An online reaction time test measures from the moment the page drew the change to the moment the browser stamped your click. That span includes the screen, the mouse and part of the browser, not only you.
- The screen adds time before you can see the change. Our model takes half to one and a half refresh intervals: 8 to 25 ms at 60 Hz, 3 to 10 ms at 144 Hz and 2 to 6 ms at 240 Hz.
- The mouse adds up to one report interval: up to 8 ms at 125 Hz, up to 1 ms at 1000 Hz.
- The panel's processing and pixel response, the mouse's switch and the path to the browser cannot be measured from a page. They need a photodiode and a sensor on the button.
The short answer
A reaction time test measures you plus your setup. The reaction time test on this site measures the refresh interval and the mouse's report interval on the page, estimates what they add, and gives your share as a range. It does not rank you against other people: there is no population behind it to rank against.
We have no recorded runs of this test in our validation notes yet, so this guide quotes no measured results. Every line and constant below is the one the test's code applies.
What is timed
Each of 5 trials waits a random 1.5 to 4 seconds, then changes colour. The time runs from the animation frame the change was drawn in to the click event's own timestamp. A click before the change is early; one within 100 ms of it is a guess, our line, and neither is counted. No click within 2 seconds drops the trial. The figure is the median, because one fast trial can be luck.
What the setup adds, and what is still inside
A frame begun at the change is shown at the next refresh at the earliest, and the scan reaches the middle of the screen half a refresh later, so the screen's share is a range. The mouse's click waits for its next report. The test subtracts both, as ranges, using the intervals measured on the page.
Still inside the figure: the monitor's own processing and pixel response, the mouse's switch and debounce, and the USB and system path before the browser stamps the event. The browser's delay from the event to the next paint is reported beside the result, not subtracted, because the click is timed from the event's own timestamp.
Comparing two results
Compare medians from the same setup, the same browser and the same time of day. A faster monitor or mouse shortens what the setup adds, by the amounts above, and leaves your own reaction where it was.
Terms used here
Questions
- Why is my reaction time different on another site?
- Each site times from a different point, draws differently, and runs on whatever screen and mouse you have. Two figures from different sites or setups are not the same measurement.
- Can a browser measure end-to-end input latency?
- No. From the screen changing to the click arriving needs a photodiode on the screen and a sensor on the button. A page can measure only the parts it times itself.