

- #BAD GRAPHIC CARD BENCHMARK EXAMPLE DRIVERS#
- #BAD GRAPHIC CARD BENCHMARK EXAMPLE DRIVER#
- #BAD GRAPHIC CARD BENCHMARK EXAMPLE SOFTWARE#
- #BAD GRAPHIC CARD BENCHMARK EXAMPLE WINDOWS#

Light-04, based on Discreet Logic's Lightscape Visualization System, has several tests for lighting, smooth shading, blending, line antialiasing, Z-buffering, and texture mapping. Radiosity simulates the light reflecting off surfaces and onto others, resulting in soft shadows. Lightscape Visualization System, or Light-04, is used to test radiosity visualization. It displays a set of particle traces through a vector-flow field using IBM's Visualization Data Explorer program for scientific visualization and analysis.
#BAD GRAPHIC CARD BENCHMARK EXAMPLE SOFTWARE#
The software models piping, equipment, and structural elements, such as I-beams, HVAC ducting, and electrical raceways in plants.ĭata Explorer, or DX-06, measures data visualization. It measures the most common operations performed by Intergraph's DesignReview. It measures shading and rendering of the most common operations performed by Alias Wavefront's Advanced Visualizer, a workstationbased 3D animation system for modeling, rendering, image composition, and video output.ĭesignReview, or DRV-07, gauges 3D modeling. Larger scores indicate greater speed or more frames per second.Īdvanced Visualization Test, or AWadvs04, is a key measurement of 3D geometry. Viewperf provides view sets which are combinations of tests using specific applications to test OpenGL performance.

#BAD GRAPHIC CARD BENCHMARK EXAMPLE WINDOWS#
Operating systems covered by the SPECopc group include OS/2, Unix, and Windows NT and 95. SPECopc is now a project group within the Graphics Performance Characterization Group managing OpenGL benchmarks. Since then, SGI, the former Digital, and other SPECopc members have updated and modified it. The Viewperf testing program was originally developed by IBM and written in C. Viewperf measures performance in frames per second. More are found at Here's a brief glossary of benchmarks and what they test. The accompanying table compares just two cards.
#BAD GRAPHIC CARD BENCHMARK EXAMPLE DRIVERS#
It measures window management performance as well as traditional graphics operations.Įxamining the graphic-driver capabilities of a variety of cards and operating systems helps determine which drivers get the most out of which cards. The SPEC Web site ( describes each test in detail.įor 2D operations, look for Xlibbased X11perf benchmarks.
#BAD GRAPHIC CARD BENCHMARK EXAMPLE DRIVER#
It contains five suites to measure performance of computers, cards, and driver combinations in five different tasks, such as animation, mechanical design, and modeling. It uses SPECviewperf 6.1.2 suites of 3D-graphic benchmark tests. This information is necessary to design high-performance drivers.īenchmarks for graphics cards running on Windows with drivers from card manufacturers are available from several Web sites and SPEC, the industry-sponsored Standard Performance Evaluation Corp. These drivers use proprietary registerlevel information provided under nondisclosure agreements. Not all support is the same.Ĭompanies, such as Xi Graphics Inc., independently develop graphics-driver software for 2D and OpenGL 3D cards, and for various Unix-based computers equipped with graphic cards from over 20 different card manufacturers. You might also research the history of support provided by card-chip manufacturers. When drivers are developed by a single card and chip manufacturer, look for benchmarks on their sites. The good news is that there is abundant benchmark information from several sources that makes it easy to compare performance of drivers and cards on several systems, especially for cards that support a library of graphic operations called OpenGL. But find the right mix, and the results can be impressive. And it's no surprise that lowcost drivers can ruin the performance of good cards. While drivers are critical elements in graphic systems, old, slow, or out-of-date graphics cards are not improved by new drivers. Getting high-performance 3D graphics out of Windows and Unix-based computers requires top-notch video hardware and drivers.
