Mouse Test

Click and scroll inside the pad to see buttons, wheel and double-click problems at once

Press the mouse buttons and roll the wheel inside the test pad above. Each button you press lights up on the mouse drawing, and its press count builds up in the same spot. The signal waveform below draws a bar for each button from the moment it goes down to the moment it comes up, while the cards record the gap between presses of the same button and how long it was held, in ms (thousandths of a second). When all seven inputs are detected (left, wheel and right buttons, back and forward buttons, wheel up and wheel down), the counter reads Inputs seen 7 / 7.

If you are worried that one click opens things twice, choose Double-click check and press the same button slowly, about once per second, 20 times. A press that registers again within 100 ms of release is not a click you made, so it is counted separately as an extra press. To find out how many clicks per second you manage, pick 1, 5, 10, 30 or 60 seconds under CPS test. Everything is processed inside this browser; nothing is saved or sent to a server.

When to check a mouse: double presses, broken drags, jumpy wheels

Mouse faults usually show up only now and then, which makes the cause hard to pin down. If you repeat the same action in the situations below and record the signals, it becomes easier to tell a settings problem from a hardware problem, and to explain it to the seller.

• One click opens a folder or selects twice: when a switch contact bounces (chattering), a single press can be split into press, release, press and arrive at the computer twice. Run the Double-click check and see whether extra presses keep appearing.
• A dragged file drops halfway: if the contact separates briefly while you hold the button, the drag breaks. Hold the button down for about 2 seconds; on the waveform a break shows up as one bar split in two, with a highlight at that spot.
• The page jumps up while you scroll down: a worn wheel sensor can mix in reverse signals even when you roll in one direction. Watch whether Reverse blips on the wheel card keeps increasing.
• Buying or selling a used mouse, or receiving a new one: confirm that all five buttons and both wheel directions register, then use Copy results to keep a record you can refer to after the deal.
• After changing gaming mouse settings: use the movement report rate to roughly confirm that the polling rate you chose in the vendor software has really changed.

How to use it

1. Button check: keep the pointer inside the test pad and press and release the left and right buttons one at a time. Then press the wheel button (pushing the wheel straight down), the back and forward buttons on the side, and roll the wheel up and down. Skip any button your mouse does not have. Inside the pad, the context menu, wheel scrolling and the side buttons' page navigation are blocked.
2. Read the gaps and hold times: press the same button a few times and look at the latest, shortest and average gap on its card, plus the hold time. The length of a bar on the waveform is the hold time.
3. Double-click check: choose Double-click check on the bar above, pick the button to test (left, right or wheel button), then press it slowly, about once per second, 20 times. The 20 dots fill in one by one, and a click that produced an extra press turns into an accent-colored dot.
4. Movement report rate: move the mouse quickly in big circles inside the pad. The scale in the Button check panel shows the current value and the highest value so far.
5. CPS test: choose CPS test, pick the duration and the button to count, and start pressing that button inside the pad. The clock starts with your first press. When it ends you get the CPS and a bar chart per interval; press Measure again to start over. Pressing Esc or switching to another window cancels a measurement in progress.
6. Keeping a record: if you want to compare results, use Copy results, paste them into a note, then Reset. The record disappears when you reload or close the page.

The five buttons are told apart by a number the browser assigns

A web page does not see where a button physically sits on the mouse. It receives the button number (MouseEvent.button) assigned by the operating system and browser. If you have swapped the primary button to the right side for left-handed use, it is normal for pressing the right button to increase the Left button (primary) card.

If you have remapped a side button to a keyboard shortcut, it arrives as a key signal instead of a mouse button. In that case, check separately with a keyboard test tool that the key comes through. Pressing two buttons at the same time is counted for each button. Pointer events do not send a new press signal when a second button goes down while another is already held, so this tool counts with the per-button mouse signals (mousedown and mouseup) instead.

Double-click problems are read as a "re-press after release"

The switch under a mouse button is a metal contact that closes and opens. At the moment it closes or opens it vibrates briefly (chattering), and the mouse firmware ignores that vibration for a set time (debounce). For example, Pulsar's support page tells you to raise the debounce value if the mouse double-clicks, and says that needing to raise it beyond 10 ms points to a faulty switch. The debounce settings on the same company's eS model are 0, 3, 5, 7, 10 and 12 ms. When a worn contact bounces for longer than that, an extra "press" arrives within a few ms of releasing or pressing.

That is why this tool separately measures the time from releasing a button to pressing the same button again (the re-press gap). If this value is 100 ms or less, it is counted in the card's Re-press ≤100 ms and highlighted on the waveform (exactly 100 ms is included). Looking only at the gap from one press to the next would miss bounce that happens at the end of a long press. In the third row of the table below, the button was held for 180 ms and bounced on release: the gap between the two presses is 184 ms, but the re-press gap is only 4 ms.

As the last row shows, deliberate fast clicking also pushes the re-press gap below 100 ms. That is why short re-presses naturally pile up in the Button check and CPS test. To judge extra presses you need the condition that every press is slow and single, as in the Double-click check. When you press about once per second, the gap between releasing and pressing again is normally several hundred ms, so a press that registers again within 100 ms was not made by you. If your deliberate clicks follow each other less than 0.25 seconds apart, the test panel tells you to slow down and try again.

How to read the Double-click check result

• 0 extra presses in all 20: no bounce showed up in this test. If the problem only happens now and then, run it a few more times, using the button you use in the app where it occurred.
• Only one extra press: your finger may simply have slipped and pressed twice. Repeat under the same conditions two or three more times and see whether it recurs.
• Extra presses in every test: save the result with Copy results, then run the same test on another computer or another USB port. Logitech's support page also advises, when a single press produces two or more clicks, turning off the Windows option for opening items with a single click, and contacting customer support if the problem also appears on another computer.
• If the vendor software has a debounce setting: raise the value one step and test again. Writing down the value and the result helps when you contact support.
• The count rose by less than 20 after 20 presses: this is the opposite of an extra press, a press that went missing. For a wireless mouse, check the battery and the distance to the receiver; for a wired one, try a different cable and port.

The browser's dblclick signal is a different record from an extra press. The browser sends a dblclick signal when two clicks fall within a set time, and that time follows the operating system setting. Windows uses 500 ms when the value has not been set and allows up to 5,000 ms. So even slow presses can raise the dblclick count if they are within 0.5 seconds, and that alone does not indicate a fault.

Wheel signal count and reverse blips

When you roll the wheel, the browser sends a wheel signal carrying a vertical amount (deltaY) and a horizontal amount (deltaX). A negative vertical value counts as up and a positive one as down. The count is not the number of "clicks" the wheel notches through but the number of signals the browser sent. Depending on the operating system and browser, one notch may send one signal, or a high-resolution wheel may split it into many. Two-finger scrolling on a trackpad sends the same signal. Zoom gestures made by holding Ctrl while scrolling, and signals with zero movement, are not counted.

A reverse blip is when, after you have rolled in one direction two or more times, a signal in the opposite direction sneaks in once or twice and then the original direction resumes. It is only counted when every pair of neighboring signals is 200 ms apart or less. Rolling down and then deliberately turning back up is not counted, because the opposite direction continues. On the wheel row of the waveform, down signals are drawn on the lower tick and up signals on the upper tick, with blips in the accent color.

How the movement report rate (polling rate) is measured

Polling rate is how many times per second the mouse reports a change of position to the computer. Logitech explains that at 125 Hz the mouse reports once every 8 ms and at 1,000 Hz once every 1 ms. This tool counts every movement signal received inside the pad and works out the rate over the last 0.5 seconds. Browsers bundle movement signals once per screen frame, so the tool uses the function that unpacks them (getCoalescedEvents) to count each individual signal. It is supported from Chrome 58, Firefox 59 and Safari 18.2; without it, only the number of signals bundled per frame is known, so the tool does not measure.

A mouse only reports while it moves, so slow movement gives a low value. To see the highest value, move big and fast. The quality of a wireless link and the browser's limit on time precision (performance.now() is usually in 100 µs steps) also affect the value, and at 4,000 Hz and above the browser or operating system may thin out the signals so the result comes out lower than the setting. A few percent of difference from the setting is therefore normal, and the tool is better used to tell apart different steps such as 125 Hz and 1,000 Hz.

CPS is the number of presses within the chosen time, counted from the first press

CPS (clicks per second) is the number of presses of the chosen button within the measurement time, divided by the time in seconds. The moment of the first press is the start, and that first press counts as 1. The interval includes the start and excludes the end: [start, end). With 31 presses in a 5-second test, the result is 31 ÷ 5 = 6.20 CPS. Even if you stopped clicking early, the divisor is the whole 5 seconds you chose, not the time until your last click.

A single number cannot tell a run that started fast and faded from one that stayed even to the end. That is why the result shows the number of presses per interval as bars, and lists the CPS of the first half · second half separately. For example, if you press 36 times in the first 5 seconds and 26 times in the last 5 seconds of a 10-second test, the total is 6.20 CPS, but the first half is 7.20 and the second half 5.20 CPS. When you compare, keep the duration and the button the same. If you want to know how quickly you react after seeing a signal rather than how fast you can repeat a press, the reaction time test is the right tool.

Common mistakes

• Pressing outside the pad: clicks outside the pad are not counted. If you press the button inside the pad and release it outside, the release is still counted.
• Reading a short re-press as a fault: when you click fast, a re-press gap below 100 ms is normal. Judge with the Double-click check, where you press slowly.
• Treating the browser dblclick signal as an extra press: a dblclick signal only means two clicks fell within the operating system's double-click time.
• Reading the wheel signal count as the number of notches: with smooth scrolling or a trackpad, one roll sends several signals.
• Judging the polling rate as low after moving slowly: look at the highest value reached while moving fast.
• Testing with a finger on a phone: touch and pen input are not mouse signals, so they are not counted. You can test by connecting a mouse to a phone or tablet.

Limits

• This tool sees the signals that reach the browser. It does not measure the moment the switch really closed, the mouse's input latency, or sensor accuracy (DPI).
• The 100 ms re-press criterion and the 200 ms reverse blip criterion are observation criteria set by this tool. They are not a manufacturer's defect standard; for exchange or repair, follow the seller's and manufacturer's guidance.
• Browsers may deliberately reduce time precision (performance.now() is usually 100 µs, and 5 µs on cross-origin isolated pages). Do not trust differences within 1 ms at face value.
• If the operating system, browser or mouse vendor software handles the side buttons and wheel button first, the signal never reaches the page. The pad blocks page navigation, but this cannot be guaranteed on every device, so be careful testing the side buttons in a tab that holds important work.
• A trackpad and a mouse send the same kinds of signals, so a browser alone cannot tell them apart.
• Results are not saved, and there are no member records or rankings. This describes how the measurement data is handled and does not mean the site as a whole makes no network requests.

Sources

• 01MDN: MouseEvent.buttonButton numbers: 0 primary, 1 middle, 2 secondary, 3 back, 4 forward; roles swap in left-handed settings
• 02W3C: Pointer EventsWhen another button is pressed while one is held (chorded), no separate pointerdown or pointerup is fired
• 03MDN: Element dblclick eventThe dblclick signal that follows two clicks
• 04Microsoft Learn: SetDoubleClickTimeWindows default double-click time 500 ms, maximum 5,000 ms
• 05MDN: Element wheel eventDifference between a wheel signal and actual scrolling; trackpad input
• 06MDN: WheelEvent.deltaModeMovement units: 0 pixels, 1 lines, 2 pages
• 07MDN: PointerEvent.getCoalescedEvents()Individual signals bundled into one movement event; supported in Chrome 58, Firefox 59, Safari 18.2
• 08MDN: performance.now()Time precision 100 µs, 5 µs on cross-origin isolated pages
• 09Logitech: What Is Polling Rate on a Mouse?125 Hz = a report every 8 ms, 1,000 Hz = every 1 ms
• 10Pulsar Support: How to use the mouse Windows softwareDebounce time: raise it if the mouse double-clicks; beyond 10 ms suggests a faulty switch
• 11Pulsar Support: Default settings for eS modelDebounce settings 0, 3, 5, 7, 10, 12 ms
• 12Logitech Support: Gaming mouse double-clicks when clicked onceTurn off single-click opening, test on another computer, contact support if it persists

Sources checked 2026-10-21. The 100 ms re-press, 200 ms reverse blip and 0.25 second "too fast" notice are observation criteria set by this tool.

Frequently asked questions

Related tools