RigidBody3D only moving on a global axis with apply_force()

[hasty oak]

In the first video I use -basis.z, in the second video I use -global_basis.z

The yellow box is the ships position + -basis.z * 16
Same for the blue box except it uses -global_basis.z

This is the desired outcome

But instead the thruster that moves forward, moves north

And the side thrusters move west and east

Etc

(The side thrusters should also move the ship slightly but mostly turn it)

Here is my thruster code

I got it working

after days of trying

but it doesnt make any sense to me

i used vector3.ZERO instead of position

technically, this is solved but i'd like someone to give an explanation

[split vapor]

The docs have the following to say on the matter: position is the offset from the body origin in global coordinates. In other words, the final parameter already starts at your node's position, passing the current position would result in the force being applied from somewhere way off to the side, rather than centered on the ship.

[hasty oak]

But where does it get the nodes position? Apply force is called in the ship node

From my understanding, a basis only provides rotation and scale

Unless apply force somehow gets the node that called the method

@midnight canopy Figured I should @ you here as we were both stuck on this

[split vapor]

It gets the position of the node it "belongs to" as it were. So it will be force centered on the ship, even though it's being called from the thruster.

[hasty oak]

Do you mean the function gets the position of the node that calls it?

[split vapor]

No, the position of the node that the function is on, in this case the Ship.

[hasty oak]

This is all strange to me, I'd expect it work like apply central force but it does apply torque

I have no clue how from the information given it would calculate torque

[split vapor]

The docs state on apply central force "This is equivalent to using apply_force at the body's center of mass."

It has all the information it would need to calculate torque - it knows how much force, in what direction, and where on the object the force is pushing.

The first two both being contained by the force vector, and the second in the positional offset vector.

[hasty oak]

But theres no positional offset

So how would it know these side thrusters should apply torque

Ship is, by the way, a parent

[split vapor]

I'm not sure what you're asking. Can you be clear: what is happening, and what are you expecting to happen.

[hasty oak]

In my mind, apply force with a position of 0,0,0 should work exactly like apply central force

Because 0,0,0 means no offset from center

Apply central force cannot apply torque

Because torque would need the force to be positioned away from the ship on atleast 2 axes

But my side thrusters do

What Im about confused is

How does it know from the info in that function that the thruster is offset in a way that creates torque

To be clear it is intentional that the side thrusters do that

And the forward and backward thrusters do go forward and backward like intended

Everything works

But i dont see where the thrusters offset is passed to the ship’s apply force function

The thrusters are children of the ship btw

[hasty oak]

I found the source code for apply force