Stick Jitter and Noise Test

Hold your thumb perfectly still on an analog stick and the number your controller reports should sit perfectly still too. Often it does not. It hisses around by a few thousandths, or it jumps once every few seconds, or it steps between fixed values instead of moving smoothly. That wandering is jitter, and it is the reason a slow tracking shot creeps when you are not asking it to and a hovering drone will not stay put. This test holds a magnifying glass over a stationary stick and measures exactly how much the reading moves when you are not moving it.

Measure Your Stick Noise

Results

Axis breakdown

Would a game see this

Report

The CSV holds every individual reading with its timestamp, which is what you want if you plan to plot it yourself. The JSON holds the finished numbers.

How to Run It Properly

1. Use a cable for your first run. Wireless adds its own noise on top of the stick, and you want to know which one you are looking at.
2. Press any button on the controller. Browsers keep gamepads hidden until they see one input.
3. Put the controller down on a desk for the resting test. Holding it in your hands adds tiny tremors that are yours, not the controller's.
4. For the mid hold, push the stick roughly halfway out and rest your thumb against the edge of the cap rather than pressing down on it. Steady beats accurate here.
5. For the full hold, push all the way into the housing and hold firm. This one is usually the quietest and it is a useful comparison point.
6. Press Start and do not move for the whole run. If you shift position partway through, the tool will tell you, and the result is not usable.
7. Run all three hold positions. A stick that is silent at rest and noisy in the middle is extremely common, and testing only at rest will completely miss it.

Before you measure the noise, it helps to know where your center actually sits. The stick drift test gives you the offset, and the full controller tester is the quickest way to confirm the rest of the pad is behaving.

What Jitter Actually Is

An analog stick reports its position by measuring a voltage. That voltage passes through a sensor, gets converted from an analog signal into a number, travels down a cable or over a radio link, and eventually arrives in your browser. Noise can enter at every one of those stages, and none of it requires the stick to physically move.

The result is a reading that wanders while your thumb sits still. Sometimes it wanders by a few thousandths, which nothing will ever notice. Sometimes it wanders by two or three percent, which is enough for a game to treat it as a genuine input and start moving your character.

What makes jitter awkward to diagnose is that it is invisible in normal use. Fast movements swamp it completely. It only shows up when you are trying to do something slow and precise, which is exactly when you are least able to tell whether the problem is the controller or you.

The middle is the hard part

Here is the thing almost every controller testing page gets wrong. Testing a stick at rest is not the same as testing a stick for noise.

On a traditional potentiometer stick, a metal wiper drags along a carbon track. The electrical contact between them is not perfect, and it is at its worst somewhere in the middle of the travel where the wiper is neither settled at one end nor pressed hard into a stop. That is where the dust collects, where the track wears fastest, and where a failing stick starts to hiss first.

Meanwhile, at rest, all sorts of things conspire to make the reading look clean. Many controllers and drivers apply a small internal deadzone and snap anything near center to exactly zero. Some firmware filters the signal aggressively near the middle. The stick can be genuinely noisy and still report a flawless string of zeroes, because the noise is being swallowed before it ever leaves the controller.

Which is why this test has three hold positions instead of one. If your rest reading is perfect and your mid hold reading is a mess, you have learned something real that a rest only test would have hidden from you completely.

Reading Your Numbers

Noise level

The typical distance between each reading and the middle of the cloud, calculated across the whole run. This is the headline number. It describes the general hiss rather than the worst single moment.

Widest excursion

The furthest any single reading strayed from the middle of the cloud. If this is close to the noise level, you have steady hiss. If it is many times larger, you have occasional jumps, and jumps have completely different causes from hiss.

Resting offset

How far the middle of your cloud sits from true zero. This is drift rather than jitter, and it appears here because the two are constantly confused. A stick can be perfectly quiet and badly offset, or perfectly centered and horribly noisy. If the offset is the larger problem, the stick drift test is the page that goes into it properly.

Smallest step

The smallest gap the tool could find between two different values your controller reported. This is not noise. It is resolution, and it tells you something interesting about the hardware. A step around 0.008 means the controller is sending you eight bit values, roughly 256 positions across the full travel. Around 0.002 means ten bit. Below 0.0005 means sixteen bit or that something is interpolating on the way through. Coarse resolution feels like stepping rather than hissing, and it shows up on the plot as a grid of separated dots rather than a smear.

Spikes caught

Readings that jumped far outside the normal spread. A handful over a long run is nothing. A steady rhythm of them usually points at the connection rather than the stick, and the connection stability test is where to look next.

