SPECviewperf^® 13

Benchmark Run Rules

1. The system under test must perform all of the respective graphics API’s functionality and other work requested by the benchmark.
2. The systems under test must be OpenGL Conformant for the pixel format or visual used by the benchmark.
3. Settings for environment variables, registry variables and hints must not disable compliant behavior.
4. No interaction is allowed with the system under test during the benchmark, unless required by the benchmark.
5. The system under test cannot skip frames during the benchmark run.
6. It is not permissible to change the system configuration during the running of a given benchmark. For example, one cannot power off the system, make some changes, then power back on and run the rest of the benchmark.
7. Screen grabs for SPECviewperf will be full window size.
8. The color depth used must be at least 24 bits (true color), with at least 8 bits of red, 8 bits of green and 8 bits of blue.
9. If a depth buffer is requested, it must have at least 24 bits of resolution.
10. The display raster resolution must be at least as high as that selected in the “Resolution” dropdown of the SPECviewperf user interface and must fully display all of the benchmark tests being submitted. Higher desktop resolutions may be used, and while the operating system may create the appearance of a higher raster resolution, the actual raster resolution corresponds to that selected and will be noted in the results file and log files. An optional feature to adjust the raster resolution to the native resolution of the display (less a ten-pixel border) may be selected, but not all viewsets will support this method of scaling. Therefore, native resolution runs are not considered official and results are unacceptable for publication.
11. The screen resolution must be large enough to run the individual tests at their requested window size, with no reduction or clipping of test window contents.
12. Results to be made public must be generated by the official benchmark which may not be changed, except to correct errors in reporting by the benchmark, and these changes shall be disclosed voluntarily upon submission.
13. Screen DPI scaling must be set to 100%, and no UI or other window elements (including but not limited to task bars, tool bars, menu bars, and title bars) may occlude the viewport rendering context. Such elements must be configured to auto-hide or to be sized to allow all test window contents to be visible on-screen without being clipped or obscured. Results which show viewport warnings in viewset log files are unacceptable for publication.
14. Tests may be run with or without a desktop/window manager, but must be run on some native windowing system.
15. Virtualized configurations, defined as any operating system configuration running on a hypervisor or virtualization layer of any kind, must include the word “virtualized” in the comment field of the config.txt file. This information must be populated before the execution of the benchmark to ensure the results reflect this attribute.
16. Virtualized configurations as defined above must also include a declaration of the transport layer name and version used by the virtualization software in parenthesis after the graphics accelerator listed in the Graphics Accelerator field of the Graphics Hardware Configuration section of the result file.
17. Virtualized configurations as defined above must also include a declaration of the hypervisor name and version used by the virtualization software in parenthesis after the model listed in the Model field of the System Hardware Configuration section of the result file.
18. All configurations shall include a link (URL) to a publicly-accessible device driver package used to generate the submission.