I’m playing with the 4.22 introduced Animation Sharing plugin and have a question specific to that plugin. I am able to let a crowd of actors play looping animations, such as idle, running forward and walking forward. Now I’d like to have certain actors play a one-off animation, which would be the use case for when an actor performs an action, for example performing a single jump, before transitioning to a looping animation.
The Animation Sharing Setup settings have a concept of on-demand animation states, which are described as “this means that we kick off a unique animation when needed”. If I understand correctly, when an actor registered to the Animation Sharing Manager enters an on-demand animation state, an animation blueprint is instanced for this actor alone. It makes sense to me that the use case for on-demand instances would be to play one-off actions. On-demand animation states have settings to “return to previous state” or “set next state”. Because of this, I would assume that when an actor must play a one-off animation, he is temporarily in that on-demand state and then automatically returns or moves into another state.
However, the user overridden function Process Actor State from Animation Sharing State Processor takes input parameters “Current State” and “On Demand State”. My question would be, is an on-demand state supposed be active in parallel to a non-on-demand state? If not, why are there separate values for Current State and On Demand State? And if they are supposed to play in parallel, how do you trigger the on-demand animation one-off, since Process Actor State’s return value sets the animation state so there seems to be only one animation channel? I appreciate input from anyone that has toyed with the Animation Sharing plugin!