What Is 3D Graphics And How It Works?

What is 3D Graphics?

We see 3D graphics every day, but we don’t think about how it works from the inside. Let’s take a look behind the screen.

Modern computers generate graphics that are almost indistinguishable from how we see life. And if you have always wondered how realistic masterpieces are created (or you are doing something similar), then it will be useful to know how 3D works in general. 

Let’s see how ordinary dots turn into 3D graphics:

Creation of 3D models

Characters, weapons, cars, donuts, landscapes … everything you see in games and movies using 3D graphics is made up of points, faces, and planes. 

It seems that it is just a smooth ball, but in fact, it consists of many points – vertices:

The more vertices, the more detailed the model looks and the more resources the computer needs to draw such an object on the screen.

Vertices are connected to each other and form ribs and faces:

All this forms a polygon mesh (English polygon mesh or just a mesh, geometry) – a set of vertices, edges, and faces (planes), which determines the shape of the object.

Each vertex has its own coordinates along the X, Y, and  Z axes. Matrix is used in computer graphics. For the calculation of matrices, you can use an online matrix multiplication calculator. And in order to find the determinant of the matrix, an online determinant calculator is also available.

3D modeling in Blender

To create solid objects (Eng. Hard surface ) 3D artists usually manually change the position of the faces.

Sculpting in Blender

But geometry is not the last step in creating a 3D model. 

For example, sculpted models have poor topology (how the mesh works) – too many vertices are involved.

To fix this, they use special tools for retopology  – this is when unnecessary edges are removed in order to optimize the model.

A source

You also need to prepare the material  – this is how different faces are painted or the whole model. Both a simple color and an image or pattern are possible.

There are many other important points: animation, texture baking, normal mapping, and so on. All this is worth closely studying for those who are going to model 3D. Now we will talk about a more technical issue.

Displaying 3D models on the screen

How to show a three-dimensional model on a two-dimensional screen? There is so much mathematics in this question that it might seem like it’s some kind of magic.

The space in which the objects are located is called the stage. Everything on it exists so far only in the computer’s memory in the form of data on geometry, materials, and so on.

For a computer to understand how to display all this, an observer is needed, with whose eyes he will look at the scene – a camera. And so that we can at least make out something, we need a light source.

This is where the magic begins: the computer has to figure out what the scene would look like from the camera’s point of view. 

This is how it works:

We only see what is between the clipping regions. Everything else, as you might guess, is cut off. The computer needs to figure out what colors to display on the monitor in each of the pixels. To do this, he sends rays from the camera and looks at what they hit.

If the ray hits an object, then the computer further checks which polygon it hit, what material the object has, how the light falls, at what distance the object is from the camera, and many other variables.

All this is broadcast on the projection plane (English viewport ) – a two-dimensional square in three-dimensional space. This plane is already used to compose the image that will be shown on the monitor.

The 3D-scene translation process is called a 2D-image rendering.

3D motion

We learned how one image is displayed, but after all, 3D also happens in films and games, where there is always some movement. In fact, we still use the same principle of animation as we did several centuries ago.

In 1877, the praxinoscope was invented – a drum pasted over from the inside with a sequence of images. Furthermore, in its center, there is another smaller drum covered with mirrors. If you look at the center of the device as it rotates, you can see the illusion of movement:

Now it looks like this:

• The rendered scene is broadcasted to the monitor.
• The position of objects on it changes slightly.
• And the updated image is displayed on the screen.
• Most modern monitors can display 60 pictures (frames) per second ( Frames Per Second, FPS ), which creates a smooth feeling.

Comparison of smooth movements in 30 FPS and 60 FPS games

In the case of games, all frames are rendered in real-time. That is, while the user is playing, the position of objects on the stage changes, the computer checks how it all looks 60 times per second and updates the image on the monitor.

Of course, this imposes restrictions on the image quality. For example, in games, it has only recently emerged technology ray tracing, which allows the software to calculate the dispersion of light rays.

Ray Tracing technology makes lights and shadows realistic, so even cube games like Minecraft look very believable.

In animation, there are almost no such restrictions:

• 3D artists compose the scene.
• They prescribe the behavior of the camera and objects.
• And they start rendering the video.
• It takes a lot of time, but the final video can be played even on very weak devices. Provided they support video resolution.