Skip Nav U.S. Army Research Laboratory DoD Supercomputing Resource Center
Sitemap Contact Us Quick Links

Scientific Visualization

Who We Are

ZSU

Facilities, Tools, and Experience for
Understanding Physics Based Simulations

 Columbia

Scientific Visualization at the US Army Research Laboratory is focused on providing the facilities, software tools, and the experience to enable the understanding of physics based simulations. These types of simulations regularly produce terabytes of output. Therefore, to enable the understanding and presentation of this massive amount of data, we are focusing on three main areas:

Projection and Displays

Vision Dome with HMMWV Analysis A standard desktop monitor is simply insufficient to handle the visual complexity of terabytes of simulation data. For example, a typical isosurface of this type of data will have a depth complexity of 5 - 10 triangles per pixel, and in extreme cases, over 300 triangles per pixel. To enable the understanding of these simulations, we are exploring several large format display technologies both commercial and locally assembled. Systems such as the curved screen Collaboratorium, the RAVE II, along with the rear projected PowerWall and the multiple LCD panel DisplayWall, allow for sufficient visual real estate and pixel density to explore modern simulations. Active and passive stereoscopic viewing systems allow the exploration to be accomplished in stunning 3D virtuality.

Graphics and Multi-Channel Sound Hardware

Driving the projection and display systems is a combination of commodity and high performance graphics hardware. This allows us to take advantage of the latest advances in graphics technologies, which are currently being driven by the gaming industry, and applying them to our unique needs. Adding feedback techniques, both audible and tactile, allows for experiencing more "dimensions" of data simultaneously.

Advanced Software Tools and Packages

Vis5d Software Package Providing, and truly understanding how to take full advantage of commercially available scientific visualization packages allows us to provide a stable, yet state of the art, data exploration environment for many of today's simulations. Looking to the future, we will be using a combination of commercial, open source, and locally developed tools to enable runtime visualization, multivariate volume visualization, and High Performance Code coupling.