Refresh Rate Test

What is this tool?

The Refresh Rate Test times consecutive animation frames and works backwards to a refresh rate. It reports an estimated Hz figure, the median frame time behind it, how stable the frames were, and how much it trusts its own answer.

There is no browser API that returns a monitor's refresh rate, so the number is inferred from timing. The engine filters the samples before converting them — a median and a trimmed mean — so one hiccup does not move the result.

Alongside the number it draws a moving pattern. That is the part your eyes judge: whether motion looks evenly paced, which is a separate question from what the average frame time says.

How to use it

1. Pick a measurement modeSTANDARD is the everyday setting. PRECISION filters harder — it discards the top and bottom 15% of samples instead of 5% — so it is the one to use when the first answer looked unstable. MOTION prioritises the visual pattern over the readout, for judging pacing by eye.
2. Choose a duration3s, 5s, 10s, 15s or 30s. The confidence rating depends directly on sample count, so short runs cannot reach HIGH CONFIDENCE. If you want a trustworthy figure, use 10s or longer.
3. Choose a pattern and speedPatterns are Moving Bars, Grid, Moving Circle, Checkerboard and Smooth Gradient, at Slow, Normal or Fast. Moving Bars at Fast is the most revealing for uneven pacing; Smooth Gradient is the gentlest if the fast patterns are uncomfortable to watch.
4. Run it in fullscreen, then read the confidence labelFullscreen usually gives the browser a cleaner compositing path and a better sample. When the run ends, read the confidence level before the Hz number — a LOW CONFIDENCE result means re-run it rather than believe it.

What the results mean

Estimated Hz

Derived from a trimmed mean of the frame intervals: the samples are sorted, a slice is discarded from each end, and the remaining average is inverted — 1000 divided by the interval in milliseconds. It is shown to one decimal place because it is a measurement, not a lookup.

Confidence: HIGH, MEDIUM or LOW

HIGH CONFIDENCE needs at least 200 samples, 90% frame stability and jitter under 1.2 ms. MEDIUM needs at least 100 samples, 75% stability and jitter under 3.0 ms. Anything else is LOW CONFIDENCE. The label describes the quality of the sample, not the quality of your monitor.

Median frame time and jitter

Median frame time is the middle interval in the run — about 16.67 ms at 60 Hz, 6.94 ms at 144 Hz, 4.17 ms at 240 Hz. Jitter is the average change from one interval to the next. A clean run has a median close to the theoretical figure and jitter well under a millisecond.

Frame pacing rating

Excellent needs 93% stability with jitter under 0.8 ms; Good needs 82% with jitter under 2.0 ms; below that it reads Fair or Poor. Frame stability here counts frames landing within ±5% of the median interval, which is a stricter window than the FPS test uses.

The refresh-rate class

The estimate is labelled Extreme at 240 Hz and above, Ultra High at 165, High Refresh at 144, Very High at 120, High at 75, Standard at 60, Low at 30 and Very Low below that. These are naming conventions for the number, nothing more.

What it can detect

• Whether the browser is receiving frames at the rate the display is set to
• A display running at 60 Hz when it is capable of more — a common state after a driver update or a cable swap
• Uneven frame pacing, which the pattern shows and the jitter figure quantifies
• Dropped and long frames during the measurement window
• How much a fullscreen tab differs from a windowed one on the same machine

Limits worth knowing

A browser can only report what the platform gives it. These are the honest boundaries of this page, so a result is never read as more than it is.

• There is no browser API for the panel's refresh rate. The page's hardware panel says so plainly — it lists the refresh-rate API status as Not available and the physical Hz as “Not directly available”. Everything shown is inferred from frame timing.
• The estimate is bounded by the browser. If the browser caps animation frames, the estimate reflects that cap rather than the monitor's capability, which is why the page's own FAQ asks “Can a browser detect 240Hz?” and answers it carefully.
• On a variable-refresh display the refresh rate genuinely changes during the run, so a single number is a snapshot of that moment.
• The test cannot damage a display, and the page says so in its FAQ. It only draws normal moving graphics.

Tips

• Measure in fullscreen, on the monitor you actually want to test, with other tabs closed. Those three things move the number more than anything else.
• If the first run comes back LOW CONFIDENCE, switch to PRECISION mode and 15s or 30s before drawing any conclusion.
• Compare the estimate against what your operating system's display settings claim. A mismatch is the finding — it usually means the OS reverted to 60 Hz.

Troubleshooting

Related documentation

• DisplayDisplay FPS TestMeasure the frame rate your browser is actually delivering, with 1% lows and jitter.
• DisplayScreen Tearing TestScrolling patterns at chosen frame rates, so a torn frame becomes obvious.
• DisplayMotion Blur TestMoving patterns at controlled speeds, so you can judge how sharp motion stays.
• DisplayScreen Flicker TestSteady and alternating fields for spotting flicker, at brightness levels you choose.
• DisplayDisplay Response Time TestControlled colour transitions for judging pixel response, trailing and overshoot by eye.