Deadzone needed

On a resting test, the smallest deadzone that would swallow every reading you produced, with a small margin added. This is a genuinely useful number to carry into a game settings menu. If you want to understand the trade you are making when you set it, the joystick deadzone test covers that ground.

What the Cloud Shape Tells You

Hiss and Jumps Are Different Problems

Two things get called jitter and they have almost nothing in common.

Hiss is continuous. Every reading is slightly different from the last, in small amounts, forever. It comes from the sensor itself, from the analog to digital conversion, or from electrical noise picked up along the way. A worn carbon track hisses. A cheap unshielded cable hisses. A controller sharing a USB port with something power hungry hisses.

Jumps are occasional. The reading sits quietly and then leaps somewhere else for one or two samples before returning. That is not the sensor. That is a contact briefly failing, a packet arriving late, or a wireless link losing a moment. A stick that hisses needs cleaning or replacing. A stick that jumps usually needs a different cable, a different port, or a look at the connection.

The widest excursion number is how you tell them apart without staring at the plot. If it sits within about three times the noise level, you have hiss. If it is five or ten times larger, you have jumps.

What Causes It

Dust and debris on the track

The most common cause on any controller with some age. Skin oil, dust and general grit work their way into the housing and settle on the resistive track. The wiper drags through it and the contact becomes unreliable. This is also the most fixable cause, and it is why compressed air is the first thing to try on any noisy stick.

A worn wiper

The contact itself flattens and loses tension over years of use. Less pressure means a worse connection means more noise. There is no cleaning fix for this one.

Power quality

An underpowered USB port, a cheap unpowered hub, or a long low quality cable all raise the noise floor. This is genuinely common and genuinely overlooked. If a controller is noisy on a hub and clean in a rear motherboard port, the stick was never the problem.

Wireless interference

Crowded 2.4GHz environments, USB 3 devices sitting next to a wireless receiver, and microwaves all add noise or drop packets. This shows up as jumps rather than hiss. The connection stability test is the direct way to check it.

Low battery

Some controllers get measurably noisier as the pack drains, because the reference voltage the sensor is measured against becomes less stable. If your readings differ between the start and end of a session, run the battery health test before you blame the stick.

Sensor type

This one matters here, unlike with snapback. Hall effect and TMR sticks have no physical contact between the moving part and the sensor, so they do not develop contact noise as they age. They can still pick up electrical noise from a bad power supply, and their resolution is still limited by the conversion step, but the specific problem of a wiper dragging through dust simply does not apply to them. A hall effect stick that is hissing badly is telling you to look at the wiring rather than the stick.

What Counts as Normal

Compare against your own baseline where you can. These ranges are a starting point rather than a standard, because normalisation differs between controllers and browsers.

Two pieces of context. First, a rest reading of exactly zero across the whole run is a suspicious result rather than a perfect one, because it usually means filtering rather than silence. Second, whether any of this matters depends entirely on the deadzone your games run. A player using a fifteen percent deadzone will never notice noise at two percent. A player who has pushed their deadzone down to two percent chasing precision has deliberately removed the thing that was hiding it.

Jitter Sits Alongside Four Other Stick Faults

These get bundled together constantly and each one has a different fix.

Jitter is the reading moving while the stick does not. It is a signal quality problem, and this page measures it.

Drift is the reading sitting away from zero while the stick rests. It is a position problem, covered by the stick drift test.

Deadzone is the threshold you set to ignore both of the above. It is a setting, and the joystick deadzone test finds the smallest one that works.

Snapback is the stick overshooting past center after you let go. It is a spring and damping problem, measured on the snapback test.

Circularity is the shape the stick traces at full travel. It is a range problem, drawn by the circularity test.

Running the set takes about fifteen minutes and the overlaps are where the useful information lives. Noise on one axis plus a short gate on the same axis points at one worn part. Noise everywhere with clean drift, clean snapback and a clean gate points at the wiring rather than the stick. Starting from the main gamepad tester and working down the list gives you the whole picture instead of one slice of it.

Fixing a Noisy Stick

Try these first, in this order

1. Change the port. Move to a rear motherboard port, off any hub. Free, takes ten seconds, and it solves more noise problems than people expect.
2. Change the cable. Thin unshielded charging cables are a real source of noise. A shorter, better made cable often drops the noise floor noticeably.
3. Go wired. If the noise disappears on a cable, you were measuring the radio link and not the stick.
4. Charge it fully. Then measure again and compare.
5. Compressed air. Work the stick through its full range while blowing around the base. This clears loose debris and costs nothing.
6. Close anything sitting in between. Steam Input, DS4Windows, reWASD and vendor software can all filter or reshape the signal. The controller mapping tool shows you the raw axis values arriving, which makes interference obvious.

