Pros and cons of model design programs

[Laurence Finston]

Originally Posted by mbauer

Still doing my drawings using 2D in AutoCAD 2006, switching to 2019 and feature based 3D modeling is a pain!

The last time I used AutoCAD must have been 1993 or so and I was using it for 3D. That's the reason I started using it in the first place. I got the impression that Fusion 360 is basically AutoCAD with CNC and a lot of other extras.

3D certainly involves a certain amount of extra work but it's nowhere near as much as trying to do two different perspective views of the same object by hand. I've done perspective drawings by hand and I'm glad I learned how to do it, but I don't plan on ever doing it again.

[Laurence Finston]

Originally Posted by mbauer

It never goes on-line now, bought three exact backup machines when Microsoft quit supporting 7. If I have any issues, I use the two spares to backup the original, simple to switch the machines out and keep drawing. Hoping they (3-machines) last for the next 20-years.

My great-uncle did something like this with his Model T, except that he used the spares for parts.

[Laurence Finston]

Originally Posted by Laurence Finston

With computers, there are only ever numerical solutions since they haven't invented a computer that can perform integration (or many other things) yet.

Not true. If you've got a formula, of course you can implement an analytic solution. In my program, GNU 3DLDF, I usually use both implicit and parametric equations for geometric figures, if they exist. I use the parametric equation to find the points on the figure when creating it and the implicit equation to test whether a given point lies on the figure or not. The latter is needed, for example, for finding intersections.

[FRD]

I'm using autocad 2023..

[Tyler Durden]

[QUOTE=mbauer;788667]Wondering if Fusion 360 shows you what the ARC length is if you select it and look at the property box for it?

Yes, it does have that capability. Its actually fairly easy to get the length and measure angles, find diameters, radius points, etc.

[Tyler Durden]

Originally Posted by Laurence Finston

The last time I used AutoCAD must have been 1993 or so and I was using it for 3D. That's the reason I started using it in the first place. I got the impression that Fusion 360 is basically AutoCAD with CNC and a lot of other extras.

3D certainly involves a certain amount of extra work but it's nowhere near as much as trying to do two different perspective views of the same object by hand. I've done perspective drawings by hand and I'm glad I learned how to do it, but I don't plan on ever doing it again.

I learned how to do one and two point perspectives years ago in HS when they still had shop classes. I have a book which I acquired a long time ago by William Garden on yacht designs that had a chapter on projected perspectives working from two drawings to work up an accurate 3D drawing at any perspective length and angle. I believe the aviation artist Keith Ferris also had a section on developing projected perspectives which he used extensively in his paintings.

I managed to learn the basics of it but it is extremely time consuming, and anyways CAD was just starting to really go mainstream and was replacing the mechanical drafting I had been trained on. I was forced to part with my drafting table for lack of space...

[Laurence Finston]

Originally Posted by Tyler Durden

Yes, it does have that capability. Its actually fairly easy to get the length and measure angles, find diameters, radius points, etc.

If it will do this, then you can make developments or unroll your model, if you prefer. If it implements AutoLISP or has an interpreter for some other programming language, you could write a program to do this or at least some functions to assist in doing this.

It might not exactly be easy, but it would certainly be possible.

[Tyler Durden]

Originally Posted by FRD

I'm using autocad 2023..

Do you make your models as a solid model first? Or as a surface model? How do you unroll the models?

I don't use NURBS for modeling since my dabbling in CAD is usually working off of prints with dimensions which NURBS is pretty much useless for. Most of the time I am creating solid models which are then used for toolpaths for machining or converted into models for 3D printing. The workflow for turning a CAD generated model into a set of parts for a paper model is where I am stuck since there doesn't really seem to be a clear-cut process and every modeler seems to be different in their approach.

[Laurence Finston]

Originally Posted by Tyler Durden

I managed to learn the basics of it but it is extremely time consuming, and anyways CAD was just starting to really go mainstream and was replacing the mechanical drafting I had been trained on. I was forced to part with my drafting table for lack of space...

The real trick is three-point or more accurately oblique perspective, one-point being linear and two-point being angular perspective. Linear perspective (as the umbrella term) is "just" an application of projective geometry and inferior to the way it's done in computer programs, which is by means of homogeneous coordinates (four coordinates per point), 4 x 4 matrices and matrix multiplications.

Linear perspective is a difficult way of doing something that is very easy to do in 3D, like the way sundials are laid out (another application of projective geometry). Doing it by hand is tedious, time-consuming and error-prone and when you're done, all you have is a single view. If you want another one, you have to start again from scratch. Glad I did it, never want to do it again.

Too bad about the drafting table. I always wanted one but now I find I like drawing on a horizontal table because I always turn the drawings, which would be hard to do on a slanting surface.

[Laurence Finston]

Originally Posted by Tyler Durden

I don't use NURBS for modeling since my dabbling in CAD is usually working off of prints with dimensions which NURBS is pretty much useless for.

That's not what I meant. Implementing curves as NURBs is standard for 3D graphics programs, so straight lines, circles, ellipses, etc., are actually NURBs from the point of view of the software. Many programs allow you to explicitly define NURBs but even if you define a curve by, say, defining a number of points on it and having the software draw a smooth curve through it, it will be a NURB.

Originally Posted by Tyler Durden

The workflow for turning a CAD generated model into a set of parts for a paper model is where I am stuck since there doesn't really seem to be a clear-cut process and every modeler seems to be different in their approach.

Most parts for manufactured objects are either rectangular, polygonal, cuboidal, circular or spherical anyway, so it's not too hard to make sets of plans for most objects. Things like gear teeth which are made using other curves are often flat, so that's no problem, either. For shoes and clothing, there are traditional methods and I've never seen a mathematical treatment of this subject.