Class TCastleAbstractRootTransform

The major axis of gravity vector: 0, 1 or 2. This is trivially derived from the known camera GravityUp value. It can only truly express GravityUp vector values (1,0,0) or (0,1,0) or (0,0,1), although in practice this is enough for normal games (normal 3D scenes use up either +Y or +Z).

We try to avoid using it in the engine, and use full GravityUp vector wherever possible. Full GravityUp vector may allow for more fun with weird gravity in future games.

Collisions with world. They call corresponding methods without the World prefix, automatically taking into account some knowledge about this world.

Calling these methods to check collisions makes sense if your collision query is not initiated by any existing TCastleTransform instance.

If your query originates from some existing TCastleTransform instance, you usually do not want to call these WorldXxx methods. Instead call TCastleTransform.MoveAllowed, TCastleTransform.Height methods. Underneath, they still call World.WorldMoveAllowed and World.WorldHeight, additionally making sure that the object does not collide with itself.

What is hit by this ray. Returns the TCastleTransform that is hit (this is the "leaf" TCastleTransform in the TCastleTransform tree that is hit) and a distance from RayOrigin to the hit point. Returns Nil (Distance is undefined in this case) if nothing was hit. Use WorldRay for a more advanced version of this, with more complicated result.

Camera position, direction, up and gravity up vectors. Expressed in the coordinate space of this TCastleAbstractRootTransform, which means: the coordinate space of enclosing TCastleViewport.

Note that some features of TCastleScene (like LOD or Billboard or ProximitySensor) need to transform this camera to scene local space. Which is not possible if the same scene instance is used multiple times (e.g. under many different TCastleTransform parents). If you need to use these features of TCastleScene, then simply do not use the same scene reference multiple times (instead clone the scene by TCastleScene.Clone).

CameraKnown = False means that CameraChanged was never called, and so camera settings are not known, and so other Camera* properties have undefined values.

Yoo can temporarily disable physics (no transformations will be updated by the physics engine) by setting this property to False.

Limit the movement allowed by WorldMoveAllowed. Ignored when empty (default).

This property allows to easily limit the possible places where player and creatures go. Player is honoring this if it uses WorldMoveAllowed, in particular our TCastleWalkNavigation navigation honors it. Creatures honor it if they use WorldMoveAllowed for their decision, in particular all creatures in CastleCreatures use it.

Note that the TLevel.Load always assigns this property to be non-empty. It either determines it by CasMoveLimit placeholder in the level 3D model, or by calculating to include level bounding box + some space for flying.

The central camera, that controls the features that require a single "main" camera (features that do not make sense with multiple cameras from multiple viewports).

This camera controls:

- the X3D nodes that "sense" camera like ProximitySensor, Billboard. - an audio listener (controlling the spatial sound). - the headlight. - when X3D nodes change Viewport/NavigationInfo, they apply these changes to this camera.

Note that it means that "headlight" is assigned to one camera in case of multiple viewports looking at the same world. You cannot have a different "headlight" in each viewport, this would cause subtle problems since it's not how it would work in reality (where every light is visible in all viewports), e.g. mirror textures (like GeneratedCubeMapTexture) would need different contents in different viewpoints.

By default this is set to TCastleViewport.Camera of the TCastleViewport that created this TCastleAbstractRootTransform instance. So in simple cases (when you just create one TCastleViewport and add your scenes to it's already-created TCastleViewport.Items) you don't have to do anything, it just works. In general, you can change this to any camera of any associated TCastleViewport, or Nil (in case no camera should be that "central" camera).

TODO: Use free notification to automatically nil this. For now, be sure to unassign it early enough, before freeing the camera.

Adjust physics behaviour.

Time scale used when not Paused.

Pausing means that no events (key, mouse, update) are processed. So time doesn't move, and input is not processed.

Navigation also doesn't work (this part is implemented by TCastleViewport and each TCastleNavigation).

This is useful if you want to unconditionally make your world temporary still (for example, useful when entering some modal dialog box and you want the world to behave as a still background).

See also: For other pausing methods, there are other methods of pausing / disabling some events processing for the world:

• You can set TCastleScene.TimePlaying to False. This is roughly equivalent to not running their Update methods. This means that time will "stand still" for them, so their animations will not play. Although they may still react and change in response to mouse clicks / key presses, if TCastleScene.ProcessEvents.

• You can set TCastleScene.ProcessEvents to False. This means that scene will not receive and process any key / mouse and other events (through VRML/X3D sensors). Some animations (not depending on VRML/X3D events processing) may still run, for example MovieTexture will still animate, if only TCastleScene.TimePlaying.

• For navigation, you can set TCastleNavigation.Input := [] to ignore key / mouse clicks.

  Or you can set TCastleNavigation.Exists to False, this is actually equivalent to what pausing does now for TCastleNavigation.