Inno3D Home Page Inno3D Home Page

FAQ News Search Archive Forum Articles Tweaks Technology Files Prices SETI
Visit NVIDIA's home page.
Preview Menu
  Trans. & Lighting
  Shading Processor
  Unified Drivers

  Quake 3 Arena
  Soldier of Fortune
  Starsiege Tribes


Direct 3D
  Unreal Tourney
  3DWinMark 2000
  3DMark 2000
Wrap Up
  HDTV Processor
Favorite Pics
Click to Enlarge
OCZ Tech Titan 3
1.0GHz Pentium III
eVGA MX Shootout
nForce Preview
VisionTek GeForce3
2001 Spring Lineup
GeForce3 Preview
eVGA TwinView Plus
VisionTek GF2 Ultra
OCZ GeForce2 Pro
Winfast GeForce2 MX
GeForce2 vs Quake3
Linksys Cable Router
GF2 FSAA Shootout
GeForce2 MX Preview
Benchmarking Guide
Quake 3 Tune-Up
Power DVD Review
Live! Experiences
Memory from

FastCounter by bCentral

 Visitors Are Online
Powered by
3D Chipset
Gamers Ammo
Reactor Critical
GeForce FAQ
Dayton's Misc.
G-Force X Sweden
Maximum Reboot
Media Xplosion
nV Italia
Riva Station
nV News Home Page

NVIDIA GeForce2 GTS Preview

Indy 3D Benchmarks

Indy3D is an OpenGL MCAD (Mechanical Computer Aided Design) benchmark which is geared toward animation and visual simulation professionals. Since the GeForce Quadro is being offered as an integral part of low-cost CAD workstations, a GeForce preview would be complete without throwing in a few comparisons using Indy3D.

I went against the grain in benchmarking Indy 3D as I only tested at a resolution of 1024x768 in 16-bit color. This was intentionally done to limit the memory bandwidth required under 32-bit color and to provide a clearer picture of fill rate and transform and lighting enhancements of the GeForce2 GTS under OpenGL. For example, at a resolution of 1024x768 in 16-bit color, the Indy3D fill rate results were reported as follows:

  • GeForce DDR - 407 Mpixels/sec
  • GeForce2 GTS - 633 Mpixels/sec

The increased fill rate of the GeForce2 GTS is also responsbile for the huge jump in performance in the MCAD40 and MCAD150 performance tests. Following the Indy 3D performance benchmarks, are detailed results from lighting stress tests which also show increased performance on the GeForce2 GTS.

The Indy3D benchmarks used in this preview consist of the following tests all of which were rendered in a 1024x768@16bpp window:

  • MCAD40 - consists of approximately 42,267 meshed, z-buffered, Gouraud-shaded, lit polygons.
  • MCAD150 - consists of approximately 150,248 meshed, z-buffered, Gouraud-shaded, lit polygons.
  • Animation - illustrates a human animation via a "walking man" figure which consists of 22,342 meshed, z-buffered, Gouraud-shaded, bilinear filtered, fogged, lit polygons.
  • Simulation - simulates a sailboat containing 7,260 meshed, z-buffered, Gouraud-shaded, translucent, fogged, trilinear filtered, MIP mapped lit polygons.


Click to enlarge - 122K
MCAD 150

Click to enlarge - 124K

Click to enlarge - 74K

Click to enlarge - 91K

Indy3D Performance Results - 1024x768@16bpp

Percent Increase in Frames Per Second

The lighting stress test was conducted by adding spotlights to the MCAD40 benchmark. Indy3D allows up to 6 spotlights and results are reported in both frames and polygons rendered per second. Unlike the official Indy3D tests, where 1/3 of the MCAD scores are based on running in an antialiased wireframe mode, this test is based entirely on polygon throughput.

MCAD40 - 3 Spotlights

Click to enlarge - 131K
MCAD40 - 6 Spotlights

Click to enlarge - 137K

MCAD40 Lighting Stress - 1024x768@16bpp

Spotlights GeForce 256 DDR GeForce2 GTS
  Frames/sec Poly/sec Frames/sec Poly/sec
0 71.4 2,984,419 108.7 4,578,261
2 53.0 2,230,932 84.5 3,557,433
4 35.0 1,474,790 55.6 2,340,000
6 25.9 1,090,062 42.2 1,778,716

With a 50-60% increase in the number of polygons rendered based on multiple spotlights, the transform and lighting capabilites of the GeForce2 GTS exceed that of the GeForce 256 DDR. This is similar to the increase in the 3DMark 2000 high polygon count benchmark.

Next Page: Unreal Tournament Benchmarks

Last Updated on May 12, 2000

All trademarks used are properties of their respective owners.