Last month a buyer sent me a spec sheet and asked a fair question: "Three suppliers all say 7000 nits. Why does the cheapest one look like a grey rectangle at 2 pm?"
The number was the same in all three documents. The screens were not.
Brightness is the most quoted and least comparable figure on an outdoor LED datasheet. It is measured under conditions that most buyers never see, and it is quoted at a point in the product's life that most projects never reach. Here is how I read it now, after too many site visits where the screen was fine on paper and disappointing in daylight.
First, the unit is not the problem
1 nit = 1 cd/m². There is no ambiguity in the unit itself, and no unit conversion trick hiding here. When two suppliers say 7000 nits they mean the same physical quantity.
The ambiguity is entirely in when and where that quantity was taken.
The test condition is the spec
A brightness figure is meaningless without four pieces of context. Ask for all four in writing:
- Measurement geometry. Module on a bench, or a full assembled cabinet, or a complete screen after installation? A single module under a lab meter will read higher than a finished wall, because the finished wall includes masking, louvres, glass, and the seams between cabinets.
- Colour temperature and pattern. Standard practice is full-white at around 6500K. A screen pushed to a warmer or bluer white can post a different number. If the datasheet does not state the white point, the figure is decorative.
- Before or after greyscale calibration. Factory brightness is usually quoted on an uncalibrated screen. Calibration to keep low-grey steps clean almost always costs some peak output. The number that matters on site is the calibrated one.
- Peak or sustained. A screen can hit a number for thirty seconds on a bench and never hold it on a roof in August. Sustained brightness is the only one your audience sees.
Heat is where the number goes
An LED screen is a heater that happens to emit light. Roughly speaking, most of the electrical input leaves as heat, and only a fraction leaves as photons. Raise the drive current to chase a higher peak and you raise junction temperature, and junction temperature is what governs both colour shift and lumen depreciation.
This is why two cabinets with identical LEDs can behave very differently at noon:
- Cabinet thermal design — heat sink mass, airflow path, whether the driver is inside the same sealed box as the modules.
- Driver headroom — a power supply running near its ceiling derates itself, and the screen dims quietly instead of failing loudly.
- Ventilation strategy — front-service cabinets with sealed drivers and a separate air path hold brightness better than a fully sealed box with everything in one chamber.
When a supplier quotes 7000 nits, the useful follow-up is not "can you do more". It is "at what ambient temperature does that 7000 become 5500, and what is the derating curve".
What the site actually needs
Ambient light is the other half of the equation, and it is measurable before you buy:
- A façade in direct sun, no shade, viewer across a road: extremely bright environment. This is where high-output cabinets earn their cost.
- A wall in open shade, or a screen facing away from the sun's path: far less demanding, and a 5000-nit cabinet can look better than an over-driven 7000 because it runs cooler and holds its greys.
- A canopy, a shopfront recess, an indoor atrium behind glass: 2500–5000 nits is normally plenty, and running higher just warms the room.
Two screens at the same nominal brightness can also look very different because of contrast, not output. A cabinet with good masking and a dark face absorbs ambient light instead of reflecting it. If you have ever seen a 6500-nit screen beat a 7000-nit screen at the same site, this is usually why.
Where brightness goes over three years
Nobody buys a screen for its first afternoon. Ask what happens to the number:
- LED binning. Tight bins keep modules consistent so the screen does not develop bright patches as different bins age at different rates.
- Driver IC choice. Constant-current drivers with good low-grey performance keep the bottom of the greyscale usable after the screen has run a few thousand hours.
- Sealing and cleaning. Dust on the face and moisture in the cabinet both eat apparent brightness, and neither is covered by the datasheet figure.
- Dimming range. A screen that can only run between 100% and 60% will be run at 100% at night, which is exactly when neighbours complain and when the LEDs age fastest.
The practical spec here is not "7000 nits peak". It is "7000 nits calibrated, stable at 45 °C ambient, dimmable to single-digit percentage with 14-bit greyscale intact".
Six questions I now ask before signing
- Is the quoted brightness measured on a module, a cabinet, or an installed screen?
- At what white point, and before or after greyscale calibration?
- What is the sustained brightness at 45 °C ambient, and what does the derating curve look like?
- What is the power consumption at 6000 nits in normal content, not full white?
- What is the brightness uniformity across the wall, and how is it measured?
- What is the minimum dimming level, and how many greyscale bits survive at that level?
A supplier who can answer these quickly is telling you something more useful than a bigger number on page two.
For reference, the cabinets we build for full-sun façades are here — AS10-2S outdoor advertising screen and the AS10 outdoor panel. What I would actually want you to take from this post is the question list above, whichever supplier you end up buying from.
MuenLED
I work with MuenLED, a Shenzhen LED display manufacturer. The engineering opinions above are mine, and the question list is the same one I use on our own product pages.
Top comments (0)