1 Comparing Immersive Virtual Reality with Other Display Modes for Visualizing Complex 3D Geometry David W. Mizell Stephen P. Jones Boeing David.Mizell@boeing.com Stephen.Jones@boeing.com Mel Slater Bernhard Spanlang University College London m.slater@cs.ucl.ac.uk b.spanlang@cs.ucl.ac.uk Abstract This experimental research was aimed at determining whether or not immersive virtual reality (IVR) technology gives a user a measurable advantage over more conventional display methods when visualizing complex 3D geometry. Subjects were shown an abstract rod sculpture in a variety of display and display-control modes, and were tasked with assembling a physical replica of the sculpture they were visualizing. They were scored on the speed and accuracy with which they assembled their replica sculptures. Head-tracked immersive VR was shown to have a statistically significant advantage over joystick-controlled display modes, especially in the case where the displayed sculpture was shown in super-scale, surrounding the subject. 1. Introduction Why use immersive virtual reality? IVR equipment is still rare and expensive. For viewing graphical datasets of any interesting size and geometric complexity, powerful, expensive graphics computers are also required. Standards, languages, APIs, tools and hardware interfaces are still not yet well established and widely accepted. Thus specialized technical people are also required for the installation and operation of the IVR system. So what do we get for the money? What is the economic case to be made to the company CFO on behalf of IVR? Can we point to a realizable benefit? Can we cost-justify IVR? A technically sophisticated CFO will ask why we should buy an IVR system when, with the same amount of money, we could buy several quite powerful graphics workstations. Hearing these kinds of questions asked in the industrial context was one of the motivations for this work. The economic questions prompted a research question: Can we demonstrate a measurable advantage IVR has over desktop displays, or other conventional human-computer interfaces? It is clear that at least certain forms of VR offer a capability that simply doesn’t exist for conventional, “flat screen” interfaces. The primary example is the quasi-physical interaction made possible within a virtual environment by such devices as data gloves, with tracking sensors on the hand and other parts of the body. An IVR user can wear an animated graphical body, i.e., an “avatar,” and reach toward, grasp and manipulate virtual objects. With the advent of haptics technology, it is even becoming possible to feel the solidity and weight of the virtual objects. We decided to focus this inquiry entirely on the aspect of visualization. Granted that IVR offers unique forms of physical or quasi-physical interaction coupled with visualization, does IVR provide any advantage over conventional displays for visualization alone? We refined this question further, again guided by aerospace industry experience: Does IVR provide an advantage over conventional displays for visualizing and understanding complex collections of 3D geometry? Large aerospace manufacturers such as Boeing design their products using 3D CAD. The typical design review entails a group of engineers in a conference room, looking at the CAD geometry on a large, front-projection screen on the wall. They get lost a lot, or at least confused about what it is they are looking at. This doesn’t happen to them inside a real airplane. Would IVR mimic their interactions with the real world closely enough that they wouldn’t get confused about what they were looking at in a virtual environment, either?