reach_task

reach_task drives squad members toward their assigned simulation target. It is part of generic stalker alife behavior, not a hand-authored active section with LTX parameters.

Parameters

reach_task has no scheme-specific LTX fields in the current TypeScript implementation.

Runtime constants:

PATROL_UPDATE_PERIOD

Value: 1000.

Milliseconds between movement-order updates.

FORMATIONS.back

Value: built-in formation list.

Default follower offsets behind the squad commander.

Runtime behavior

SchemeReachTask.setup() installs the SMART_TERRAIN_TASK evaluator and action inside the nested alife planner. The evaluator returns true when the NPC’s squad has a REACH_TARGET action and the assigned simulation target is not yet reached.

The action initializes movement toward the squad assigned target. The squad commander moves toward the target game vertex and level vertex. Other squad members follow orders from ReachTaskPatrolManager, which keeps them in formation behind the commander and accelerates members that fall behind.

Objects are removed from the patrol manager when they die or switch offline.

Runtime sequence

1. SchemeReachTask.setup replaces the nested ALife planner’s SMART_TERRAIN_TASK evaluator and action.
2. EvaluatorReachedTaskLocation returns true only when the NPC’s squad is doing REACH_TARGET and the assigned simulation target still reports not reached.
3. SchemeReachTask.add subscribes the existing nested SMART_TERRAIN_TASK action for scheme events.
4. The action sends the commander toward the assigned target and keeps followers in formation.

Example

[logic]
active = walker@idle

[walker@idle]
path_walk = idle_walk
path_look = idle_look

reach_task is driven by squad simulation state. It is not normally configured with a dedicated [reach_task] section.

Notes

• Movement switches between game-path and level-path movement depending on whether the commander is on the target game vertex.
• During surge, reach-task movement uses running with free mental state.
• Debug this through squad simulation state first; there is normally no [reach_task] LTX section to inspect.