In the last two decades, compute power, storage density, and
network bandwidth have improved by more than three orders of
magnitude. In the same time period, however, display resolution
has merely doubled. The increased processor and storage capacities
have enabled the computational sciences, but have simultaneously
unleashed a data avalanche on the scientific community. As a result,
scientific research is limited by data analysis and visualization
capabilities. In order to improve visualization capabilities,
we started Princeton Scalable Display Wall project in 1998 to
investigate issues in building and using a large-format, scalable
display system. Since then, we have built several display walls and
software systems. Our latest display wall built in our lab uses
24 projectors to drive about 20 million pixels onto an 18' by 8' screen.
The system uses an automatic alignment mechanism to achieve sub-pixel
geometric alignment accuracy and is capable of running various 2D and
3D visualization applications. Our recent focus is to work with
computational science researchers closely and help them use display
walls in their data explorations. This talk gives an overview of
the project and reports our lessons and experiences in building and
using scalable display wall systems. |