Why a GPU benchmark drops in a background tab

Your graphics benchmark looks fine until you switch to another tab, then the score drops or the test appears to crawl. That change can be alarming, but a hidden browser tab does not always get the same access to graphics and processing time as the tab on screen.

Here is how to check whether tab behavior explains the drop, compare scores fairly, and decide when a low result might point to a separate graphics problem. The Graphics Test measures shading power and fill rate at 1080p.

Compare the GPU test in the foreground and background

1. Open the Graphics Test and run it with its tab visible. Let the test finish without switching away or using other demanding apps.
2. Note the score and the shading and fill-rate results. These are browser test results, not a direct measurement of every graphics task your computer can do.
3. Run the test again under the same conditions. If the two visible-tab results are close, you have a useful baseline. If they vary a lot, close other demanding apps and repeat before drawing a conclusion.
4. For a comparison, start another run and switch to a different tab briefly. Return to the test, let it finish if it is still running, and note what changed. Do not treat this hidden-tab run as equivalent to the visible-tab baseline.
5. Run the test once more with its tab visible. If the score returns near the baseline, the drop is probably tied to background-tab behavior rather than a sudden change in the GPU.

What the result means

The score drops only when the tab is hidden

This is usually a browser scheduling effect. Browsers often reduce work in hidden tabs to save battery and processing time. They may slow or pause animation updates, including the repeated drawing a graphics test needs. If fewer frames are produced during the test, its result can fall even though the graphics hardware has not changed.

Use the visible-tab result when you want to compare the device with its own earlier test or with another computer. Keep the tab in front for the whole run.

The test pauses and catches up when you return

The browser may have limited updates while the page was hidden. A test that appears frozen in the background is not automatically evidence that the GPU stopped working. Run it again in a visible tab and watch for a smooth, consistent result.

The score stays low with the tab visible

Background throttling does not explain a result that remains low during a foreground run. Other possibilities include a power-saving mode, heavy activity from another app, a browser using software rendering instead of GPU acceleration, or a graphics limit in the device. A browser benchmark cannot identify a defective graphics chip by itself.

Results vary even when the tab stays visible

Short runs can be affected by other work on the computer, heat, power settings, and browser scheduling. Close demanding apps, wait for the computer to cool if it is hot, and repeat the test a few times under similar conditions. Look for a consistent pattern rather than relying on one score.

How to reduce score changes

1. Keep the test visible. Avoid switching tabs, minimizing the browser, or covering it with another window during a run. A screen lock or display sleep can also interrupt the test.
2. Close demanding work. Pause games, video exports, 3D apps, and video streams before comparing results. These can use the GPU or processor at the same time as the browser benchmark.
3. Check browser graphics acceleration. In Chrome or Edge, open the browser's settings and look under System or System and performance for Use graphics acceleration when available. Turn it on if it is off, then restart the browser. The wording and location can differ by version.
4. Use a consistent power mode. On Windows 11, open Settings › System › Power & battery › Power mode and choose a performance-focused option for the comparison. On macOS, open System Settings › Battery and check that Low Power Mode is not limiting performance. Keep the laptop plugged in for repeat tests, if convenient.
5. Check whether the browser discards inactive tabs. Some browsers have a memory-saving feature that can unload tabs you have not used recently. If the test reloads when you return, look in the browser's performance settings and keep the test page active or exempt it from tab sleeping during the run.
6. Repeat after a restart. If visible-tab results remain unexpectedly low, restart the browser and computer, then test again before changing graphics drivers or other system settings.

When it may be a graphics problem

Consider further checks if the visible-tab score is consistently much lower than before, and you also notice graphics glitches, crashes, or poor performance in other applications. Compare only results from the same browser, device, power mode, and similar room conditions. A browser test can show a change in its own workload, but it cannot prove that a GPU is damaged or measure every part of its performance.

If the issue began after a driver or system update, check for an available graphics driver update from your computer or graphics hardware manufacturer. If problems continue across multiple applications, save any error messages and contact the device maker or a repair professional before replacing hardware.

Why browser tests behave differently in the background

A browser page is not a desktop benchmark with guaranteed access to the computer. The browser decides how much attention each tab receives, and may reduce work on a page that is not visible. Battery-saving features and memory-saving features can add more variation. The exact behavior depends on the browser, operating system, and power state.

For repeatable results, compare foreground runs and keep the setup consistent. Treat a background run as an example of what happens when the page is hidden, not as a clean measure of the GPU's full performance.