If those do not help

• Contact cleaner. A small amount applied to the potentiometer, worked through the full range, then left to dry. It can genuinely restore a noisy stick. It can also strip the grease that damps the gimbal and give you snapback in exchange, so test both before and after. The drift repair guide covers the procedure and the risks.
• Replace the module. The permanent answer for a worn track or wiper. The part is inexpensive and the soldering is the real cost.
• Raise your deadzone. A workaround rather than a fix. It hides the noise at the price of precision near center for everything else you do.
• Do not reach for smoothing. Filtering the signal averages the noise away and adds delay in exchange. Measure what that delay costs you with the latency test before you accept the trade.

Method and Honest Limits

Sample rate

This page reads the controller in step with your display refresh, so a 60Hz screen gives you about 60 readings a second and a 144Hz screen gives you well over twice that. For measuring the size of noise, that is plenty. For measuring the frequency of it, it is not, because anything oscillating faster than half your sample rate cannot be seen correctly. If you want to know how fast the noise is rather than how big, that needs different tooling. The sample rate readout tells you what you are actually getting.

Hold genuinely still

The tool compares the beginning of your run against the end and warns you if the cloud moved. Set the controller on a desk for the resting test. Human hands have a natural tremor of a few hertz and it is easily large enough to contaminate a measurement at this scale.

Run each position

Rest, mid and full tell you three different things. Mid hold is where problems appear first, so if you only have time for one, make it that one.

Test both sticks

They wear at different rates because they do different jobs, and on most controllers the right stick sees far more small precise movements.

Save it

Noise creeps up slowly. A number today next to a number from six months ago is evidence. A number on its own is just a number.

Browser and Platform Notes

• Chrome and Edge report axis values most consistently. Firefox works. Safari is the least predictable.
• Some browsers and drivers apply their own small deadzone before you ever see the value. That is exactly why a perfect rest reading deserves a mid hold test to confirm it.
• Keep this tab focused. Background tabs get throttled and the sampling stops.
• A few third party controllers only expose their axes properly in DirectInput mode, usually reached with a button combination involving Home.
• If you are unsure which axis numbers your pad uses, the controller mapping tool prints them, and the gamepad viewer shows the values live.
• Everything runs locally in the page. Readings never leave your device and the exports are files your browser writes for you.

Questions People Actually Ask

My rest reading is exactly zero. Is my stick perfect?

Maybe, but do not celebrate yet. Many controllers snap small values to zero before sending them, which produces a flawless rest reading on a stick that is actually noisy. Run the mid hold test. If that is clean too, then yes, you have a quiet stick.

Is a bit of jitter normal?

Completely. Every analog sensor has a noise floor. The question is whether it is large enough to cross the deadzone your games use, and the table in the results answers exactly that.

Why is my noise worse in the middle than at the edges?

Because that is where a potentiometer wiper has the least reliable contact and where dust settles. It is the expected pattern rather than a strange one.

My cloud is a grid of dots instead of a blob. What is wrong?

Nothing. You are seeing the resolution of the controller, not noise. The smallest step reading tells you how fine the steps are.

Can jitter cause drift?

Not directly, but they share causes. A dirty track produces both, which is why cleaning often improves both at once.

Will a hall effect controller fix this?

For contact noise, yes, because there is no contact to wear. For electrical noise from a bad port, cable or power supply, no. Diagnose which one you have before spending money.

Does jitter get worse when the controller warms up?

Sometimes. If your numbers differ between a cold run and one at the end of a session, that is worth noting in the report, because intermittent faults are the hardest ones to get taken seriously by support.

How long should I sample for?

Ten seconds is enough to characterise steady hiss. If you are trying to catch something that happens every few seconds, run thirty or sixty and watch the spike count instead of the noise level.

Related Tests on Gpadtester

Jitter covers signal quality while the stick sits still. These pages cover everything else, and readings that agree across two of them are worth far more than either alone.

Sticks and triggers

Whole controller

Connection and response

Read next

How to fix controller drift covers cleaning, contact cleaner and module replacement. The same work that clears a dirty track also clears the noise coming off it.

Other hardware

Away from the controller: keyboard test, mouse test, mic test, frame rate test and refresh rate test.