Terms of Service
By using this tool you acknowledge the notes below — less a contract than a clear description of what the test does and how to read the result.
Last updated: November 2026Nature of the tool
VolumeShader BM is a browser-based WebGL GPU stress and performance benchmark intended for comparison. It is not a hardware diagnostic tool, does not replace professional test software, and is not purchasing, overclocking or repair advice.
Heavy-load risk
The run keeps the GPU under sustained heavy compute, similar to a demanding game. Devices may heat up, spin up fans, drain battery, drop frames or briefly stop responding; older or poorly cooled hardware may freeze or reboot. Save open work first, ensure adequate power, and decide for yourself whether to run it.
The visual contains colour and brightness changes. Users sensitive to flicker or high-contrast imagery should avoid or exercise caution.
Interpreting results
Scores depend on drivers, power mode, cooling, background apps, hardware acceleration, browser implementation and display refresh rate. Different results on the same machine in different conditions are normal. We discard the first and last 5 seconds and flag likely vsync caps or unstable runs, but cross-device comparability cannot be guaranteed.
Tiers and what gets ranked
The test offers four presets — Simple, Standard, Advanced and Extreme — which change only the fractal iteration count and the ray-marching step cap; all other parameters are identical on every device. Only the Standard tier is ranked. Results from other tiers are for local reference and never enter the leaderboard. Heavier tiers keep the GPU loaded for longer, so decide for yourself whether your machine is up to it.
Relationship to the original project
This is an independent community project based on the open-source Volumeshader_bm shader. It is not affiliated with or endorsed by the original author or any GPU manufacturer. The rendering kernel follows that project; the project link and licence details will be credited prominently once verified.
Changes
We may adjust presets, scoring or leaderboard rules. Preset changes bump the version, and older results are archived separately rather than mixed with the new board. We may suspend or modify the service and do not guarantee permanent availability of results.
Limitation of liability
To the extent permitted by law, we are not liable for hardware damage, data loss or decisions made on the basis of a result. Please assess the risk before running.