confusion over vizcave and viewpoint
 

I'm using a head-tracker and power wall setup. I use to allow the frustum to be automatically adjusted according to head-tracker coordinates. Alternatively, I could manually do this with and an appropriate timer, though I'm not clear why this would be useful (the documentation does not explain this well).

But how should I deal with the viewpoint? From reading the documentation on sensors and linking it seemed to me that the appropriate way was to use but vizcave provides the view matrix, which I alternatively linked to the main view with no apparent difference. What is the appropriate way of updating the viewpoint? And are these two methods completely equivalent?

When using the CaveView object, you don't need to manually link it to the viewpoint, it automatically does that for you. In most cases, creating a CaveView object is no different than simply linking your tracker to the viewpoint.

The CaveView object comes in handy when you need to move the virtual viewpoint away from the origin. In this case, you can treat the cave view object like a standard node object and use the setPosition, setEuler, etc.. methods to move the viewpoint around in your virtual world while still being linked to the tracker. The vizcave section of the documentation has some sample code that shows how to use the CaveView object.

OK, so CaveView gives me more flexibility than simply linking my tracker to the MainView. Forgive me for being totally naive, but in what type of head-tracked situation would you want this additional flexibility to move the viewpoint? Wouldn't that look really odd to the user wearing the head-tracker if the viewpoint suddenly changed NOT as a result of their movements?

Oh, and would you mind also explaining in what situation I might want to use vizcave.update() instead of vizcave.setTracker?

Sorry to ask questions about options that may not apply to me, but I can't be sure that they don't apply if I don't totally understand what they mean. Besides, I can hope that some other user may find this forum thread useful. I couldn't find anything on this topic here.

A simple example would be someone viewing a large indoor environment. Let's say you wanted to show them the second floor, or a room on the other side of the building. They cannot physically walk there, so you would move the virtual viewpoint origin to the other room, and now they can walk relative to the center of that room.

Realistically, you will never need to use the cave.update() function. There might be a case where someone would have to manually manipulate the tracking data before passing it on to the cave object. In this case they could manually update the cave after they have manipulated the data.