What is a 3d object?

So we’ve been doing this thing for long enough that there’s people that didnt see this developed but rather were born into a world where cg animated films and videogames exists

Now with all the flashy new developments in technology and fantastic things you can do today I thought there had to be a way to bring things back to basic 

So .. what is a 3d object? 

A 3d object in its most basic form is an object that lives in a 3 dimensional space  the D stands for dimension.  Think back to your geometry class  where we had an x and a y and now we’re adding a z  that’s the frame of reference for or 3d object 

Now  if we want to talk about what 3d object in CGI is  then lets say its a representation of this in the computer 

Usually this refers to a container that holds data as a generalization most 3d softwares create by default 

A mesh ( this is a coordinate representation of an object

A shader(the properties of the material that the object is made of ) 

Modifiers (that affect the mesh )

or Some Extra properties. (this can be any type of data attached to the object) 

The important thing is that when you create a cube.. you really are creating by default, a container, a mesh and often a shader. There’s a 4th thing you’re creating.. construction history 

Every software treats construction history a little bit different but in essence it’s the track of what you do to the generated mesh.. 

Here’s where you get into destructive softwares or linear based editors and non destructive or non linear editors.. Let’s not get too deep into it but basically with the first kind the tendency is that once you edit stuff it is expected to be broken if you then change the order of the operators .. not the case with non linear editors. 

Now I like to keep videos to a single topic it makes it easier for people to find what they’re looking for but I feel talking about mesh and shaders is within this scope

A mesh is different from a 3d object in fact you should visualize it in the following hierarchy, the object has a lot more information than the mesh. 

Going back to our 3d object, an object can be empty or just have one point .. the important thing is that within an object you usually get one type of mesh , there are many type of mesh types, cuves, polygons, subdibs nurbs, grease pencil in blender. 

To go into each of the mesh types is beyond the scope of this but I want to talk about polygons  which is a mesh type that has  a few pieces inside ...components. 

Polygons are made of  points, edges and faces. in some softwares you’ll find that points and vertices are different things. I'm looking at you houdini but let's keep it simple. 

One thing I often see students get confused by is object space and world space, in essence theres a hirarchical order( parent child relationship) between the object and the scene/world an then the object and the mesh

As you can see in the image to the right the scene is the parent of everything then the object is a child of scecen and mesh is a child of object. you can move the individual elements of a mesh, the object or the scene and if you move an element that has children the children move with the parent.

On a very basic level all we’re doing are simple translations and rotations of points in 3d space.

The first unit of a polygon is point or vertex . .points live inside the object container that on itself lives in a  3d space. This is where we get Object space and world space..  they’re if you will , two different places to describe the position of an element.. 

the position of the point is defined from the object space. we put the coordinates of where it is.. without getting too deep into linear algebra, just know that the point is defined form the object and then the position of the object gets added to the point to get the world space.

Points can be inside a mesh without being connected to anything else, in fact there's a popular kind of data called a point cloud that is just that. and points can hold data and information, anything from color to random attributes. Scientists use this all the time to define terrains and even space simulations. 

So we’ve created a point that lives inside our mesh, this mesh itself is inside an 3d  object

Now we’re gonna duplicate this point and create a “line” between the first and second one .. 

this is called an edge, this is still a two dimensional object not what we consider a 3d object although we might move the individual points in the 3d space. 

Finally if we connect 3 edges or 3 vertices we get a face, with a face we introduce a  few more elements the first one is a face normal. A normal is a device we use to tell the computer which is the positive side and which is the negative , the in and out.. 

The other thing we are gonna get when we create a face is the possibility to map things, specifically 2d images into this 3d space, those coordinates are applied using uv maps.. So just to recap we have points, edges, faces, those faces have normals( the direction ) and uv maps wich is how to project 2d images into our 3d objects.. 

it’s important to note that  U would correspond to the horizontal axis or the x and V to the vertical axis or y ..3d packages have either the Y as the top down axis or the Z as the top down