WebGL shadow mapping test
WebGL 2 renders a depth texture and filters it to shade a scene. Changing shadow quality changes map size and sampling. Compare FPS and shadow appearance under the same lighting settings.
WebGL 2
Before you start
Save your work, close competing apps, and keep the device ventilated. Use a current browser with hardware acceleration where available. Start with a short run and modest settings. Stop if the device becomes unresponsive or displays unexpected glitches.
Run and compare
Start with the defaults. Observe FPS and the workload count for 30–60 seconds. Where configurable, change only Shadow Quality, then repeat under the same conditions. Some demonstrations adapt their workload automatically; compare both the workload count and FPS. Stop with the button or Escape.
Interpret the result
A steady low FPS can simply mean a demanding workload. High FPS at a display refresh limit does not prove maximum GPU performance. Compare repeated runs with identical settings; differences between browsers also include their rendering engines.
A blank canvas or context loss can result from unsupported WebGL, browser memory limits, a driver reset, or software rendering. Reduce the workload and retry. Reproduce a problem in another application before drawing a hardware conclusion.
What this test cannot establish
This shader workload reports browser frame rate, not TFLOPS, memory bandwidth, or complete GPU utilization. GPU drivers and browser limits affect the result.
This site cannot read GPU temperature, power draw, clock speed, or utilization through WebGL. Use a separate hardware monitor for those readings. A successful run does not guarantee stability in games, rendering software, or other workloads.