OLED Burn-In Census: What Owners See, by Panel and Hours

Whether an OLED monitor will burn in is argued in every thread about buying one. The evidence on the table is mostly long tests of single monitors. Monitors Unboxed has used one MSI MPG 321URX for desktop work for 21 months and reported on it at intervals, and Notebookcheck’s account of the latest update puts it at about 5,000 hours, with burn-in that keeps worsening in the areas where it first appeared. That is one unit, used one way.

What nobody holds is the spread. How many owners see nothing after two years? How many see a faint taskbar line on a gray slide, and how many sent the monitor back? A search in September 2026 turned up no dataset of owner reports that records the panel, the hours and what the owner sees on a fixed scale.

This page opens that census. One row is one monitor, judged once by the person who owns it. The data, the field definitions, the validator and the report form live in the open repository: github.com/iBlessi/techfuelhq-census (CC BY 4.0).

My own OLED is the Gigabyte AORUS FO48U in this review. If you are choosing between an OLED television and a monitor for a desk, that comparison is in OLED TV as a PC monitor.

How results are reported

For each panel type the census gives the number of monitors, how many sit at each step of the scale, the median months in use, and how many warranty claims were approved or denied. Hours are handled with more care. Only hours read from the monitor’s own menu feed a median, and only once five rows in the group carry them.

It publishes no verdict on a model or a brand. Two owners of the same monitor can use it in ways that have nothing in common, one on spreadsheets for ten hours a day and the other on games for two, so the row records the use beside the result.

The scale

The steps are fixed. A change to their wording would make old and new rows incomparable, so it would be a new major version of the dataset with the old rows marked.

What one row records

A row is one monitor, judged once. Of its 18 fields, you fill in 16 and the maintainer sets 2.

How to judge a panel

1. Let the monitor run for half an hour, then dim the room.
2. Open the test slides in the builder below. Start with dark gray, where faint marks show first.
3. Look where static things live on your screen: the taskbar, the edges of windows you keep in one place, the HUD of the game you play most.
4. Pick the step of the scale that matches what you see. If you see a mark only on a slide and never in normal content, that is a 1.
5. If the monitor’s menu shows panel hours, read them there. Leave the field empty if you would be guessing.

Send a report

Sending needs a GitHub account, because the report is filed as an issue. No account? Copy the report and email it to contact@techfuelhq.com.

What has been published

Current count: 0 reports. No group has rows yet, so nothing is published.

A panel type publishes once it holds five monitors that have been in regular use for a month or more. Below that it is listed as collecting, with its count and no figures.

Limits of this data

• Reports come from people who chose to send them. An owner who found a mark is more likely to report than one who never looked.
• The middle of the scale is a judgement. Steps 0, 1 and 4 are close to facts; the line between 2 and 3 is where two careful owners can differ.
• Hours from a monitor’s menu are kept apart from estimates, and only menu hours feed a published median.
• A row is one look at one monitor. It says nothing about what the same panel shows a year later unless its owner reports again.

Use the data

The dataset is one file. submissions.csv holds one row per monitor, in the field order above, under Creative Commons Attribution 4.0. Attribute it as “TechFuelHQ open census” with a link to this page.

The coil whine census and the GPU sag census are the sibling datasets.

Change log

• 0.1.0 (2026-09-29): fields, the five-step scale, test slides, validator and intake form. The census opened with 0 rows.