Posted on Thursday, July 18 2019 @ 9:05 CEST by Thomas De Maesschalck
UL Benchmarks announces they will introduce the 3DMark Variable-Rate Shading benchmark on August 26. This new test will measure the performance gained by enabling Variable-Rate Shading (VRS), a DirectX 12 feature that can selectively reduce the level detail in certain parts of the frame, without having a noticeable impact on image quality. The test will be available for 3DMark Advanced Edition and for 3DMark Professional Edition.
What is Variable-Rate Shading?
Variable-Rate Shading (VRS) is a new feature in DirectX 12 that allows developers to improve performance by selectively reducing the level of detail in parts of the frame where there will be little noticeable effect on image quality.
Shading rate refers to the number of pixel shader operations called for each pixel. Higher shading rates improve accuracy but are more demanding for the GPU. Lower shading rates improve performance at the cost of visual fidelity.
With Variable-Rate Shading, developers can vary the shading rate within a single frame. By using VRS to lower the shading rate for parts of the frame that are in deep shadow, far from the camera, or peripheral to the player's focus, for example, a game can run at a higher frame rate with little perceptible loss in visual quality.
3DMark VRS feature test
3DMark feature tests are special tests designed to highlight specific techniques, functions or capabilities. The 3DMark VRS feature test is designed to help you compare differences in performance and image quality when using variable-rate shading.
The test features a forest scene with floating lanterns. VRS is disabled on the first pass of the test to provide a baseline for comparison.
Variable-rate shading is enabled for the second pass of the test.
With VRS, a single pixel shader operation can be applied to a block of pixels, for example shading a 4x4 block of pixels with one operation rather than 16 separate operations.
In the 3DMark VRS feature test, the shading rate varies with camera distance when VRS is enabled: 4×4 for the furthest objects, (the blue areas in the image below), 2×2 for the middle-distance geometry (green areas), and 1×1 (red areas) for the closest foreground objects.
The test reports the average frame rate for each pass and calculates the performance gained with VRS. There is also an interactive mode that lets you change VRS settings on the fly and export frames for comparison.
