Reaction time test. You, and what your setup adds.
Every reaction time test measures you plus your screen, mouse and browser. This one says how much of that it can account for. It times each trial from the frame the green was drawn in to the click's own timestamp, measures the screen's refresh rate and, if you let it, the mouse's report interval on this page, and shows your median with that share taken out, as a range. The rest of the chain needs hardware to measure, and the page says so.
One check · Five trials
Asks for: A mouse, a touch screen or the Space bar. It runs in its own window and nothing leaves this page.
In short
Reaction time test: the answer first
A reaction time test times you from a change on screen to your click, and every online one includes the screen, the mouse and the browser. This one measures the refresh rate and the mouse's report interval on the page, shows how much of your median they are estimated to add, and gives your time with that share taken out, as a range.
Reading your result
- The median, not the best
- One fast trial can be a lucky guess. The median of five is the figure to compare.
- The setup's share is a range
- The screen's share depends on when the frame is shown and where on the screen the box sits, so it is a range, from the refresh interval measured here.
- What is still inside
- The panel's processing and pixel response, the mouse's switch and the path to the browser stay in the figure, because only hardware can measure them.
What it measures
Measured here, not assumed.
- Your time
- From the frame the box turned green in, its requestAnimationFrame timestamp, to the click's event timestamp. Five trials after a random wait of 1.5 to 4 seconds (our values); the median is the figure. A click before green, or within 100 ms of it (our line), is a guess and is not counted.
- The screen's share, estimated
- The refresh interval measured from frames on this page. Our model: the changed frame reaches the middle of the screen between half and one and a half refresh intervals after it is drawn. At 60 Hz that is 8 to 25 ms; at 240 Hz, 2 to 6 ms.
- The mouse's share, estimated
- Move the mouse over the pad and its report interval is measured with the mouse test's own method. A click waits up to one interval for the next report.
- You, with that taken out
- The median minus the high and the low ends of the estimate: a range, never one falsely exact figure.
- The browser, beside it
- From the click's timestamp to the page's handler, and to the next paint where the browser reports it (Event Timing, rounded to 8 ms). Not subtracted, because they come after the timestamp the trial is timed to.
If it fails
The usual failures, and what fixes each.
- Your times vary a lot
- Sit as you would to play, keep a finger resting on the button, and run more trials. The median ignores the odd slow one.
- The screen's share is large
- A higher refresh rate shrinks it. Check that the display runs at its rated rate in the system's display settings; the refresh rate test shows the rate it really runs at.
- The refresh rate reads low
- On a laptop, plug it in: many lower the refresh rate on battery. Close other tabs that animate.
Where a browser stops
What it cannot tell you.
- End-to-end latency. From the screen changing to the click arriving needs a photodiode on the screen and a sensor on the button; no browser can measure it.
- The panel's own processing and pixel response, which add to every trial and stay inside the figure.
- The mouse's switch and debounce, and the USB and system path before the browser stamps the click.
- A ranking against other people. There is no population here to compare with, so none is shown.
Privacy
Nothing leaves this page.
The camera and microphone are read on this page and stopped as soon as they have been measured. The clip you can play back lives in the browser's memory and is discarded when you leave. The location check discards the position. The internet test sends random bytes to this site and nothing else. The result is stored nowhere unless you ask for it by email, and then only the fact that an email was sent is kept.
Questions
Reaction time test, answered
- Are online reaction time tests accurate?
- They measure you plus your screen, mouse and browser. This test measures the screen's refresh and the mouse's report interval and shows your time with them taken out, as a range. The panel's own delay still needs hardware to measure.
- What does the screen add to my reaction time?
- At least part of a refresh interval: the changed frame has to be shown and scanned out. This page measures the interval and estimates that share as half to one and a half intervals.
- Can I take the test on a phone?
- Yes, tap the box. A touch screen's own scanning adds a delay this page cannot measure, so compare phone results with phone results.
- Does a 240 Hz monitor improve reaction time?
- It shortens what the screen adds: this page's estimate of the screen's share is half to one and a half refresh intervals, 8 to 25 ms at 60 Hz and 2 to 6 ms at 240 Hz. Your own reaction does not change.
Read next