Transition Rules

Transition Setup

To control which states can blend to another, you can create transitions, which are links between states that define the structure of your State Machine.

To create a transition, drag from a state border to another state. In this example the Idle state is connected bi-directionally to the Run state, which is a common setup for basic locomotion State Machines. Transitions are single-direction, so if two states are intended to transition back and forth, you need to create a transition for each direction.

You can also rebind existing transition logic by selecting a transition node and dragging it to a different state. You can rebind multiple transition nodes at once by dragging the transition arrow to a new binding.

Transition Rules

While transitions control the structure of inter-state blending, transition rules control when a state can transition. In other words, it is not enough to simply define a transition, you must also define how and when the transition happens.

Transition rules are automatically created when you create a transition. Like states, you can see and access them from either the My Blueprint panel, or by double clicking the transition icon in the State Machine graph.

Within transition rules, you can create any variety of blueprint logic for checking and comparing, all with the purpose of outputting a boolean (true or false) value. A true value is used to determine whether the state can transition to the next state.

For example, when transitioning from idle to run, and then back to idle, the logic might look like this. In this example, a bool variable is being used to provide the transition rule. When returning to the default state, the inverse of the bool is used.

Transition rules are typically informed by the movement component and other character variables. To learn more about how to get common properties for a controllable character, refer to the How to Get Animation Variables page.

Once your variables are correctly set up with the intended logic, meeting the conditions of those variables during gameplay causes the transition to occur.

Transition Functions

Within the transition rule graph, you can use the following specialized transition-only functions to augment your logic:

Animation Notify Functions

There are also several Animation Notify functions that you can use within the transition graph. You can add these functions by right-clicking in the transition graph and locating Was Anim Notify… functions.

The main differentiators between the notify function checks are if they are searching:

• In any state, which searches for the notify in all active State Machines.
• In the source state, which searches for the notify in the previously active state (the state being transitioned from).
• In the State Machine, which searches for the notify in a specific State Machine.

Transition Interrupt

While transitioning, if another state becomes active, then the transition will "interrupt" and transition to that new state instead. When this interruption occurs, you can link certain Animation Notifies to the interruption. This makes those notifies execute when the interruption occurs.

To set up transition interruption notify behavior, select the transition and locate the Transition Interrupt properties in the Details panel.

You can specify either a Notify or Notify State Class to link to the interruption. If a Notify State is used, then the interruption executes both the start and end notify events in sequence on subsequent frames.

The other properties can be referenced in the Montage Notify page.

Transition Blend Types

There are three main types of state transition blending you can use when deciding how you want states to transition: Standard Blend, Inertialization, or Custom. You can choose any of these types by selecting the transition and locating the Blend Logic property in the Details panel.

Standard Blend

Standard Blend is the default transition option and contains settings for duration, curve, and other basic controls. You can locate and adjust these settings by selecting the transition and locating the Blend Settings category in the Details panel.

Inertialization

Rather than simply blending from one state to another, Inertialization can be used where the animation velocity and acceleration occurring at the time of the switch to the new animation are used to carry forward the motion. Refer to the Inertialization document for more information.

If you are using inertialization as a blend type, you must also ensure that the Inertialization node is used in the Anim Graph. It must be placed after your State Machine evaluates.

Inertialization does not produce good results on all transitions, depending on the animations used in the states bordering them. Some general rules to keep in mind when using interialization blending are:

• Keeping your blend duration short, less than 0.4 seconds is best.
• When the poses are extremely different, do not use inertialization.

Custom

Custom blends are blends you can graph and customize in their own Anim Graph layer, with its duration and curve shape determined by the standard blend settings.

When Blend Logic is set to Custom, you can either click Edit Blend Graph next to the dropdown menu, or double click the custom blend graph in My Blueprint to open the graph.

Within a custom blend graph, you can use the following special functions to read transition time and state weight information to inform your graph:

With custom blends, you can create a wide variety of blending logic. As a simple example, you can replicate a standard blend by using a normal Blend node with the Get Transition Time Elapsed (ratio) function.

Transition Properties

Transitions contain the following properties: