WHY ARE 3D MODELS MADE OF POLYGONS?

3D models typically start with the creation of a mesh. Most 3D software allows for a mesh to be viewed in a wireframe mode, which allows the user to see the lines and vertexes that create the 3D shape. The mesh contains important data that computers use to properly render the 3D model. Thus, polygons become a natural necessity due to the nature of how meshes are created and used in real-time. This is also an advantage 3D has over traditional art forms because a mesh can be animated and changed in real-time non-destructively.

WHAT ARE POLYGONS IN RENDERING?

A 3D mesh is a net of polygons, connected by lines and vertexes. Alone, a polygon is just a 2D shape made up of at least 3 vertexes and lines, also referred to as edges. These points joined together create faces, often referred to as triangles, quads or N-Gons depending on how many vertexes and edges form them. In 3D modelling, N-Gons refer to any polygon that has more than four vertexes and edges and should be considered undesirable. Good topology practice recommends the use of tris and quads only for the cleanest results.

WHAT IS POLYGON MODELLING USED FOR?

Polygon modelling is considered the standard for 3D modelling due to its adoption in nearly every commercial 3D software. It is used to create 3D meshes with accurate and deliberate topology. Polygon modelling is preferred in VFX and gaming especially. Because gaming often requires many assets to be rendered in real-time at once, developers must be mindful of polygon counts in relation to hardware constraints. Different consoles and PCs will all have varying limitations as to how much they can handle. Thus, 3D artists who work in these spaces, prefer polygon modelling due to the heightened control they have over each model.

What is a polygon in 3D modelling?

Polygons are 2D shapes that are used to create 3D meshes. Did you ever make those paper cutouts that turned into shapes like pyramids or cubes in school when learning about 3D shapes? Well, in polygonal modelling, a mesh is just like that.

Explore Substance 3D

3D polygon modeling of a bird sitting on a tree

Polygons are fundamental to 3D modelling.

Every polygon is formed by at least three vertexes and edges coming together to create a closed shape. Three vertexes or points, makes a triangle—tri for short. Four of them makes a quad. It’s industry standard to only use tris and quads in a mesh as much as possible. There is a term for polygons that have more edges than quads: they’re called N-Gons and really should be avoided because they are more likely to cause artefacts in your textures and materials.

Pieces of a Polygon.

Understanding what constitutes a polygon will help you better manipulate and edit them while modelling. Let’s look more closely at the pieces of a polygon and where they fit in a 3D model’s mesh.

Vertex

A vertex is the point where two edges meet and join. If you move the position of a vertex, you will also affect the position of the two edges simultaneously.

Edge

An edge appears in modelling programmes as a line. An edge has two vertexes, one on each end. When at least three edges join at their vertexes, a polygon is formed: specifically, a triangle.

Face

A face is the flat 2D shape created when a polygon is formed. Many polygonal modelling software will allow you to select a face specifically and change its position editing your mesh. Moving a face would of course, simultaneously move the edges and vertexes that form it.

Create any shape imaginable using polygons.

To illustrate a simple example of polygons and complete 3D models, consider the construction of a basic cube. A cube has six faces, 12 edges and 8 vertexes linked together to create one object. With these basic principles, CG artists can use polygons to create any shape they’d like.

If you look at the mesh or wireframe of any 3D object, you will see the many polygons wove together to create that shape, from a human face to something born wholly from their imagination.

The difference between high-poly, low-poly and retopology.

When it comes to polygonal modelling, there are a few key industry terms you should be aware of.

High-Poly

A high-poly model refers to a mesh that was created with extreme detail and usually no regard for the final poly count. Often, high poly models aren’t designed with polygons at all. Typically, they are sculpted using voxel-based 3D methods.

Think of a single voxel like a pixel in 2D art. Each pixel shops colour information. A voxel is no different. It’s a single point that shops relevant 3D information and each has a place in a three-dimensional grid.

Using these methods allows artists to sculpt 3D objects in real-time. When complete, they can export the 3D model as a mesh, in which case, computer algorithms take care of the polygonal modelling. The results are often impressive, but bloated when polygons are concerned. Hence the name, high poly.

Low-Poly

In many industries, having 3D models with low numbers of polygons is desired. Often the term low-poly is used to describe a model that not only has a small number of polygons, but has a mesh that has been recreated or optimised to lower its poly count.

Retopology

This brings us to retopology: the process of turning a high-poly model into a low-poly one. There are several methods and various software may even have specific tools available to aid the retopology process. For many artists, this is a necessary step in their workflow.

Most commonly, artists will simplify the topology, but bake the more complex, high-poly version onto the objects surface textures. This way very little detail is lost, despite greatly optimising a model’s performance.

Hyper-realistic 3D rendering of bird using 3D polygon modeling

Image by Olivier Beaugrand.

style

Center, grid width 8, static links

Simple polygonal modelling techniques you can use.

Here are basic techniques artists can use when creating polygons for 3D modelling.

Extrusion

Extrusion is a basic modelling tool that enables artists to select a piece of a mesh, usually a face or an edge and extrude it inwards or outwards. This causes an extension of the original shape or can literally tunnel into the shape.

Subdivision

Subdivision is a useful tool and technique for creating smooth 3D models out of polygons. Take our earlier example of a simple cube that only has six faces. One could subdivide each face of the cube, dividing it into smaller quadrants. For example, an artist could subdivide the cube once vertically and once horizontally, thus splitting each face into four smaller faces.

Subdivision is a great way to increase poly count while letting a basic shape become the starting point for modelling. In general, subdivision can be completed in two types.

Uniform subdivision. This is when the entire object is subdivided evenly. This will affect the entire mesh.
Selective subdivision. In this method of subdividing, the artist only selects a single quadrant, for example one face and imposes the desired number of subdivisions. The remainder of the mesh is unaffected.

Bevels

On their own, the edges of a model are perfectly sharp. Consider again the cube example, where two faces share an edge. You could manipulate the angle between two faces making it square, obtuse or acute, but the edge itself would remain sharp.

Beveling tools allow artists to select an edge and then add any number of bevels to soften an edge. This is a useful tool but should be used carefully as it can easily add unwanted polygons or cause conflicts with topology if too much detail is added to one bevel.

Use Adobe Substance 3D to create inspiring art without fussing over polygons.

For many 3D workflows, polygonal modelling may seem unavoidable. Managing topology can feel like a hassle, especially in situations when having a perfectly streamlined mesh may not be required. This is especially the case in product-focused industries that want to take advantage of 3D.

In these situations, having a software that not only allows you to block out shapes quickly and efficiently, but can also produce high-quality results for both hard surface and organic surface modelling, can make all the difference in your success in 3D.

Adobe Substance 3D Modeler is a voxel-based sculpting software that aims to make 3D modelling feel as gestural and natural as working with real clay. From start to finish, artists don’t need to worry about polygons, topology or UV mapping. All these steps are handled by the software at export and the artist has export options available to help guide this process.

Discover modeller and the rest of the Substance 3D apps to learn more.

Thank you for exploring the world of Adobe. We can’t wait to see what you create.

style

Center, grid width 8, static links

style

M spacing