Refresh rate
6.94 msbetween frame opportunities
Timing reference
Hz ↔ milliseconds timing calculator
Translate display refresh and input polling frequency into the interval between opportunities. The frequency-to-period math is exact; the result is not a measurement of total end-to-end latency.
Hz ↔ milliseconds calculator
Use 1000 ÷ Hz to convert frequency into the time between nominal opportunities. These periods are timing context, not total latency.
Polling rate
1.00 msbetween nominal report opportunities
Frequency → interval
The conversion is simple: interval in milliseconds = 1000 ÷ frequency in hertz. A 144 Hz display cadence therefore has about 6.94 ms between frame opportunities; a nominal 1000 Hz polling cadence has 1 ms between report opportunities.
The calculator keeps display refresh and input polling separate because they describe different stages of the path. Do not add the two periods and call the result “total latency.”
A period is one cadence window, not the whole latency path
1000 ÷ Hz converts frequency into time between opportunities. Device, OS, application, compositor, display and human response remain separate parts of the path.
Common display refresh intervals
| Refresh rate | Frame interval | What the number means |
|---|---|---|
| 60 Hz | 16.67 ms | One new frame opportunity roughly every 16.67 ms |
| 75 Hz | 13.33 ms | One opportunity roughly every 13.33 ms |
| 120 Hz | 8.33 ms | One opportunity roughly every 8.33 ms |
| 144 Hz | 6.94 ms | One opportunity roughly every 6.94 ms |
| 165 Hz | 6.06 ms | One opportunity roughly every 6.06 ms |
| 240 Hz | 4.17 ms | One opportunity roughly every 4.17 ms |
| 360 Hz | 2.78 ms | One opportunity roughly every 2.78 ms |
| 500 Hz | 2.00 ms | One opportunity roughly every 2.00 ms |
Use the browser refresh-rate estimate to observe callback cadence in this tab, then use this table to translate the Hz number into frame interval.
Common polling intervals
| Nominal rate | Report interval | What the number means |
|---|---|---|
| 125 Hz | 8.00 ms | One nominal report opportunity every 8 ms |
| 250 Hz | 4.00 ms | One nominal report opportunity every 4 ms |
| 500 Hz | 2.00 ms | One nominal report opportunity every 2 ms |
| 1000 Hz | 1.00 ms | One nominal report opportunity every 1 ms |
| 2000 Hz | 0.500 ms | One nominal report opportunity every 0.5 ms |
| 4000 Hz | 0.250 ms | One nominal report opportunity every 0.25 ms |
| 8000 Hz | 0.125 ms | One nominal report opportunity every 0.125 ms |
The BenchRush polling test estimates pointer-event cadence visible to JavaScript. That browser result is not the same thing as raw USB polling, so treat the conversion as timing context rather than hardware certification.
Why these intervals are not total latency
A real click-to-photon path can include device scanning, USB or wireless transport, operating-system scheduling, application/game processing, browser/compositor timing, display scanout and pixel response. Frequency-to-period math describes one cadence window inside that chain. It cannot tell you the total delay by itself.
This is also why a higher Hz number should not be turned into a guaranteed “X ms faster” claim. The interval gets shorter mathematically; the real end-to-end outcome depends on the rest of the system and when an event lands inside each scheduling cycle.
Use the calculator with a controlled test
- Measure first. Run the polling estimate or refresh estimate.
- Translate the cadence. Put the observed or configured Hz into this calculator.
- Change one variable. If you change a display mode, browser, port or power setting, leave the rest stable.
- Compare repeated runs. Use Setup Compare for A/B values rather than reading too much into one sample.
- Check the whole-system result separately. A reaction baseline includes human response plus the visible/input path and should not be “corrected” by subtracting these interval numbers.
Questions people ask
Does a 1000 Hz mouse add exactly 1 ms of latency?
No. One millisecond is the nominal period between 1000 report opportunities per second. It is not a measurement of the mouse's total input latency or the whole click-to-screen path.
Is 240 Hz twice as fast as 120 Hz?
The frame interval is half as long: about 4.17 ms instead of 8.33 ms. That mathematical cadence change does not mean every real task or end-to-end response becomes twice as fast.
Can I add the polling interval and frame interval together?
Not as a valid total-latency measurement. They are separate scheduling periods inside a larger asynchronous pipeline, and actual waiting time depends on when events arrive plus many other processing stages.
Privacy by default
Your personal bests, XP and streak are stored locally in this browser. The current tool does not require an account or upload your individual run to a global leaderboard.