mathematicians in comparison.

[goccia]

catia allows you to design objects both in parametric modeling (gsd) and in explicit (non-parametric) modeling more direct, allowing you to work also on the individual control points that does not seem possible in gsd environment.
both generate surfaces, both transportable in step-type formats.
I wonder: is there a difference between the two geometries, mathematically speaking? i.e. gsd generates better mathematicians for the proposition to video of the surface and above all it ends an industrial processing with very low tolerances?
I also noticed that modelers free forms different type rhino (in the face of greater laboriousness) allow to create geometries with a high control of continuity: is the mathematics present in rhino more poor, if so can be said, of a freestyle or gsd generation? or is it just personal preferences?
These questions arise from the fact that I am interested in producing clean surfaces of high quality, read from the mathematical point of view (reducing to the minimum the degrees of curvature and the points of control for example) for future purposes of design and industrial production.

 

[Lettore espatriato]

goodwill drop,
I wonder and ask you: if the modeling difference between a "rhino win" and an icem (perhaps the comparison would be more correct with alias, I think), was all about managing a g3 on g2 connection, do you not think that tens of style centers and technical offices in the world would already have gone to cheaper programs? Of course, it is also true that large organizations are, in some cases, "lent" to react or tied to "foreign" evaluations, but many times the choices are made also because of real skills, modeling or "global" management of the "product", better.
then, I saw beautiful patterns made with rhino, with catia, with alias, with autocad and with the tecnigraph, but it is not so much the "matita" that makes the difference, is the "manic".
class a, intended as a modeling of surfaces, did it in the 60s with the shape plans to tecnigraph (perhaps someone, hidden, still today) and the Baptist masters. It doesn't seem to me that jaguars or ferraris or pilasters were all boxes with sheet metal folded at 90°. But maybe I'm wrong.
this to say that the process of class a or similar is a long and expensive process, which in order to preserve all the advantages of a certain production process, must provide adequate tools upstream and downstream of what the individual designer does.
It is useless to check with zebra to verify a radius in g2 or g3, if then, to make an example, the mold you go to realize is made more "bigly", removing or adding 0.5 mm here or there, with many manual processing to fix other imperfections, perhaps also due to a carefree "mechanical" design. or if the cam program that gives "meal" to the milling, approximates the perfect class surfaces to arc sequences (to simplify).
Keep in mind that, as I said, the product falls, must be seen as a whole . I do not doubt that there are single products that do better a certain thing, who the connection in g3, who the toothed wheel, who the pipe with holes along the axis. the advantage for a company is to have a single product that, on average, does a good job on everything and everything. with v6, so much to say one, they went even more towards the product "only" for everything: cad, cas, cae, etc...
nx, with subdivision surface, went, as had already done catia (which had already ima, besides icem) towards alias, to try to fill the "initial" phase of the style.
If your goal is "only" to make beautiful surfaces, it can be that rhino perfectly meets your needs.
then, as always, I'm not experienced in anything, so don't take my "ragionaments" as "colated gold". ...

 

[goccia]

that the development of the product does not depend on the only design is there and I am aware of it, but if my piece is to make a perfect surface because it is what I sell everything else does not matter, they will be made of those who did not use the most suitable tools for production. So, returning to mathematicians? What about me?

 

[goccia]

But perhaps I have an answer. (ii) the production process corrects the errors of the cad.

 

[gianni55]

catia allows you to design objects both in parametric modeling (gsd) and in explicit (non-parametric) modeling more direct, allowing you to work also on the individual control points that does not seem possible in gsd environment.
both generate surfaces, both transportable in step-type formats.
I wonder: is there a difference between the two geometries, mathematically speaking?

Hello, drop,
what the reader said is right, I also consider:
1) between a parametric mathematics and an explicit one, there is no difference if you generate it the same way.
2) at the time when you carry out and (import) the mathematics, in step, iges or other formats happens a "decadency" but usually is minimal, to this you must add the worsening of the rings of the production chain (pu, milling, materials etc.), that do not correct the errors of the cad but increase them (for this they are born of the master equipment to keep under control the drifts).
3) if you generate mathematicians with different software, while using the same methods of constructing and at the same level of commands, you can have diversity, as it increases the complexity of geometry (a plane, a sphere an extrusion, or sweep).
4) good level cads allow you to do everything, but as always it is as important as you know how to use them and the militias you have learned with the experience, without those even the best software remains to the pole.
5) beyond the best software for this or that, in the real world you will have to use what the customer imposes on you.
6) for class surfaces a: alias and icem surf are the most used (and there will be a reason), if you have to model a spit for a dentist's armchair and there are no quattrini, it's okay.

as for photography, better nikon or canon?....... better a good photographer.

Hi.

 

[goccia]

This last statement seems a bit to contradict the first paragraph 1.

 

[goccia]

I read wrongly sorry, thank you both, you were clear.