OLED Burn-In Test
Full-screen flat fields and a slow wipe to reveal image retention on OLED, QD-OLED and AMOLED — plus a timed pixel-refresher wash. Everything runs locally in your browser.
Sweeps the chosen field across the panel every 9 seconds. A moving edge makes faint ghosts pop — your eye is far better at spotting a difference beside a moving boundary than a static one. Toggle it from inside with W.
Nothing is measured, uploaded or recorded here — this page has no camera, microphone or screen-capture code at all. It drives your panel with clean signals; the detection is done by your eyes.
♻ Pixel refresher wash
Runs a timed full-screen wash that ramps the panel through white, black and the six primary and secondary colours in complementary pairs, with a slowly drifting noise field on top. Each loop takes 40 seconds and gives the red, green and blue subpixels an identical drive profile — every one is swept from dark to bright and back the same number of times — which is what evens out temporary retention. The drifting noise deliberately holds the extremes a little short of absolute black and absolute white, so no pixel is parked at a fixed level for even one stop.
Checking screen wake lock support…
Turn the brightness down first: an hour of full-brightness wash puts real runtime hours and heat into the panel. The timer pauses if you switch tabs or minimise, because a wash that isn't on screen isn't washing anything. If you are photosensitive, don't watch it — the colours ramp slowly and never flash, but it is an hour of changing full-screen light.
How to run the check
- Dim the room and turn off adaptive brightness. Ambient light hides low-contrast ghosting, and a panel that dims itself mid-test changes the very thing you are looking at. On Windows also turn HDR off for this — in HDR the desktop is tone-mapped and a "5% grey" is no longer 5%.
- Start with 5% grey. Look across the whole panel from about arm's length, then move your head. Retention announces itself as a shape you recognise: a taskbar strip along one edge, a game HUD, a row of browser tabs, a channel logo in a corner.
- Step up through 10%, 20% and 50%. Use ← / → or tap. A mark that is obvious at 5% and gone by 50% is mild; a mark you can still see on the 50% field is a large differential and worth acting on.
- Turn on the wipe. The field sweeps across the panel and ghosts pop out as the edge passes them. This is the single most sensitive step — faint marks that you talked yourself out of on a static field become undeniable.
- Check white, red, green and blue. The colour of the ghost tells you which subpixel wore: a mark that is clear on the blue field but invisible on red means the blue emitters in that area have aged, which is the usual OLED failure. The black field at the end is there to catch bright or stuck subpixels while you are already looking — for a proper grid sweep use the dead pixel test.
- Then re-test. Run the pixel refresher for 10–20 minutes and go back to the same grey field. A ghost that has faded or gone was image retention. A ghost that is pixel-for-pixel identical is burn-in.
Before you trust what you see
- Near-black is not perfectly uniform on any panel. Faint blotches, a vertical band or a warm/cool tint on a 5% field are normal OLED near-black behaviour, not burn-in. Burn-in has an edge and a shape you can name.
- Viewing angle lies. OLED shifts colour off-axis. Judge from dead centre, at a normal distance, then confirm from the other side of the screen — real wear stays put; angle tint moves with you.
- Clean the screen first. A dry microfibre pass removes the smudge you are about to diagnose as a permanent defect.
- Percentages are code values, not luminance. "50% grey" means sRGB code value 128. On a standard gamma-2.2 display that is roughly 22% of peak luminance. On a wide-gamut panel the OS colour pipeline may also nudge these values — harmless for spotting retention, but "full red" here may not be your panel's most saturated red.
- LCDs do this too. IPS and VA panels get genuine image retention from voltage sticking in the liquid crystal; it looks identical and almost always clears within an hour. Permanent burn-in on an LCD is rare.
- There is no sensor in this page. A browser cannot measure panel wear. This is a controlled stimulus plus your eyes — the only number here is the wash timer.
What is actually happening in the panel
Image retention is a charge-trapping effect. Drive one region hard for hours and the transistor and organic layers under it settle into a slightly different operating point; show a flat field afterwards and that region reads back a little brighter or darker. It relaxes on its own — minutes for a browser tab bar, a few hours for a HUD you left up all weekend.
