BD - Can See Object - Angle Offset 2D / 3D?

darkAbacus247

New member
Does Behavior Designer offer offset settings for 3D angles?

Attached is a video example of my goal. The target is located & tracked when in front of the enemy but is out of sync with the displayed Gizmo Graphic.
Setting the angle value on the Can See Object back to zero and it's back in sync again. AKA, I don't know how to properly use this tool....

This enemy has eyes on side of its head, so I'd like to replicate this if possible. I've researched the forms and everything, if someone could point me in the right direction I'd be grateful!

Thanks everyone :)
 
Ah, for this I would create a new Can See Object ability. Instead of doing a single check for the single position you'll want to perform two different checks for each eye. There isn't a way to do a blind spot with the current task.
 
Thanks Justin :)
If I'm following correctly, I think this is what I've done?

Attached is a screenshot of my current behavior, a real work in progress ;p

I have placed 2 Can See Object controllers, one for each eye. Everything seemingly works, as in when stepping in front of the enemy you're seen. The forward direction is that of the character though, I'm not following how to offset the Can See Object rotation(s)? The 2D offset is perfect but of course doesn't work in a 3d environment shown in my example video. I hope I'm describing this well enough....

My only thought is to create 2 behaviors, one for each eye that are assigned empty game objects placed & rotated where the eyes are? but...well that comes with it's own issues like sending messages back / forth to the main behavior - haven't gotten nearly that good yet...I'm sure I'm just missing the obvious here, yet again :9



behaviorExample.jpgfwdFacing.jpgoffsetFacing.jpg
 
The Can See Object task will use the agent's rotation value so by just setting the offset you aren't going to get the effects that you want. For this specific situation I would not use the actual Can See Object task - I would instead create a new task similar to Can See Object which has a reference to both eyes. You can then perform the can see check for each eye within that task. This will then use the correct rotation values since you can use the eye's transform.
 
I think that it's different from what you said. For this situation I would script a completely new task that is similar to the Can See Object task. This task will then allow you to specify the locations of each eye.
 
Top