General Rigging

Here we can find tips on Getting Started and FAQs for Rigging.

Getting Started
You're given a model! Ask: With this information, you are ready to start rigging! Ask: IMPORTANT NOTE: Who will be animating this rig? If you are not 100% sure about what is needed, ask.
 * 1) What is this model?  Is it mechanical, or organic?  Is it a vehicle?  A character?  Does it have a face?
 * 2) What will this character/prop be doing? Is it a microwave with a swinging door, or is it a main character?  What actions will they be doing over the show?  Will they have dialogue?
 * 3) What is the animation style of the show?  Will characters and props move realistically, or will it be stylized, with over the top exaggerated arcs and motion?
 * 4) Does the meshflow support the level of deformation needed?  If the character's arms are going to bow into large swooping S curves, will it need additional edges?
 * 1) What sort of controls will the animators require? IK, FK, spline or ribbon?
 * 2) Can some of the motions be automated? Will muscle sim or nCloth be required?
 * 3) What rig elements should be prioritized? In other words, what do the animators need right now?

FAQs
Q. What is the difference between IK and FK?

A. IK stands for Inverse Kinematics: FK stands for Forward Kinematics.

In FK, each joint in the chain is controlled individually, each child inheriting transform information from its parent. In Ik, a single controller's movement is able to inversely control rotations for the entire chain.

For example: In FK, in order to raise the characters arm, you must rotate the shoulder controller. In IK, you would simply translate the wrist control and the rotation values for the shoulder and elbow joints are calculated based on the distance and location of that controller.

Q. What is meant by the parent child relationship?

A. A child is a node which is controlled by a parent. These terms are used in many situations. For example: In an object hierarchy, when the root node is moved, all nodes underneath inherit those transformations and move along with it. In this case, the root node can be considered the parent and all other nodes beneath it, its children. The terms grand-parent and grand-child may also be used when working in a multi-leveled hierarchy.

Q. What is the difference between a parent constraint and a parent?

A. Parent constraints act similar to parent-child relationships. In order to differentiate between the two, the terms driver and driven are often substituted for parent and child when dealing with constraints. In both scenarios, the "child" (driven) object moves along with the "parent" (driver). However, the way this happens is fundamentally different. Parent constrained objects do not have to live in the hierarchy of the driver node. Instead, transformation values (rotate, translate, scale) are calculated and applied directly to the driven object, effectively locking those attributes from being manually assigned. In a traditional parent-child relationship, the child's transform values are free to manipulate because the child lives under the parent, inheriting its world-space from the parent and adding it to it's own. One benefit of a parent constraint, is that multiple parent-constraints can be applied to a single object. In other words, one object can have multiple drivers who's weight values can be increased or decreased allowing for a dynamic parenting behavior like what is commonly referred to as 'space-switching'.