CG animation is a huge subject, and there are many different parts that make up a fully CG production. But most projects work in much the same way. Here is a quick list of the processes involved in virtually all CG animation productions.
First up, modelling. This is the process of building the 3D objects that will feature within the animation. Without any objects you can’t make an animation, so this is a pretty fundamental step in production. This can be done from scratch, using modelling tools in applications such as Cinema 4D, 3DS Max or Maya. Or you can import 2D illustrations to extrude into 3D shapes, for example a logo. In some cases there will be existing CAD (Computer Aided Design) files to work with, but these often need refinement for use in animation, and sometimes contain way more information than you actually need.
This is where materials are added to the 3D objects to help make them look real. If you’ve just modelled a table, you might want to make it look like wood. Throw a wood material on there and hey presto, a photoreal table. Well, not quite, see the additional steps below. There are many sites that sell materials, such as Poliigon and Greyscale Gorilla, where you can buy wood, stone, metal, cotton, liquids – you get the idea.
One of the most important aspects for achieving a photoreal look, lighting is a skill and an artform. In the same way lighting is done on a live action film set, you can set the mood of the animation, by adjusting the size, number, direction and type of lights you use. The terminology in 3D software is much the same as in film production too, with soft boxes, spotlights and area lights used to create different looks on your 3D film set. You might want to go soft and bright for a beauty product, or dark and moody for a new tech product. The way you choose and manipulate your lights will determine the look. Get it wrong and you won’t even reach the depths of the uncanny valley, let alone achieve photoreal greatness.
To be able to see what the viewer will see, you will need a camera in your 3D scene. You can choose a regular camera which can be moved freely, or a target camera which is locked to aim at a particular point in your scene. There are also a whole heap of different camera rigs that people have built for different purposes, but let’s not overcomplicate things at this point.
This is the fun bit. Where you make the 3D objects that you have built move. You could write a whole book on how to make things move. In fact, a quick search of Amazon will reveal literally hundreds of books on the subject. The process within a 3D application will usually involve a timeline, where you set keyframes at different points along that timeline, locking the position, rotation and scale of an object at that moment. You then set another keyframe a few seconds along the timeline with different position, rotation and scale, and there you have it, an animation. Animating the camera is also a really important part. You’ll rarely see a CG animation with a static camera.
Not essential to all CG animation, more of an offshoot if required, but simulations are how you can create realistic liquid, fire, smoke, and altogether more abstract swirling animation effects that clog up mograph Instagram like hair in a perfectly rendered CG plughole.
Rendering is the process whereby the computer puts everything you’ve done into files that you can actually show someone. It’s like developing a photograph from film or baking a cake once you’ve thrown all the ingredients in. You just sit back and wait for the oven (computer) to bake (render) the cake (animation). Like lighting, rendering is another area that requires a lot of knowledge and skill in order to achieve a photoreal look. And here’s the extra complication: there are as many renderers as there are 3D applications. Most 3D applications include a renderer within them, but most animators will now be using an external renderer to handle this part of the process. Popular choices at the moment are Redshift, V-Ray, Octane, Cycles and Arnold. There was a huge shift from CPU to GPU rendering a few years back but that’s a whole other blog post.
The final part in the process is compositing. This is the process of refining the look of the things you’ve rendered. A 3D render is typically formed of a number of passes – layers of information that you can use to change the look of particular parts. Shadows, lighting, reflections, refractions, colours, depth of field, motion blur can all be tweaked in the compositing phase to create the perfect final look.
So there you have it. This is by no means an exhaustive list of every detail of production, but all CG animations will require most, if not all of these processes, in a similar order as set out above.
To discuss starting a CG Animation project, get in touch with Frantic today.