LEFT STICK
CORRECTED X0.0000
CORRECTED Y0.0000
CORRECTED DISTANCE0.00%
Controller Lab // Calibration Reference
CONTROLLER CALIBRATION TEST
Inspect the neutral position of both sticks, set the current position as a temporary zero reference, and record the raw axis range you can reach. This helps separate a centre-offset problem from limited or uneven stick travel.
TEMPORARY ONLYThe zero adjustment on this page exists only in your browser and does not rewrite controller firmware or operating-system calibration.
Live diagnostic
Neutral offset and axis range
AWAITING CONTROLLER
Connect a controller, place both sticks at rest, then press any button.Keep your hands off the sticks before setting the temporary zero reference. Standard mapping normally uses axes 0–3 for the two sticks; non-standard devices can expose different controls on those indices.
RIGHT STICK
CORRECTED X0.0000
CORRECTED Y0.0000
CORRECTED DISTANCE0.00%
RAW AXIS RANGE OBSERVED
LEFT X MIN / MAX0 / 0
LEFT Y MIN / MAX0 / 0
RIGHT X MIN / MAX0 / 0
RIGHT Y MIN / MAX0 / 0
Neutral reference has not been sampled yet. GearCheck Lab averages roughly 0.7 seconds of browser values and rejects the reference if obvious stick movement occurs during the sample.
For a useful range check, rotate each stick around its full physical edge several times after resetting the sample. Values near -1 and +1 are common on standard axes, but exact endpoints vary by controller and calibration.
How to run a useful calibration check
- Connect the controller, press a button if needed, place it on a stable surface and release both sticks.
- Watch the raw centre values for 10–15 seconds before changing anything. Note whether the offset is steady or wandering.
- Select Set current position as zero, then leave the sticks untouched during the 0.7-second sample. Moving axes are rejected instead of being saved as a reference.
- Select Reset range sample. Rotate each stick around its full edge twice, then make two slow horizontal and vertical passes.
- Compare left/right symmetry, repeatability and whether the corrected centre remains still. Finally, reset the temporary zero and confirm the original behavior returns.
What this browser test actually changes
The page subtracts a short average of axes 0–3 from the values shown in this tab. It does not write firmware, change an operating-system profile or alter a game's deadzone.
On a standard Gamepad API mapping, axes 0–3 usually represent the two sticks. On a non-standard mapping, verify which physical controls move these fields before interpreting them.
No automatic pass/fail verdict: controller models, firmware and game deadzones differ. Use repeated patterns and manufacturer guidance, not one universal endpoint or centre threshold.
How to interpret the patterns
| Observed pattern | What it suggests | Next check |
|---|---|---|
| Stable off-centre value that disappears after temporary zeroing | A repeatable static offset. A supported device calibration or modest in-game deadzone may compensate for it. | Reset the browser zero, repeat the neutral sample three times, then use the manufacturer's calibration path if all samples agree. |
| Centre keeps moving after zeroing | The input is changing rather than holding one fixed offset. Temporary calibration cannot remove that behavior. | Run the Stick Drift Test, compare another USB/Bluetooth connection, and check firmware. |
| One direction repeatedly stops short | Possible restricted range, calibration mismatch or physical wear. Exact endpoints vary, so repeatability matters more than reaching exactly ±1. | Reset the range sample and repeat two slow full-edge rotations. Compare with the Circularity & Range Test. |
| Values jump or skip during slow movement | Possible noisy input, dead spots or a non-standard axis mapping. | Confirm the mapping, reconnect the controller and compare the same movement in the full Controller Test. |
Calibration, deadzone and repair are different
- Calibration teaches supported software or firmware where centre and travel limits are.
- Deadzone ignores a small area around centre; it can hide minor input but also reduces fine control.
- Repair addresses changing, noisy or mechanically limited input that software compensation cannot reliably correct.
Do not increase a deadzone until the symptom disappears without recording the original values. That makes later comparisons harder and can mask a worsening fault.
Browser and hardware limits
- The Gamepad API exposes browser-normalized values, not electrical sensor measurements.
- Some browsers reveal a controller only after a button press; mobile and embedded browsers may not support gamepads.
- Wireless interference, low battery, adapters and remapping software can change what reaches the browser.
- This page cannot identify contamination, wear or a failed component by itself.
- Manufacturer or console calibration may use device-specific data that a web page cannot access.
Controller calibration FAQ
Does this page recalibrate my controller?
No. The temporary zero reference affects only the values displayed in this browser tab and resets when you reload or leave.
Should I calibrate away stick drift?
A small, stable centre offset may be manageable with a supported calibration or deadzone. A centre that wanders or keeps moving after zeroing is evidence of a changing input problem, not a fixed offset.
Why do my axes not reach exactly -1.000 and +1.000?
Small endpoint differences can be normal. Compare repeated passes, direction symmetry and whether the controller reaches the input expected by games instead of relying on one universal number.
Why are the calibration controls disabled?
The controls remain disabled until the browser detects a connected gamepad. Some browsers require a button press after the controller is connected.
Does the test upload controller input?
No. The page reads the browser Gamepad API and performs its calculations locally in the current tab. Read the privacy policy for the site's broader data practices.