Burn-in is not an effect at all, it is wear. Every OLED emitter loses efficiency as it runs: for the same drive current it emits less light. Emitters that ran at full white all day age faster than the ones next to them that spent the day at black. Once that gap opens, showing a uniform grey no longer produces a uniform grey — you see the negative of whatever was parked there. Nothing restores it, because there is nothing to restore: the material is spent.
Why blue subpixels go first
Blue light is the highest-energy visible light, and in shipping OLED stacks the blue emitter is still the weak link — red and green have used efficient phosphorescent emitters for years while blue has largely stayed fluorescent or TADF. Lower efficiency means the blue subpixel must be driven harder to hit the same perceived brightness, and higher-energy excitons chew through the host material faster. Both effects push the same way, so blue ages first and a worn area drifts warm and dim.
The panel type changes how that shows up. WOLED (most LG-panel TVs) puts a white stack behind colour filters plus an unfiltered white subpixel, so wear tends to read as an overall dimming with a colour cast. QD-OLED (Samsung Display) generates all of its light from a blue OLED layer and converts part of it to red and green with quantum dots — so blue-layer ageing touches every colour on screen. AMOLED phones use per-colour emitters and are the classic case: a status bar and a navigation bar that never move for two years.
What causes it, and what actually helps
The culprits are always the same: the Windows taskbar or macOS dock, a game HUD with a minimap and ammo counter, a broadcaster's channel logo, a news ticker, letterbox bars on 21:9 content, a desktop left on the same wallpaper with the same icons, and a game left paused for three hours.
- Auto-hide the taskbar or dock, and use a dark theme and a dark wallpaper. This is the single highest-value change for a desktop OLED.
- Turn on logo dimming and pixel shift. TVs call these Logo Luminance Adjustment and Screen Shift / Screen Move; monitors usually have a Pixel Orbit or Screen Move toggle. They cost you a couple of pixels of image and they genuinely work.
- Drop peak brightness for static work. Wear scales with drive current. Editing a spreadsheet at 250 nits ages the panel dramatically faster than at 120, and you will not miss the difference indoors.
- Set a short screen-off timer — two to five minutes. Use a moving screensaver or, better, a blank one. A static screensaver is worse than no screensaver.
- Let the firmware do its job. Don't kill the power at the wall the second you switch off; that is when the short compensation cycle runs.
- Vary the content. The failure mode is a fixed image for hundreds of hours. Anything that moves is fine — a panel that plays film and games all evening is not the one that develops burn-in.
FAQ
What is the difference between image retention and burn-in?
Image retention is temporary. A bright static element leaves a faint ghost that fades on its own once the panel shows varied content again, usually within minutes to a few hours. Burn-in is permanent: the organic emitters under that static element have aged more than their neighbours, so they output less light forever. Retention clears on its own; burn-in can only be compensated for, never reversed.
Why does this test start with 5% and 10% grey instead of white?
Uneven wear shows up as a difference in light output, and that difference is easiest to see when the panel is barely lit. On a full white field a worn area is a fraction of a percent dimmer and your eye will miss it. On a 5% or 10% grey field the same worn area can look clearly darker or tinted, which is why near-black fields are the standard first check.
Can this tool fix burn-in?
No, and neither can any other software. The pixel refresher wash evens out temporary image retention by driving every pixel through the full range of colours, which can clear a ghost that has not yet become permanent. Real burn-in is physical wear in the emitter layer, and no pattern, video or app can restore light output that is already gone.
How long should I run the pixel refresher?
Start with 10 to 20 minutes at a moderate brightness. If a ghost is still there afterwards, leave the panel showing varied content for a few hours and check again rather than immediately running another long wash. The wash itself adds runtime hours and heat to the panel, so longer is not automatically better.
Does my OLED already run its own pixel refresh?
Almost certainly. Most OLED TVs and monitors run a short compensation cycle automatically when you switch them off after a few hours of use, plus a much longer panel-refresh cycle after a couple of thousand hours. That firmware routine measures and compensates individual pixels and is far more capable than anything a web page can do. This wash does not replace it and does not reset its counter.
Will burn-in show up in a screenshot or a photo?
Not in a screenshot. A screenshot captures the image the computer sent to the display, and burn-in happens inside the panel itself. A phone photo of the screen can capture it, but only if you turn off HDR capture and shoot a flat grey field straight on with the exposure locked; otherwise the camera's processing hides the very low-contrast difference you are looking for.