Meet The 2012 Graphics Charts: How We're Testing This Year

Synthetic Graphics Benchmarks

Synthetic Benchmarks

Synthetic benchmarks tend to exercise graphics functionality at a level that is unlikely to be seen in a typical game. Consequently, they're often regarded as worst-case scenarios, and should thus be part of any benchmark suite to provide a baseline, ideally uninfluenced by IHV interference. Apart from the obligatory 3DMark 11, we also added Unigine Heaven 2.5 and Unigine Sanctuary, which we first used in 2011.

Synthetic: 3DMark 11

3DMark is one of the oldest, most widely-used synthetics, and it has been subject to its own controversies and abuses over time. Its current iteration is 3DMark 11, which provides some of the most rigorous performance testing available today. Consequently, it's the only benchmark where we loosen our resolution requirements, as there is no graphics card on the market that can tackle more than 1080p in 3DMark 11. Since our goal is to only assess graphics speed, we don't both with the CPU scores in our final result, only tallying graphics tests one through four.

Unigine Heaven 2.5

This benchmark is a classic for assessing tessellation, and it is also excellent for assessing shader performance. It's well-suited to illustrate architectural changes and other technological advancements.

Unigine Sanctuary 2.3

Unigine Sanctuary is good for evaluating dynamic lighting, HDR rendering, parallax occlusion mapping, ambient occlusion mapping, translucence, volumetric light and fog, particle systems, and post-processing. Consequently, it can stand in for older DirectX 10 games, which we're now dropping from the charts in order to keep our game selection current.