How to Test Any Controller Online in 60 Seconds (No Downloads, No Software)

JoyPad Tester — Thu, 22 Oct 2026 11:14:13 +0000

Your character is moving on their own again. Or one face button only registers when you press it a little harder than the rest. Or you are about to hand over cash for a used controller and the seller swears it works perfectly.

I have been in all three spots. When one of my own controllers started drifting a while back, I went looking for a quick way to check it and found the usual mess: random software downloads, a Windows settings page that shows almost nothing useful, and tester pages buried under ads. That frustration is honestly why I built [JoyPadTester. I wanted one page that tells me everything about a controller the second I connect it.
The good news is that a real controller health check takes about a minute, and your browser already has everything it needs to do it. Here is the exact routine I run.

Why this works:

your browser can already read your controller
Modern browsers can read gamepad input directly through something called the Gamepad API. It is the same stream of data your games use, just shown to you raw: every button, both sticks, both triggers, as it happens. A browser tester simply puts a face on that data, so instead of guessing whether something feels off, you can see it.

Step 1: Connect, then press any button

Plug your controller in with USB or pair it over Bluetooth, then open JoyPadTester in Chrome or Edge, which give the most reliable results (Safari support is still limited). Now press any button on the controller.
That press matters more than people realize. Browsers deliberately hide connected controllers until you give an input, as a privacy measure. So if the page says no controller found, one button press is usually the entire fix.

Step 2: Check the sticks without touching them

This part is counterintuitive: the most important test is doing absolutely nothing.
Put the controller down and look at the resting values of both sticks. A healthy stick sits at or very near 0.000 on both axes. When I tested my drifting controller, the left stick was sitting at 0.09 on one axis with nobody touching it. Question settled in five seconds: that is drift, measured, not imagined.
Then move each stick slowly in a full circle. The trace should form a smooth, round circle that reaches the edges evenly all the way around. A flat spot or a lopsided circle means the stick has lost range in one direction, something repair folks call a circularity problem, and it will show up in games as uneven aim.
**

Step 3: Press every button, squeeze both triggers

Do one clean pass over everything: face buttons, bumpers, D-pad, stick clicks, menu buttons. Each press should register once and release cleanly. Flickering, or a single press registering twice, points to a worn or dirty switch.
Triggers are analog, so treat them differently. Squeeze slowly and watch the value: it should climb smoothly from 0 to 1. A trigger that jumps straight to 1, or stalls around 0.7 no matter how hard you press, has a problem that will cost you in racing games and shooters long before it completely dies.

Step 4: Run the vibration test

Finish with vibration. Both rumble motors should spin up together and feel even. Weak or one-sided vibration means a tired motor, and here is the thing: in an actual game you would almost never notice which side is failing. In a ten second test, it is obvious.

Reading the numbers, quickly
Resting stick values at 0.000, or tiny jitter under about 0.01: healthy, forget about it.
Resting values sitting around 0.05 to 0.08, or slowly climbing: real drift. Whether you feel it in a game depends on that game's deadzone, but it will only get worse.

Only small drift? Before spending anything, try the deadzone slider on the tester. It previews whether a slightly bigger in-game deadzone would hide the drift. That is a legitimate short term fix. It is not a cure, and anyone who tells you otherwise is selling something.
The honest limitation

A browser tester shows you exactly what your operating system reports, and nothing beyond that. It cannot measure true end to end input latency (that takes hardware, like a high speed camera setup), and it cannot repair a thing. If a stick's potentiometer is physically worn out, no website will bring it back.
What it does give you is evidence. Real numbers, in hand, before you spend money on parts, argue a warranty claim, or open the controller up with a screwdriver and hope for the best.
Who should actually run this test
Anyone whose camera or character moves on its own: drift suspicion, confirmed or cleared in a minute.
Anyone buying a used controller: run it in front of the seller. Sixty seconds is cheaper than a bad purchase.
Anyone who just cleaned or repaired a controller and wants proof the job worked.
Game developers checking raw input before blaming their own code. I have been that person too.

Does this work with PS5, Xbox and Switch controllers?

Yes. DualSense, DualShock 4, Xbox Series and Xbox One controllers, the Switch Pro Controller and most generic USB or Bluetooth gamepads all report through the same Gamepad API. If your browser can see it, the tester can read it.

Do I need to install anything, and is my data collected?

Nothing to install and no account to make. All the processing happens locally in your browser, so your controller input never leaves your own device.

My controller is not showing up at all. What should I try?

Press any button after connecting, since browsers require that first input. If it still does not appear, switch from Bluetooth to a USB cable, close Steam or any other app that might be holding the controller, and refresh the page. Chrome and Edge cause the fewest headaches here.

One last habit worth building: run this test once while your controller is healthy and note the numbers somewhere. The day something feels off, you will know exactly what changed instead of guessing.

DEV Community: JoyPad Tester