Mouse double-click test
Click the selected button in the area below, and the page records how many milliseconds apart two consecutive presses were and sorts them into three kinds: a few milliseconds is the switch contacts bouncing (a fault, commonly called double-clicking); tens to hundreds of milliseconds is you double-clicking yourself; and anything longer is two separate clicks.
Gap log
The last 20 click gaps, newest on top. The ones in red are the switch chattering — if red rows turn up when you were not clicking fast yourself, that mouse’s switch has already started to wear out.
How to test so the result means something
- Start with a dozen single clicks at normal speed. The gaps come out in the hundreds of milliseconds or more, all classified as “separate clicks”. This step is the baseline for the fast clicks that follow.
- Then double-click deliberately a few times. You will see a batch of 60–200ms “double-click” rows in the log; those are normal, not a fault.
- Finally, look for any rows of a few milliseconds. Those are the evidence of a fault — you did not click twice, yet it reported two presses. One is enough to show the switch has a problem; a few more presses confirm it is not a one-off.
Where the reading stops
This page measures the press-to-press gap in milliseconds, and its precision is whatever timestamps the browser puts on events. Without cross-origin isolation that precision is capped at 100µs, which is more than enough for a judgement on the millisecond scale — the order of magnitude here is nothing like the polling rate test, where the job is to tell 0.125ms apart; the finest this page needs to resolve is a few milliseconds.
What really deserves attention is clicks that never got recorded. This page can only count the events it receives, and a press the mouse never reports is invisible to it — yet a click swallowed by firmware debounce shows up as exactly that, a click that never got recorded. To find out whether that is happening, use the CPS test page: there you can line up how many clicks you made against how many were counted.
Common questions
A double-click is you deliberately pressing twice; chatter is the mouse reporting an extra press on its own. The test goes by time: the fastest a hand can manage press, release, press again is 60ms or more, so two clicks less than 90ms apart must come from the switch contacts bouncing. The page counts the two separately precisely so that every double-click you make does not flash up as a “fault”.
No — they are two different things. The system’s double-click speed is the threshold the operating system uses to decide whether two presses count as one double-click action; it defaults to 500ms and you can set it to anything in Control Panel. This page decides whether two recorded presses were the same physical action, and its threshold is the mechanical limit of a human hand. The two are an order of magnitude apart and do not interfere.
To leave room for your own double-clicks. If you hammer fast double-clicks one after another, the flood of 50–500ms gaps fills the log and the one or two that actually matter get drowned out. The routine is: a dozen single clicks at normal speed, then a few deliberate fast double-clicks, and finally a look for any rows of a few milliseconds hiding among them.
That is the fault in the other direction — the firmware debounce is set too long and swallows the real second press along with the bounce. It shows up as a double-click to open a file doing nothing, or a drag dropping part-way. To test for it, double-click fast on purpose and see whether the page records 2 presses or 1. Some mice let you change the debounce time in their driver.
