Solids, nothing more than surfaces....

[ptc-4864603]

Correct me if I'm wrong on this.

I've worked with a lot of surfacing programs in the past, as well as parametric. We have an argument every few months about "solids" and "surfaces". I'm having a hard time getting the guys that have worked in Proe for years to understand that there is no such thing as a "solid" model unless you are running FEA or using a voxel based program.

A solid model is nothing more than "watertight" joined surfaces defining a closed volume. When you perform cuts, extrusions, etc. Creo is still caluculating boolean operations on surfaces. Explaining U/V paramaters falls on deaf ears. I won't even get into texuring or decals.

So is there a way in Creo to show the isocurves of "solids/surfaces"? that's the only thing I can think of to prove my point. Any other suggestions to get my point across?

[TomD.inPDX]

Ryan, it is a moot argument simply because the difference is the background operations and definitions -within- Creo that makes the difference. In other words, yes, the solids also have surfaces but the way things interact with the two "types" of features is different. So it is all made up by PTC to begin with.

[dschenken]

The only solid models are those you can pick up and throw, and even then there are huge gaps between the electrons and the nuclei.

As software goes, it's just a matter of what the interface appears to do.**

If it maintains a bounded volume description and lets any changes to that result in bounded volumes, then that's a solid model. If you could delete one face and keep the others with the loose edges dangling, then that's not a solid model.

The underlying description may or may not be explicit surfaces - look at POV-Ray, which includes solid primitives that never expose the individual surfaces to manipulation and Blender which can have water-tight geometry, but where the only surfaces are made up of bounded-linear-edge mesh elements. Even voxel programs aren't really solid - it's just a list of X,Y, Z cube origins and side-lengths.

PTC Parametric uses a B-Rep scheme that is built from non-solid geometry, usually 2D. It does so with a procedure in such a way that generates vertices, edges, and bounding surfaces that are accessible/identifiable geometry results.

**When you watch digital-TV there aren't actual people on the screen, just a software recreation of some point of view based on a program that tells which pixels to change color and when.

[ReinhardN]

Ryan maybe right. There is a TPI about that:

https://www.ptc.com/appserver/cs/view/solution.jsp?n=126505

[ptc-4864603]

all these explanations make sense to us

but explaining it to ProE vets, that only know "surfaces" from Wildfire or Creo, I get looked at like I'm a 3rd grader talking to a college grad. lol

Reinhard, that is a gem of a find for this. Substitute "quilt" with "surface" and it still is right.

I'm probably pissing in the wind with trying to explain it. The only thing I can think of to explain it is to show how the "surfaces" of the solid get reparamaterized (is that a word?) when it gets edited showing the isocurves.

I know in the end none of it matters to the user, it's all the man behind the curtain.

thanks for confirming I'm not a third grader

[ptc-4864603]

tthinking more about it, is knowing about surface construction and defining 3d geometry not even relevant anymore in the future of solid modeling in this industry?

[James62]

Yep, Ryan, It's like talking about some invisible things. It's cause Creo takes care all of that for the user, so they don't really need to know all of these things related to surface theory.

Well, I'm also coming from Rhino. Just to explain the difference between solids and surfaces... The surface normals of a solid body always point outwards, while surface normals of a closed quilt can point inwards as well as outwards.

Btw, for isocurves try this config: mesh_spline_surf yes. It shows isocurves on more complex surfaces only, which works well enough at least for me.

[neomechanikos]

How about the option to "cap" the surfaces on a cross-section (creo 2). Why would the software have to cap anything if the models were actually "solid", hmmm?

[TomD.inPDX]

For the most part, people seem to work in a solid environment with occasional surface geometry to create special "forms" or "cosmetic" shapes which are later thickened.

However, working with quilts can greatly simplify the number crunching behind the scene where you can later thicken and otherwise solidify the entire model once it is fully defined with quilts.

If you dig deep enough into the underlying algorithm of the sheetmetal module, you will find the same; a master surface and a thicken to control the distortions along bends.

It is best to simply consider surface modeling and solids modeling as two modules within the core Creo package. each have their usefulness and the the ability to mix them is a staple of the parametric modeling landscape today.