How are you? :)
Do you have any ideas?
alteous @alteous is investigating
Oh, I see what's going on now, never mind.
I see it
It's just scene graph update
Or not
wat
I'm still investigating/debugging
But not
What's the point of
mesh.set_orientation(Quat::from_angle_y(Rad(PI)));?
That mirrors the spaceship over in the Y axis in the XY plane.
I've fixed it and removed that line, no changes in rotation
Here is the changes if you're curious vitvakatu/asteroids@a76232a
Great minds think alike, eh?
Great minds can't solve problem for over a half an hour
Doesn't sound great :D
- mesh.set_orientation(Quat::euler(vec3!(0.0, PI, 0.0)));This only confuses matters. You've already fixed the vertex winding order.
- let new_orientation = Quat::axis_angle(vec3!(1.0, 0.0, 0.0), self.rotation) * self.orientation;
+ let new_orientation = Quat::axis_angle(vec3!(0.0, 0.0, 1.0), self.rotation);You don't need this extra multiplication. Now the rotation around the Z axis behaves as intended.
+ let mut dv = new_orientation.rotate(vec3!(0.0, 1.0, 0.0));This now behaves as intended, although I'm not familiar enough with quaternions to know if this should actually work.
Thank you so much
I'm not sure I really understand why it was working with cgmath before
Neither am I lol.
You need to swap these signs:
if input.hit(three::Key::A) {
self.rotation += SHIP_ROTATION_SPEED * input.delta_time();
}
if input.hit(three::Key::D) {
self.rotation -= SHIP_ROTATION_SPEED * input.delta_time();
}
If you graph on paper, the rotation angle is correct (anti-clockwise for increasing rotation around Z axis).