I will just make a section of the model from quads.

Does that mean you will not be making a mesh output from MOI with no tris? :)


- Steve

Hi Steve,

> Does that mean you will not be making a mesh
> output from MOI with no tris? :)

Well, I'd like to but it probably won't be happening anytime too soon... I've looked into various techniques for possibly doing that, and it would be possible to make something that worked ok for some simple kinds of cases but very difficult to make one that would be very robust.

- Michael

Hi Steve,

> due to you wanting correct reflection rather than
> just stopping rendering artefacts.

A reflection that looks wavery and jittery _is_ a type of rendering artifact. But of course you won't notice it if you do a rendering without any reflection in it.

Reflection tends to substantially amplify any problems like little wiggles in geometry or irregularity in normals.

It's just another area that you don't have to worry about if you render with the NURBS vertex normals being used.

- Michael

Hi Michael,

In real world, most reflections are not perfect as most surfaces are not perfectly smooth.

As for edges. I do know about adding insets/extra edge loops for controlling smoothing. The good thing with that is edges can be made to look natural/not "knife edge sharp" without added geometry, although if needed, hard edges can be added to force termination of smoothing at boundaries.

 

- Steve

Hi Steve,

> In real world, most reflections are not perfect as most
> surfaces are not perfectly smooth.

Yup, but that's one of the cool benefits to NURBS - you can do things like easily create spheres that are perfectly smooth. Then if you want them to not look smooth you put stuff like a texture on them, it gives you more options that way.

It's a whole lot easier to add in imperfections if you want them, than it is to remove imperfections if you don't want them.


That's kind of the whole reason for existence of NURBS in general - so that you can make a kind of idealized geometry design for your object. It's why manufactured objects use NURBS geometry for the model definitions for the most part.

I'm not sure if I really understand your point about "in the real world".... - surely you don't think that every kind of rendering is only about trying simulate the imperfections of the real world, right? There are various kinds of rendering styles, some of which are about making illustration type results.

- Michael

Hi Michael,

>>I'm not sure if I really understand your point about "in the real world"....<<

"In real world", = not a computer simulation/render, actually outside in the world :)

I have managed 15 minutes to make part of the model, will finish the section off (if time) when I get back from city centre.
Monday morning blues on a Tuesday :)

- Steve

Hi Steve,

> "In real world", = not a computer simulation/render,
> actually outside in the world :)

Uh ok, but the discussion here is actually about rendering...

You said:

"In real world, most reflections are not perfect as most surfaces are not perfectly smooth."


Was I supposed to interpret this as just some random fact completely disconnected to the discussion at hand?

I took it to mean something like: In the real world most reflections are not perfect, so it's ok for them not to be perfect in a rendering as well.

- Michael

Hi Michael,

I wish I had as much spare time as you to continue this fruitless discussion on what "real world" is, but I have not.


- Steve

Hi Steve, well I understand what the real world is.

What I don't follow is the purpose behind you bringing that up - were you trying to make some correlation between the real world and what a rendering should look like, or not?

If your comment had nothing to do whatsoever with the actual discussion, it might be better to just leave it unsaid.... It's pretty natural for someone to assume it had some kind of intended relevance to what was being discussed.


I have to admit, I'm pretty confused about a lot of things in this thread, like what is the motivation to do all this work to reorganize all the polygons when the ones that were just directly saved out from MoI with the accurate vertex normals on them generate an already perfect looking rendering.

I guess some of these things will just remain mysteries to me.

- Michael

Being a newcomer in these forums I didn't want to jump in, but after following this thread for a few days let me just praise Michael's composure in a discussion that might have stopped at post n.85 (or even before).
Reading some posts I had the feeling that people is speaking different languages or just refuse to understand. Meh!

