Craig P
Hmm, let's think to myself.
I have an orbital camera separated into yaw and pitch gimbals. I then steal that camera off the gimbals and move it around for another purpose.
Then I want to *move the gimbals into place* so we can have the camera stay in the same place.
Obviously this is just a matter of reverse-engineering the pitch and yaw of the gimbals, I've done it loads of times.
But I've never really touched Godot's transforms.
Craig P
I think the best thing to do would be to project the camera position onto the plane of the yaw gimble, lookat, then project it onto the plane of the pitch gimbal, lookat.
Should work...
The Seven Voyages Of Steve
@Craigp I’d do it by an arccos for pitch and arctan for yaw based on the new lookat (camera ray intersect with horizontal plane)
Craig P
@sinbad I THOUGHT that was what I was doing with the project function. It works fine for yaw, but then pitch spins on a stupid axis. My bet is that the up axis just goes bananas, I may have to simply manually set the rotation instead of using lookat.
The Seven Voyages Of Steve
@Craigp projecting a position onto a plane normally projects a new position along the normal of the plane, which isn’t what you want
The Seven Voyages Of Steve
@Craigp hmm yeah UE has separate heading / pitch funcs. From a Quick Look at the Vector3 class in Godot I think you’d need to use the angle_to for pitch (against Up) and signed_angle_to for heading/yaw, against one of the forward/right plane vectors, after flattening the camera direction vector to that plane
Craig P
@sinbad It does, but it's unclear how I'd constrain it properly. I did a few things, but I can't quite grasp where it's going wrong since there's several steps and I don't know how to examine the live scene in Godot.
GNU social JP is a social network, courtesy of GNU social JP管理人. It runs on GNU social, version 2.0.2-dev, available under the GNU Affero General Public License.
All GNU social JP content and data are available under the Creative Commons Attribution 3.0 license.