3D Camera shifts the Lightsword Hilt and Tip position
Hello
Initial situation
- I use a Lightsword Ultra (2-Point Auto) effect on a grade.
- The grade is 2D.
- The positions of the hilt and tip are each connected to a point.
- I determined the positions of the points with a tracker (tracking the position of a grip in a mans hand)
So far, everything works perfectly, the light sword is "attached" the mans hand, and moves in a correct position with the hand movement.
PROBLEM
- Now I added a 3D camera, because there should be a 3D model in the scene too.
- As soon as the 3D camera is active, the light sword hilt and tip positions are SHIFTED. They are no longer at the correct position
Conclusion
- It looks like the 3D cam affects the hilt and tip position
- As long a the 3D cameras Transform properties are reset to its default values, there is NO SHIFT
- but as soon as the 3D Camera has some movement, the light sword hilt and tip positions are moved with the camera
Question
- How can I detach the light sword positions from 3D camera movement?
- It makes no sense for me that this 2D effect is affected by a 3D camera, especially its position is tracked on a 2D media.
Many thanks for your support!
Roli
Answers
-
Hi @RoliKreis
I'm invoking @Triem23 as it seems to be related to the way layers are rendered on the layer stack. Have you tried applying the effect directly on the video layer instead of on the grade layer?
If the layers are 2D, there shouldn't be any offset as far as I know, but maybe I'm wrong.
-
Hi
Same "shift" if the Lightsword effect is directly added on the video layer.
Conclusion: The Lightsword effect interprets 2D-points as 3D, as soon as a camera is active. Perhaps a bug?
#1: Camera off: 2D points are at the correct position
#2: Camera on: 2D points are at the correct position, but Hilt and Tip are shifted by 200 to the left
#3: Camera on, points changed to 3D: Hilt and Tip are on position of the shifted points
Conclusion: The Lightsword effect interprets 2D-points as 3D, as soon as a camera is active
-
@laconstantedeplanck all I can do is verify I'm seeing the same behavior as @RoliKreis .
That's a bug.