Hello Michael,

>>I have to admit, I'm pretty confused about a lot of things in this thread, like what is the motivation to do all this work to reorganize all the polygons when the ones that were just directly saved out from MoI with the accurate vertex normals on them generate an already perfect looking rendering.<<

As I put forward earlier in the thread, I make models with quads due to that is what is wanted. I do not turn around to a customer and tell them "you really dont need that" and send them a model with n-gons/tris, then say, well it renders perfectly. If I did, then I would lose the work/contract.


- Steve

Well if the poly flow is so important then just use a poly modeler, where the toolset is all about pulling and pushing vertex.

Hoping that a nurbs package like MoI will output full quad sub-d ready mesh is just science fiction :P.

The problem with MoI meshing is that sometimes, the result is so close to a "perfect" topology , it gives the illusion that with some extra work (from Michael) the model can be used as a subd cage.

Hello PaQ,

>>Well if the poly flow is so important then just use a poly modeler, where the toolset is all about pulling and pushing vertex.<<

Some models are easier for me to create in MOI, and then cut them up to give what I want. So why would I be pushing vertex around when I have no need to? I look for the easiest solution for the output I want, not be told what tool to use that may then make the job harder. Maybe just me.

>>Hoping that a nurbs package like MoI will output full quad sub-d ready mesh is just science fiction :P.<<

There is certainly little hope of that if a mesh cannot be created without tris. But I have not put forward an expectation of such.


>>The problem with MoI meshing is that sometimes, the result is so close to a "perfect" topology , it gives the illusion that with some extra work (from Michael) the model can be used as a subd cage. <<

Now you are just being silly.

I will only make one more post to this thread, and that will be the result of my render test, once I have the section done.

All have a nice day.

"There is certainly little hope of that if a mesh cannot be created without tris. But I have not put forward an expectation of such."

There is an option to output quads and ngons. Not a triangle in sight.

Why sub-d a mesh created from nurbs? Sub-d is to create the kind of curved surfaces that are already in the nurbs model.

This thread has me baffled. The only reason to use all quads is if you are doing character animation and skinning and need to control where the bends in the mesh occur. Even then triangles usually work if they are in the right place.

Nurbs modelling is not the tool to use to create character skins.

>>There is an option to output quads and ngons. Not a triangle in sight.<<

No there is not. There is an option to output "N-gons" which will produce an output of N-gons, quads and tris. Even Michael stated earlier that it is not possible to output without tris.

Well my last post.

This is a section of that model.

Render of vertex normals



Render of smoothing.

Hi Steve,

> No there is not. There is an option to output "N-gons"
> which will produce an output of N-gons, quads and tris.
> Even Michael stated earlier that it is not possible to output
> without tris.

When you use "Output: N-gons" it outputs the underlying n-gons that were created by the refined surface intersected with the trimming boundaries, without doing any further tessellation of them into triangles.

The only way that you will get a triangle with that option is if the underlying surface fragment is actually a triangle itself. For example if you create a triangular plane, then that will create a triangle in that kind of situation.

But on any typical CAD model that will be pretty rare.

The n-gon option produces polygons that essentially follow the same topology as the original NURBS model - it has quads that follow the UV structure of the NURBS surfaces, and those quads are trimmed into n-gons where they intersect the trimming boundaries.

So for example in a case like the original block where there aren't any triangular trimming boundaries, absolutely no triangles are created:



If you have some utterly bizarre goal of wanting to have no triangles in the output mixed with n-gons, even when a even when a triangle is the most accurate thing to be created to fit on a triangle in the original CAD model, then I would suggest creating such meshes by hand in a polygon modeler - that will allow you to custom make the polygon layout to follow whatever particular odd goals that you might have...

- Michael

And just to clarify:

> Even Michael stated earlier that it is not possible to output without tris.

This is not accurate as a blanket statement. For example the block model posted earlier in this thread can be output without any triangles, here it is again:




If your model actually has a triangular plane in it, then it will naturally generate a triangle in the polygon output there as well, since a triangle is the best kind of polygon to fit on a triangular trimming boundary.

- Michael