Dead vs Stuck vs Hot Pixels — and What Warranties Cover
Every pixel on an LCD is three tiny shutters — red, green, blue — and each shutter can fail in one of two directions: jammed shut or jammed open. That’s the entire taxonomy, and it decides whether your warranty claim succeeds.
The three defects
Dead pixel. The whole pixel gets no drive signal, all three shutters stay shut, and you see a black dot on any bright background. Standards documents call this a “type 2” fault (always black). It’s permanent: the transistor or trace feeding that pixel has failed.
Hot pixel. The opposite: everything stuck open, a white dot on black. “Type 1” (always bright). Rarer, more noticeable, and — usefully for you — treated more seriously by most warranty policies, because a bright dot on a dark movie scene is impossible to unsee.
Stuck subpixel. One of the three shutters is jammed open, so you get a pure red, green, or blue dot that’s most obvious on black or on a color that should exclude that channel. This is “type 3,” the most common defect — and the only one with any realistic chance of self-repair or coaxed recovery.
A two-minute pass through the fullscreen color test identifies which one you have: dark on white means dead, bright on black means hot, colored on black means stuck.
The standard manufacturers point to
When a support rep tells you your defect is “within specification,” they’re usually leaning on ISO 9241-307, which defines pixel fault classes. The class system sets how many faults per million pixels are acceptable:
| Class | Type 1 (hot) | Type 2 (dead) | Type 3 (stuck subpixel) |
|---|---|---|---|
| Class 0 | 0 | 0 | 0 |
| Class I | 1 | 1 | 2 |
| Class II | 2 | 2 | 5 |
| Class III | 5 | 15 | 50 |
Those numbers are per million pixels, so scale them up: a 4K panel has 8.3 million pixels, which under Class II — where most consumer monitors sit — means up to 16 hot, 16 dead, and 41 stuck subpixels can be called acceptable. Almost no panel ships anywhere near that bad, but the class gives manufacturers a wide legal shelter, which is why the marketing term to look for is “Zero Bright Dot” or similar guarantees that voluntarily promise better than the class requires.
How actual policies differ
The class system is the floor; each maker’s pixel policy sits somewhere above it. Reading a stack of current policies, the patterns that matter to you are:
- Bright beats dark. Many policies replace a panel for a single always-bright pixel but tolerate several dark ones. If your defect is hot, your odds are good.
- Location matters. Several policies divide the screen into zones; a defect in the central zone qualifies with fewer faults than one at the edge.
- Premium lines promise more. “Zero bright dot” guarantees are common on flagship monitors and rare on budget ones — sometimes the guarantee is the actual difference between two otherwise similar models.
- The clock matters. Some pixel guarantees run shorter than the general warranty. A defect you noticed in month two but reported in month fourteen may have aged out.
Policies change and vary by region, so treat any specific numbers you read in forums (or here) as a prompt to check the current policy for your exact model — search “manufacturer pixel policy” and read the PDF, not the summary.
Making the claim stick
- Document on the revealing color. Photograph the defect on whichever solid color shows it best, plus one wider shot that locates it on the screen. Include something for scale.
- Count precisely. Run the full color cycle and note every defect and its type. “Three stuck subpixels, one dead pixel, locations attached” reads very differently from “some dots.”
- Cite their own policy. “This is a type 1 bright defect, which your policy covers at quantity one” shortens the conversation considerably.
- Act fast on new purchases. Within a retailer’s return window you don’t need the pixel policy at all — you need a receipt. The policy only becomes your problem after that window closes.
The honest coda
One stuck subpixel at the edge of a high-density screen is, at normal viewing distance, invisible. If the policy says no and the cycling trick fails, the healthiest move is to stop running tests that show you a defect you otherwise never see. We say this as the people who built the test.