GameHomeHub Free browser tools

GameHomeHub/Screen Test — Dead Pixels, Refresh Rate & Input Lag

Display check · dead pixels · refresh rate · input lag

Screen Test — Dead Pixels, Refresh Rate & Input Lag

Three checks in one page: hunt dead and stuck pixels with fullscreen flat colours, measure the refresh rate the browser really composites, and estimate click-to-paint input lag. All local, no install, nothing uploaded.

Three display checks

Runs in this tab · nothing uploaded

Fullscreen flat colours. Look for dots that stay the wrong colour — a pixel that does not change with the fill is stuck, and a pixel that stays black on the white fill is dead. Click or use the arrow keys to change colour; Esc leaves fullscreen.

Opens fullscreen; in-page overlay if the browser blocks it

Measures how many frames per second the browser actually composites — a two-second sample of requestAnimationFrame timestamps. Browsers throttle this to your display, so it is the real rate rather than the panel's spec sheet, but it also includes any frame pacing problems the browser has right now.

A steady 16.7 ms average is 60 Hz; 8.3 ms is 120 Hz; 6.9 ms is 144 Hz. Large gaps between the fastest and slowest frame mean dropped frames — a browser or cable problem before it is a monitor problem.

Click the target ten times in a row. Each click measures the time from the pointer event's timestamp to the next animation frame — an estimate that includes one frame of compositor latency, so treat the number as an upper bound rather than a monitor review.

no samples yet

Your display adds its own latency on top of this, and a browser cannot see it. What this number is good for is comparison: the same machine with a different monitor, cable or refresh setting will show a different figure.

Everything runs locally in the browser — nothing is uploaded, no install is needed.

How to use it

  1. Dead pixels: start the fullscreen test, then step through the colours with a click or the arrow keys. Look at each colour for a few seconds — stuck pixels are easiest to see against a fill that should be uniform.
  2. Refresh rate: press measure and leave the tab in front for two seconds. The result is what the browser composites, so check it after changing the display setting to confirm the change took effect.
  3. Input lag: click the target ten times. The average is the estimate; the per-click values show jitter, which matters more than the average for how a mouse feels.

Dead pixel test

Two or three minutes in fullscreen catches most bad pixels on a laptop or desktop panel. The fullscreen button is in the first tab; it asks the browser for fullscreen, and if the browser refuses it still runs as an in-page overlay you can maximise.

  1. Start on black. Black is where bright defects show: a pixel stuck on white, red, green or blue stands out as a coloured dot on a black fill. Desktop wallpapers and colourful pages hide these — a flat black fill does not.
  2. Give each colour five to ten seconds. Your eye needs a moment to stop compensating, and you need to sweep the whole panel — corners and edges first, because those are the defects people miss. Run the test at the display's native resolution: if the image is scaled, one image pixel is not one panel pixel and small dots get smeared.
  3. Switch to white. White is where dark defects show — a genuinely dead pixel stays black on a white fill. View from about 30 cm, and do not confuse dust on the glass with a pixel: dust moves when you wipe the screen, a bad pixel does not.
  4. Cycle red, green and blue. A single failed subpixel only shows on the fill of its own colour, so all three primary fills are needed to place a defect that a white or grey fill hides.
  5. Leave with Esc. A click or the arrow keys move to the next colour; Esc or the exit button closes the test. Turn off night light, colour filters and any HDR mode first — they change what the flat fills look like.

Write down what you see. Note the fill colour the defect appears on and roughly where on the panel it is. A defect that appears on every fill is dead; one that appears on a single fill is a failed subpixel; one that stays lit in a colour across all fills is stuck. Manufacturers' dead-pixel policies count these differently, so the colours matter when you make a claim.

Is "massaging" a stuck pixel reliable? Not really. The usual tricks — pressing gently on the area with a soft cloth, or running a rapid colour-cycle film over it for 10–30 minutes — sometimes recover a stuck pixel and often do nothing; pressing harder can leave marks on the panel. A dead pixel (black on white) does not come back. If the display is under warranty, open a claim instead of experimenting.

What each test measures

Dead pixels is a visual check, not an automated one — the steps above are the whole method. This tool cannot scan the panel for you; no browser can read individual pixels off the display.

Refresh rate comes from two seconds of requestAnimationFrame timestamps. Browsers throttle animation to the display's rate, so the number reflects the real end-to-end setting — including a cable or adapter that cannot carry the bandwidth. Expect 16.7 ms per frame at 60 Hz, 8.3 ms at 120 Hz, 6.9 ms at 144 Hz.

Input lag measures pointer event timestamp to the next animation frame. It is an upper-bound estimate: it includes one frame of compositor latency and cannot see the panel's response time. Use it to compare setups, not to grade a monitor.

Common problems

  • The fullscreen test will not start. Browsers require a user gesture for fullscreen; the button click is that gesture. If the browser refuses, the overlay still runs in-page — move the window somewhere dark or maximise it.
  • Refresh rate reads 60 Hz on a faster monitor. Check the OS display setting, the cable (HDMI before 2.0 caps at 60 Hz for 4K), and any adapter in the chain before blaming the browser.
  • Input lag numbers jump around. Other applications, power saving, and background tabs all add latency. Close what you do not need and keep this tab focused while measuring.
  • A dead pixel appeared after the warranty. Manufacturers publish a dead-pixel policy per panel class and region; check the model's policy before assuming it is covered.

Frequently asked

Is this a dead pixel test?

Yes. The first tab is a fullscreen dead pixel test: it fills the display with flat black, white, red, green, blue and grey so a dead or stuck pixel shows up as a dot that does not change with the fill. Click or use the arrow keys to change colour, and Esc leaves the test.

How do I find a dead pixel?

Run the fullscreen colour test and look for dots that do not change with the fill. A pixel stuck on one colour will stay wrong on every screen, and a dead pixel stays black on the white fill. Keep the display at its native resolution and turn off any colour filters first.

Why does the refresh rate test show a different number from my monitor's spec?

The test measures what the browser actually composites — the frames it hands to the display — not the panel's specification. A 144 Hz monitor can report 60 Hz if Windows or the browser is set to 60 Hz, if the cable cannot carry the bandwidth, or if the machine cannot keep up.

Is the input lag number accurate?

It is an estimate from your pointer event's timestamp to the next animation frame, so it includes one frame of compositor latency and cannot see the display's own response time. Treat it as a comparison tool: same machine, different monitor or cable, different number — not as a monitor review.

Can a stuck pixel be fixed?

Sometimes. A pixel stuck on one colour can recover after running a rapid colour-cycle over that area for several minutes, or with gentle pressure (carefully — pressure can damage the panel). A truly dead pixel, one that stays black on white, does not recover; it counts against the manufacturer's dead-pixel policy, which varies by brand and region.

More on GameHomeHub