What shading throughput means for a livestream

Your livestream preview looks smooth until you add a screen capture, animated overlays, or a second display. A graphics test can help you check whether the computer can handle a repeatable graphics workload, but its score is not a direct prediction of stream quality.

The Graphics Test measures shading power and fill rate at 1080p, then gives a score out of 100. Here is how to run it fairly and use the results to investigate stutter, dropped frames, or a system that slows down during a stream.

Measure shading throughput under repeatable conditions

1. Open the Graphics Test on the computer you use for streaming. Run it on that device, not on a different computer you use to watch the stream.
2. Before testing, close demanding games, video editors, and extra browser tabs. Keep the laptop in its usual position, connect its normal power supply, and leave the display at the resolution you use for your stream.
3. Run the test and note the shading power, fill rate at 1080p, and score out of 100. These are browser-based graphics measurements, not a diagnosis of your streaming encoder.
4. If you want to compare changes, repeat the test after changing one thing at a time, such as closing a graphics-heavy app or switching to a different power mode. Keep the other conditions as similar as possible.

Do not keep running a heavy test if the computer becomes unusually hot, the browser becomes unstable, or the computer slows down sharply. Give it time to cool before another run.

What the result means

The score is steady across repeat runs

Similar results suggest the browser graphics workload is behaving consistently under those conditions. That is useful as a baseline, especially if you plan to compare the same computer before and after a driver, power, or display change. It does not prove that a livestream will be smooth, because the stream also uses video encoding, capture, audio processing, and network capacity.

The score changes noticeably between runs

A changing score can mean the computer was doing other work, its power mode changed, or heat affected performance. Browser tabs, video playback, screen recording, and other graphics activity can compete for resources. Repeat the test with those tasks closed, and compare results only when the conditions match.

The graphics score looks strong, but the stream still stutters

The Graphics Test measures shader work and fill rate, not your complete livestream pipeline. The streaming app may be using a CPU encoder, the wrong graphics processor, a demanding scene, or an output setting your system cannot sustain. Network drops can also interrupt the broadcast even when graphics rendering is fine.

The score is low or falls during longer use

A low score by itself does not establish a fault. Browser graphics results depend on the device, browser, power limits, and current workload, so there is no universal score that guarantees a good stream. If the score falls as the computer warms up, check for airflow and power limits, then compare the actual stream with a simpler scene.

How to improve livestream performance

1. Close competing graphics work. Quit games, video editors, extra video streams, and unnecessary browser tabs before going live. A browser test and a livestream can compete for the same graphics resources.
2. Check the computer's power mode. In Windows 11, open Settings › System › Power & battery › Power mode and choose a performance-focused option if available. On a Mac laptop, open System Settings › Battery and check whether Low Power Mode is limiting performance. Retest under the same plugged-in or battery conditions you will use while streaming.
3. Check which graphics processor the streaming app uses. In Windows 11, open Settings › System › Display › Graphics, select the app, and choose Options if Windows offers a GPU preference. The available choices depend on the computer. macOS usually manages graphics selection automatically, so test with the same app and power conditions you plan to use rather than looking for a universal manual GPU selector.
4. Reduce the scene workload. Temporarily remove animated overlays, browser sources, filters, or extra camera feeds. If the stream becomes smooth, restore them one at a time to find which addition pushes the system too hard.
5. Lower the stream workload if needed. Try a lower output resolution or frame rate in your streaming software, then make a short local test recording. This changes the real workload more directly than trying to reach a particular browser-test score.
6. Update graphics software and retest. Install updates through your computer maker or operating system, then run the same browser test again. Change one variable at a time so you can tell whether the update or setting made a difference.

What shading throughput can and cannot tell you

Shading throughput describes how quickly a graphics processor handles certain visual calculations, such as applying effects to pixels. Fill rate is related to how quickly it can process or write pixel data. The Graphics Test reports these measurements for its 1080p browser workload, which makes them useful for a rough comparison on the same device.

A score out of 100 is not a standard industry rating, and it does not reveal every part of a graphics processor. It does not measure the computer's exact temperature, prove that hardware encoding is available, or certify that a particular streaming scene will hold a target frame rate. Browser graphics tests also cannot identify defective video memory from a score alone.

For a practical check, make a short test stream or local recording with your usual scene and settings. If the preview and recording stay smooth but viewers see interruptions, investigate the network separately. If graphics stutter starts when you add effects or a second display, simplify the scene and compare again. For more on the limits of browser graphics scores, see why a browser benchmark cannot predict gaming FPS.