Dead Pixel Test Guide: Check a Screen Colour by Colour

A display is a grid of tiny lights, and each point on that grid is a pixel made of red, green, and blue subpixels. When all three subpixels work, the pixel can show any colour. When they fail, the pixel misbehaves in a way that is often tiny, a single dot among millions, but surprisingly noticeable once you have seen it, because the eye is drawn to a point that never changes while everything around it moves. A dead pixel is one that stays dark on every colour, because its subpixels produce no light at all. A stuck pixel is one that stays fixed on a single colour, such as red or green, because one or more subpixels are locked on. A hot pixel is the opposite of dead: it stays bright white on every colour. The distinctions matter because they point at different causes and different remedies, and a test that only uses one colour cannot tell them apart.

The Dead Pixel Test tool shows a fullscreen solid colour in your browser and lets you look for dead or stuck pixels. The fullscreen part is essential: in a normal window, browser chrome, page content, and window borders all compete for attention, and a single faulty pixel hides among them. In fullscreen, the entire display becomes one colour, so any point that differs from that colour stands out by contrast. The test is a visual inspection aid, not a hardware diagnostic: it shows what the screen displays, and it cannot repair a pixel, measure the panel's electronics, or distinguish with certainty between a panel fault and something on the surface of the screen. What it does well is make a small fault visible enough to confirm, count, and describe accurately, which is exactly what a warranty claim or a repair conversation needs.

Colour choice is the method, not a decoration. A dead pixel, which produces no light, is invisible on a black screen, because black is also the absence of light, and it is most visible on white, where a dark dot breaks an otherwise uniform bright field. A stuck red pixel is invisible on a red screen, because it matches, and most visible on green, blue, or white, where its fixed red stands out. A stuck green pixel hides on green and shows on red, blue, or white. That is why a proper test cycles through at least black, white, red, green, and blue: 5 colours, each of which hides a different fault and reveals the others. A fault that appears on 4 of the 5 colours and vanishes on the fifth is behaving exactly like a stuck pixel of the matching colour, and that pattern is strong evidence about what kind of fault it is.

Before concluding a pixel has failed, rule out the impostors. Dust on the screen can look like a dark dot, but dust sits on the surface and moves or disappears when the screen is cleaned, while a dead pixel does not. A smudge can look like a discoloured patch, but it changes with viewing angle and cleaning, while a pixel fault does not. A software mark, such as a stuck cursor or an overlay, disappears when the test goes fullscreen or when the application changes, while a pixel fault persists across applications and across the fullscreen test. The test procedure should therefore include cleaning the screen gently with a suitable cloth, re-running the test, and checking whether the suspect point survives both steps. A point that survives cleaning and appears in the fullscreen test on multiple colours is a strong candidate for a genuine fault.

A worked example you can follow

Take a sample laptop screen with a resolution of 1920 by 1080 pixels. Multiply to get the total pixel count: 1920 times 1080 equals 2,073,600 pixels, or just over 2 million. Suppose the test reveals 1 suspect point. That is 1 out of 2,073,600, or about 0.00005 percent of the display. The smallness of that fraction is worth stating plainly, because it explains both why the fault is easy to miss and why it is still worth confirming: a single faulty pixel is a tiny share of the screen, but it sits in the same place on every image, so the eye keeps returning to it.

Now run the colour sequence on that suspect point and record the results. On white, the point appears dark. On black, it is invisible. On red, it appears dark. On green, it appears dark. On blue, it appears dark. Count the colours where it shows: 4 out of 5, with the exception being black, where a dark point cannot be distinguished from the background. That pattern, dark on every colour where darkness is visible, matches a dead pixel, which produces no light on any colour. Contrast that with a second suspect point that appears red on white, red on green, red on blue, and invisible on red, while on black it appears red. That point shows on 4 of the 5 colours as well, but the colour it shows is red on every one, which matches a stuck red pixel, not a dead one. The two points have the same count, 4 out of 5, but different colours, and the colour is what distinguishes them.

Take a third sample, a phone screen of 1170 by 2532 pixels. Multiply: 1170 times 2532 equals 2,962,440 pixels, or just under 3 million. Suppose the test shows 3 suspect points, but after cleaning the screen gently, 2 of them disappear, leaving 1. That is 3 minus 2, or 1 genuine candidate, and 2 out of 3, or about 67 percent, of the original suspects explained by surface dust. The cleaning step therefore changed the conclusion from 3 faults to 1, which is why it belongs in the procedure rather than after it. Record the surviving point's position, such as near the top-left corner, and its behaviour on each colour, because that record is what a support team needs to assess the claim without asking you to repeat the test from scratch.

Mistakes to avoid

The first mistake is testing with only one colour, which hides exactly the faults that match it. Use at least the 5 colours named above. The second mistake is skipping the cleaning step and reporting dust as dead pixels, which wastes a support conversation and undermines the genuine faults in the same report. The third mistake is pressing hard on the screen to fix a stuck pixel, a folk remedy that risks damaging the panel further and is not a repair. The fourth mistake is treating a browser test as a hardware verdict. The test shows display behaviour; a manufacturer's diagnostic or a technician's inspection is what confirms a panel fault for warranty purposes, and the test's job is to give that conversation accurate evidence, not to replace it.

The Monitor Test tool offers a broader screen check to run alongside the pixel test, and the Desktop Computer Test tool reads browser-visible values for the device itself. For background on what a 27-inch screen really measures, see Screen Size Calculator: What a 27-Inch Screen Really Measures.

Checking the result before you act on it

Treat a test result as evidence to be confirmed, not as a final diagnosis. Re-run the colour sequence after cleaning, note each suspect point's position and its behaviour on each colour, and check whether the points persist across applications and after a restart. If a point survives all of that, you have a well-documented fault to describe to support: its position, its colour behaviour, and the screen's resolution and pixel count. If it does not survive, you have saved yourself a repair trip. Either way, the test has done its job, which is to turn a vague feeling that something is wrong with the screen into a specific, checkable observation.