VGIS
(Virtual Geographic Information System)

• based on 1996 Lindstrom-Koller algorithm
• project at Georgia Tech, managed by professor Bill Ribarsky
• Frank Jiang is the research scientist/programmer/manager for VGIS
• multi-platform: SGI, SUN, and NT/OpenGL/E&S
• pc version has been run with 4, 2, and 1 processors
• only documentation is a 10-page "system paper" overview:
  • "An Integrated Global GIS and Visual Simulation System", Peter Lindstrom, David Koller, William Ribarsky, Larry F. Hodges, Augusto Op den Bosch, and Nick Faust. Abstract. GVU Technical Report 97-07.
• multi-threaded: at least 3 very active threads
  • 1: terrain manager - update current view, view culling, create display list
  • 2: terrain server
  • 3: rendering
• terrain AND texture data is stored as GSD format, converted from IMG and others
• received VGIS demo. Issues:
  • navigation is awkward (no documentation, no control of viewpoint roll)
  • threads eat 100% of CPU time on both CPUs, even with a totally static view
  • although demo supposedly includes elevation data, the landscape appears totally flat
  • lacks wireframe option, so degree of tessellation can't be seen
  • large display glitches at high altitudes (could be fault of the card's 16-bit Z-buffer)
  • threads get upset when the display window is minimized
  • what do "Elevation Threshold" (0-10) and "Texture Threshold" (0-10) do?
    • changing these values had no visible effect on the scene