Email this sample to a friend

THE MODELING OF COMPLEX OBJECTS is something that every 3D professional is bound to be required to do at one time or another. While many of the modeled elements encountered in architectural visual ization are structural forms that are often square and linear, there are times when buildings, walls, furniture, props, and architectural details will require either a bit more detail, or that they are nonlinear by nature. In many cases, the way to create this detail is through the use of subdivision modeling. Building a low polygon base mesh that will later be subdivided through the use of the TurboSmooth or MeshSmooth algorithms is a powerful way to create complex models with high levels of detail and smooth-flowing surfaces. I am sure everyone reading this has modeled and used Turbo-Smooth many times. I am also confident that there have been many occasions when your model has turned into something completely unexpected when adding the TurboSmooth modifier. The ability to create intricate models that will subdivide properly and create consistent and predictable results is what makes a modeler's skill level advanced. With this in mind, I am going to outline some of the key factors that make poly-modeling and subdivision surfaces more predictable, while also improving the quality and speed by which you produce your models.

Poly-modeling Theory

The quad polygon is the basic building block for any well-built model that is going to be subdivided. A quad polygon does not necessarily need to be square in shape, although a majority of the time it will be. Simply creating a 4-sided polygon when building a model is not going to guarantee that your model will look clean when TurboSmooth is added. There are other factors that need to be considered: polygon size, spacing, and polygon flow all impact the look of the final product when subdivided. Having long quad polygons next to short quad polygons can quickly turn a model into something that doesn't react as expected when subdivided. An even quad polygon size and spacing is not only important structurally but also will make the process of texturing your model a much more predictable process. As a general rule, using all quads for subdivided models is a good habit to get into; however, I am not strictly opposed to having non-quad polygons in my meshes. There are times when having a non-quad polygon is acceptable, especially when modeling inanimate objects. Non-quad polygons are called NGons. They have a vertex and edge count that is greater or less than 4.

Previous Page Next Page Page 2 of 27