TryYourGear

Refresh Rate Test

Press Start and keep this tab visible. The test estimates how often your browser draws frames over a few seconds.

Ready
Ready

The test takes about 4 seconds.

For the most reliable result, keep this tab visible and avoid switching tabs while the test is running.

This is an estimate based on how often your browser renders frames. It does not read your monitor's refresh rate directly.

What does this test measure?

It measures how often your browser renders frames while this page is visible, using requestAnimationFrame. That observed rate can approximate your display's refresh rate, but it is not a direct reading of the monitor's hardware.

Common refresh rates include 60, 75, 120, 144, 165 and 240 Hz.

The result can be affected by:

  • Your browser and operating system
  • Your graphics card (GPU) and drivers
  • VSync and variable refresh rate (VRR, G-Sync, FreeSync)
  • Page performance and system load
  • Background tabs and dropped frames

This is an estimate based on the frames your browser renders, measured with requestAnimationFrame. It is not a direct or certified reading of your monitor's hardware, and it can be affected by the browser, operating system, GPU, VSync, VRR (G-Sync/FreeSync), page performance, background tabs, dropped frames and system load.

How the three-second measurement runs

  1. Warm-up (half a second)

    The page draws frames for half a second that are thrown away, so the first, often uneven, frames don't count.

  2. Measurement (three seconds)

    For the next three seconds it records the timestamp of every frame the browser draws, using requestAnimationFrame.

  3. Analysis

    It needs at least 30 frames. From the gaps between them it works out a frame rate and an estimated refresh rate.

  4. Interruption

    If the tab stops being visible, the run is stopped and its data discarded, because a hidden tab is drawn less often.

How the estimate is built

Two figures come from the same frames. The frame rate is the number of gaps between frames divided by the real time that passed. The estimated refresh rate is the average of the regular gaps only: a gap longer than 1.5 times the median counts as a stalled or dropped frame, is reported separately and is left out of the estimate, so a few drops don't drag the number down.

If the estimate falls within 3% of a common rate (60, 75, 90, 100, 120, 144, 165, 180, 240 or 360 Hz), the page says the result is consistent with that rate. It never says the monitor is running at it.

The readouts, with a worked example

What each readout in the Refresh Rate Test results shows
ReadoutWhat it shows
Estimated refresh rateThe large number: 1,000 divided by the average of the regular gaps between frames, rounded to whole hertz.
Frame rateThe number of gaps divided by the seconds measured, with one decimal. Unlike the estimate, it counts the stalled gaps too.
Average frame intervalThe measured time divided by the number of gaps, in milliseconds. Stalled gaps are included.
Minimum and maximum intervalThe shortest and longest gap between two frames. A maximum far above the average means at least one frame was held up.
Frames observed and measurementHow many frames were recorded and how long the window really lasted, close to three seconds.

An invented run shows how they fit together. Suppose the page records 181 frames over 3.00 seconds. That is 180 gaps, so the frame rate is 180 ÷ 3.00 = 60.0 FPS and the average interval is 3,000 ÷ 180 = 16.67 ms. Since 1,000 ÷ 16.67 is about 60, the large number reads 60 Hz and the page says the result is consistent with 60 Hz.

Now suppose that run came from a monitor sold as 144 Hz. The numbers alone can't say whether the display is set to 60 Hz, the cable or port limits it, or the graphics settings do; that is the question the 60 Hz guide takes step by step. The large number is also shown with a leading ≈ when the frame timing was irregular, and the page then says how many gaps stalled.

Reading the result

What you might see in the Refresh Rate Test, what it usually means and what to try
What you seeWhat it usually meansWhat to try
About 60 on a monitor sold as 120, 144 or 240 HzThe display may be set to 60 Hz, or the cable, port or graphics settings limit it. The measurement can't say which link is responsible.Follow the 60 Hz troubleshooting guide, changing one thing at a time.
A value close to the rate you expectYour browser is drawing frames at a rate consistent with that setting.That's reassuring, but it doesn't certify the panel or its real refresh rate.
A value that matches no common rateVariable refresh rate, power saving, a busy system or an unusual display mode can all cause it.Repeat with other apps closed and power-saving options off.
The page reports irregular framesFrame timing wasn't steady: system load, other tabs or dropped frames.Close other tabs and apps, keep this tab visible and run it again.
No reliable estimateToo few frames, or too irregular a rhythm, to give a number.Try again with the tab in front and nothing heavy running.

What this test can't tell you

It measures the frames your browser renders, not the panel itself, so it can't read your monitor's real refresh rate or certify it. The browser or the operating system can limit how often a page is drawn, and variable refresh rate can make the number differ from the nominal one. With more than one display, the result normally reflects the display the window is on, so drag the window to the monitor you want to test before you start. It also can't see the resolution, the capability of the cable or the response time. The guide on what browser tests can and can't measure explains why a web page can't read these values.

What to do next

If the number is lower than your monitor should give, work through the 60 Hz guide in order: display settings first, then the connection, then graphics settings. Comparing the result before and after each change is what makes the test useful. The Monitor Test checks the panel itself for dead or stuck pixels, and the Reaction Time Test shows how much your display and input matter in timed responses.

Try next