Norman Vine writes: > Well the Mouse code certainly could but let's leave that alone > as I REALLY don't want the Mouse reading properties :-)
Not read them, but set them. It wouldn't much matter if the mouse code called globals->get_current_view()->set_orientation_offsets(r, p, h); or view_roll_node->setDoubleValue(r); view_pitch_node->setDoubleValue(p); view_heading_node->setDoubleValue(h); as long as the values ended up back in the viewer *and* the property tree somehow. In either case, though, I think that we need to throw out our current view_offset and view_tilt (the latter is my own evil creation) and replace them with full 6-DOF view offsets: - x (meters) - y (meters) - z (meters) - roll (degrees) - pitch (degrees) - heading (degrees) These would allow the view to move around inside and outside the 3D model in any way -- for example, the pilot could sit higher or lower in the seat, walk back through the cabin, or stick her head out the window to see where she's taxiing in a taildragger. On a separate note, some day we'll get around to making the mouse bindings user-configurable, like the joystick and keyboard bindings. I have received many e-mails asking why we haven't done that already. When that happens, properties are going to be hard to avoid. > Good enough to get you and many others hooked on FGFS :-) Absolutely. It works well and efficiently, and has served the project very, very well; the only problem is that we cannot easily extend it as FlightGear becomes more sophisticated. We all owe Norm an enormous debt of gratitude for getting all this stuff working in the program's early days, or there would have been no program for us to work on now. All the best, David -- David Megginson [EMAIL PROTECTED] _______________________________________________ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
