Evening,
I've been browsing the source and was wondering why the Scene class contains a camera rather than a View onto the Scene having the camera? At the risk of over-analysing what was quite possibly a simplification just for the book, it feels to me like there is a layer missing.
From what I've seen (so far - I'm still poking around), there's no way for different Scenes to reference the same scene-graph as the Scene constructor resets its root node. So, it looks like every camera has to have its own Scene (which feels rather inefficient) or the Scene's camera has to be changed for every view that is rendered (which may not work particularly well with multi-threading).
Have I missed something? Is there a reason why a scene-graph should-not/cannot be shared?
I've been browsing the source and was wondering why the Scene class contains a camera rather than a View onto the Scene having the camera? At the risk of over-analysing what was quite possibly a simplification just for the book, it feels to me like there is a layer missing.
From what I've seen (so far - I'm still poking around), there's no way for different Scenes to reference the same scene-graph as the Scene constructor resets its root node. So, it looks like every camera has to have its own Scene (which feels rather inefficient) or the Scene's camera has to be changed for every view that is rendered (which may not work particularly well with multi-threading).
Have I missed something? Is there a reason why a scene-graph should-not/cannot be shared?
- Arm