In Miarmy 4.0, we can make our cache as an input and drive our agents. And then our agents can also controlled by logic and physics.

This going to be very useful when the director approved some crowd results but you still want to continue add dynamics on them

This feature is an extension of previously Layer Cache. We just need to mark it as input mode and everything can be done automatically.

Workflow

• First, you need to totally understand our layer cache and the workflow.
  If you didn't know that, please go through Distributed Simulation and Layer Cache

• Then let's talk a look at the layer cache input mode.
  When we select our place node in list and enable cache, there is a question for choosing the driven mode.
  At this time, let's choose the "As Input" mode.

Choose As Input Mode

• When we choose the "As Input" mode, the system will treat the cache like an input and drive the agent, but the agent can still be affected by environment and driven by logic, even enable dynamics.
  Please notice there will be ">>" input sign prior the place node when we enable the cache as "input mode"

"==" (full drive mode) VS ">>" (input mode)

• With logic sentences driven, some agents are enable dynamics and some agents are continue driven by cache

cached agents can continue simulate and enable dynamics

• You can hybrid use this feature and make some agents fully driven by cache and some agents driven by cache and still can be controlled by logic and dynamics

hybrid "full drive" and "as input" mode