Hello! I am porting an OpenSceneGraph application to Qt3D (Qt 5.9.2). Everything works fine so far, but there is a problem regarding QOrbitCameraController: when I use the RMB to pan the view, the camera doesn't rotate about the vertical axis as expected. I found the problem looking into the source code of the QOrbitCameraController class: the up vector is hardcoded to (0, 1, 0). Since my setup uses Z as the vertical axis, I had to set the camera up vector to (0, 0, 1), so the rotation is performed about the wrong axis. I think that setting the controller to use the up vector from the camera would make more sense. It behaves as expected if I modify the QOrbitCameraController class code, by passing m_camera->upVector() to panAboutViewCenter() , instead of m_cameraUp . However, this is not an optimal solution, since it requires to recompile Qt. It would be nice to have a camera controller similar to OSG's "TrackballManipulator". Will it be possible to achieve a similar behavior with the upcoming QAbstractCameraController class that's being introduced in Qt 5.10?

Hi @ LeDrull 5.10 is already in beta phase. No chance for such changes there. You should propose and discuss such changes on the Qt Interest mailing list anyway: http://lists.qt-project.org/mailman/listinfo -Michael.

Your browser does not seem to support JavaScript. As a result, your viewing experience will be diminished, and you have been placed in read-only mode.

Please download a browser that supports JavaScript, or enable it if it's disabled (i.e. NoScript).

Qt 6.12 LTS is out! See what's new in the release blog

QOrbitCameraController not working as expected

Unsolved General and Desktop

2 Posts 2 Posters 1.3k Views 1 Watching

L Offline
L Offline
LeDrull

wrote on last edited by

#1

Hello!

I am porting an OpenSceneGraph application to Qt3D (Qt 5.9.2). Everything works fine so far, but there is a problem regarding QOrbitCameraController: when I use the RMB to pan the view, the camera doesn't rotate about the vertical axis as expected.

I found the problem looking into the source code of the QOrbitCameraController class: the up vector is hardcoded to (0, 1, 0). Since my setup uses Z as the vertical axis, I had to set the camera up vector to (0, 0, 1), so the rotation is performed about the wrong axis.

I think that setting the controller to use the up vector from the camera would make more sense. It behaves as expected if I modify the QOrbitCameraController class code, by passing m_camera->upVector() to panAboutViewCenter(), instead of m_cameraUp. However, this is not an optimal solution, since it requires to recompile Qt.

It would be nice to have a camera controller similar to OSG's "TrackballManipulator". Will it be possible to achieve a similar behavior with the upcoming QAbstractCameraController class that's being introduced in Qt 5.10?
M 1 Reply Last reply

0
L LeDrull

Hello!

I am porting an OpenSceneGraph application to Qt3D (Qt 5.9.2). Everything works fine so far, but there is a problem regarding QOrbitCameraController: when I use the RMB to pan the view, the camera doesn't rotate about the vertical axis as expected.

I found the problem looking into the source code of the QOrbitCameraController class: the up vector is hardcoded to (0, 1, 0). Since my setup uses Z as the vertical axis, I had to set the camera up vector to (0, 0, 1), so the rotation is performed about the wrong axis.

I think that setting the controller to use the up vector from the camera would make more sense. It behaves as expected if I modify the QOrbitCameraController class code, by passing m_camera->upVector() to panAboutViewCenter(), instead of m_cameraUp. However, this is not an optimal solution, since it requires to recompile Qt.

It would be nice to have a camera controller similar to OSG's "TrackballManipulator". Will it be possible to achieve a similar behavior with the upcoming QAbstractCameraController class that's being introduced in Qt 5.10?
M Offline
M Offline
m.sue

wrote on last edited by

#2

Hi @LeDrull

5.10 is already in beta phase. No chance for such changes there.

You should propose and discuss such changes on the Qt Interest mailing list anyway: http://lists.qt-project.org/mailman/listinfo

-Michael.
1 Reply Last reply

1