You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When given a negative value in ShapeCast.target_position it will register hits in any direction, even 180 degrees in the opposite direction. This bug respects the variable's magnitude, but only respects it's directionality in positive directions.
When placed inside of a RigidBody's CollisionShape, it does not register any hits even with ShapeCast.target_position set to true. It is important to clarify that it is detecting nothing, at all. This bug is inverted when the target position is negative.
Steps to reproduce
Behavior 1:
Create a new ShapeCast and rotate it 180 degrees so that it points straight up.
Position above a collider and no more that 1 meter away from touching it.
Run the scene with visible collision shapes and observe how it detects a collision despite facing away from the StaticBody.
Play around with the ShapeCast's position and rotation.
Behavior 2:
I'm unable to figure out a reliable sequence of steps to reproduce one. It's affected by the node's position and the 'target_position' variable, but it doesn't always work. It might be a byproduct of Behavior 1 based on how my attempt to recreate it in the MRP turned out, but it doesn't seem like it would be related.
I'll post a proper list of steps in the comments if I find one. Try playing around with "test 3" and "test 4" in the MRP to see if you can trigger this behavior.
Tested versions
Reproducible in: 3.6
System information
Linux MInt 22 Cinnamon - Godot 3.6.0
Issue description
ShapeCast.target_position
it will register hits in any direction, even 180 degrees in the opposite direction. This bug respects the variable's magnitude, but only respects it's directionality in positive directions.ShapeCast.target_position
set to true. It is important to clarify that it is detecting nothing, at all. This bug is inverted when the target position is negative.Steps to reproduce
Behavior 1:
Behavior 2:
I'm unable to figure out a reliable sequence of steps to reproduce one. It's affected by the node's position and the 'target_position' variable, but it doesn't always work. It might be a byproduct of Behavior 1 based on how my attempt to recreate it in the MRP turned out, but it doesn't seem like it would be related.
I'll post a proper list of steps in the comments if I find one. Try playing around with "test 3" and "test 4" in the MRP to see if you can trigger this behavior.
Minimal reproduction project (MRP)
shapecast_issues.zip
The text was updated successfully, but these errors were encountered: