Free mouse drift test. Detect idle cursor drift and sensor jitter by sampling pointer events for 30 seconds while the mouse sits still. Works in your browser, no install.
Place your hand off the mouse, press Start, and wait 30 seconds. The tool samples every pointer event and reports total drift, max delta, and movement count.
Keep your hand OFF the mouse during the test. Move the cursor into the pad once, press Start, then let go. We sample every pointer event until the timer ends.
Hover your cursor inside this pad and press Start. Do not touch the mouse until the timer finishes.
Time left: 30s
Total drift
0.00 px
Max single-event delta
0.00 px
Pointer events recorded
0
Press Start and keep your hand off the mouse
Tip: laser sensors drift more on glossy surfaces. Optical sensors drift on transparent or very dark mats. A good gaming mousepad eliminates most false positives.
Was this tool helpful?Rate Mouse Drift Test:2.3(30 ratings)Thanks for rating!
Mouse Drift Test is a free, browser-based mouse testing tool that lets you detect idle cursor drift and sensor jitter by sampling pointer events for up to 3 minutes while the mouse sits still.
Cost: Free, no signup
Install: None — runs in the browser
Privacy: Runs locally, no uploads
Platforms: Windows, macOS, Linux, Android, iOS
Time: Under a minute
How to test for mouse drift while the device is still
A mouse drift test watches for browser pointer movement while you leave the mouse untouched during the selected run. It records CSS-pixel movement, event count and the largest observed jump inside the test area. Unexpected motion is a symptom to repeat under controlled conditions; it does not by itself identify sensor noise, a surface problem or a hardware fault.
How The Drift Test Works
Each pointermove event gives us a new (x, y) in CSS pixels. We compare consecutive events, sum the Euclidean distance, track the largest single jump, and count the number of events. After 10-180 seconds we report:
Total drift — sum of distances between consecutive pointer samples (px).
Max single-event delta — the biggest jitter jump in one event (px).
Pointer events recorded — how many moves the browser reported.
Zero observed displacement means the page received no position change during that run. A small nonzero value should be repeated under the same conditions; there is no universal pixel allowance that diagnoses electronic noise.
The surface, an additional pointing device, driver settings and the input path can all contribute to unwanted cursor movement. Repeat with a clean, suitable mouse surface and one active pointing device, changing one condition at a time rather than assigning a sensor type from the trace.
When Drift Is A Real Problem
Record repeatable movement with the mouse untouched, then compare another application, surface and connection where practical. A browser pixel total cannot distinguish a damaged lens, firmware, cable, wireless path or another source of pointer input. Use model-specific support guidance if the behavior persists.
Cursor Moving By Itself?
If the cursor moves when your hand is off the mouse, use the full mouse cursor moving by itself guide to separate sensor jitter from touchpad input, controller drift, surface problems, USB or wireless issues, and security concerns.
To check individual physical keys, open the keyboard tester and press each key once.
Mouse Drift Test FAQ
Common mouse drift test questions
What is mouse drift?
Mouse drift means unwanted cursor movement while you are not intentionally moving the device. This browser test records the pointer events it receives; it cannot identify whether the source is the mouse, another input device, software or the operating system.
How long should I run the drift test?
30 seconds is enough to spot obvious drift. Run 3 minutes to detect slow, intermittent drift that only shows up occasionally.
Why does my mouse drift on a glass desk?
Optical sensors need a textured surface to track contrast. Glass and very glossy desks lack texture, which causes the sensor to misread motion. Use a proper mousepad.
Is any drift at all a problem?
There is no universal CSS-pixel cutoff that establishes a faulty mouse. Repeat the same timed run with the mouse untouched, compare another surface or input device, and record the conditions. A clean run only means no unwanted movement was observed during that sample.
Check audio and video sync with a moving marker and quiet beep. Adjust sound timing to compare speakers, Bluetooth headphones and displays without a download.
01 / 10Test before buyingIs a new drag-clicking mouse necessary?Check your current clicks, settings and comfort first. Stock photos illustrate the method.02 / 10BaselineUse the same short runChoose a 5-second test and one mouse button. Keep the pointer inside the test area.03 / 10RecordSave your starting resultNote total clicks and the event pattern. Repeat a few runs; one peak is not a stable personal score.04 / 10Compare fairlyKeep the setup fixedUse the same browser, port, button and run length. Pause between runs if your hand feels tired.05 / 10Software limitThe page filter is not firmwareChanging the tool's Min gap changes which browser events it counts. It does not reprogram your mouse.06 / 10Vendor settingChange only a supported optionUse your exact model's official app. Save one supported setting, then repeat the same test.07 / 10RepeatLook for a repeatable differenceCompare several similar runs. A changed browser count alone does not prove physical latency or firmware behavior.08 / 10Desktop checkDo slow single clicks stay single?Try normal desktop clicks after changing settings. Restore the prior value if extra clicks appear.09 / 10Know the limitsA browser does not certify hardwareIt cannot certify switch type or physical latency. Check vendor support and server rules separately.10 / 10Next stepTest, then decideKeep a mouse that fits and behaves well. If replacing it, check the exact model, software support and returns.
01 / 10Shape before specsRival Pro or Sensei Pro?Start with fit, size and button reach. The wireless headline is shared. Stock photos illustrate gaming.02 / 10Start hereWhat needs to change?Note palm contact, grip pressure and button reach on your current mouse. A replacement should solve a specific problem.03 / 10Official productRival: taller supportSteelSeries describes an ergonomic shell for palm and relaxed claw. Try the exact standard or Mini size for fit.04 / 10Official productSensei: lower profileA lower symmetrical shell, aimed at fingertip and aggressive claw. Check the exact thumb-button layout.05 / 10Compare dimensionsMini is not always lowerRival Pro Mini is 40.5 mm tall; standard Sensei Pro is 38.6 mm. Compare height as well as length.06 / 10In normal playCan you reach the controls?Aim while using the actions you need together. Check thumb-button reach and avoid accidental presses.07 / 10Vendor claimRead runtime per batterySteelSeries claims up to 16 h at 8K or 60 h at 1K per battery. The 32/120 h figures combine two batteries.08 / 10Know the limits8K is not a latency resultA browser cannot certify radio stability or physical click latency. Compare settings in your actual game.09 / 10Buying decisionUpgrade for a clear needKeep a mouse that works and fits. Confirm modes, total price and returns. Check keys with the keyboard tester.10 / 10Check before replacingTest ordinary controls firstTest clicks and scrolling, then assess fit in play. A browser cannot certify hardware claims.
01 / 10Mouse showdownM42V2 vs M42The new mouse drops 14 g and seals the shell—but also moves from a PixArt 3389 to a 3327.02 / 10Weight45 g, Cable ExcludedCHERRY lists 45 g versus 59 g for the original M42. The V2 is solid, not perforated.03 / 10Modular shellTwo Shells Are Not NewBoth mice include two rear shapes. V2’s real shell change is a hole-free body with tool-free swaps.04 / 10SensorThe Sensor Trade-OffM42V2: 6,200 CPI, 220 IPS, 30 G. Original M42: 16,000 CPI, 400 IPS, 50 G.05 / 10InterpretationA Downgrade on PaperThose limits are lower. They do not prove bad tracking; independent instrumented tests still matter.06 / 10PollingSame 1,000 Hz CeilingBoth wired mice top out at 1,000 Hz. The V2 also offers 125, 250, and 500 Hz modes.07 / 10Price$39.99 US LaunchThe V2 launched at $39.99. The older M42 also showed $39.99 on sale when checked Sept. 4.08 / 10Other changesWhat Else Changed?Huano 80M-rated switches, no RGB, PTFE-free UHMW-PE glides, and no software requirement.09 / 10Buyer cautionWho Should Wait?Low-sensitivity players focused on fast-swipe tracking should wait for lab data or test within a return window.10 / 10Your turnTest Your M42V2Compare browser-observed pointer-event consistency at 1,000 Hz. A browser cannot certify raw USB polling.