“1000 nits” is the number most often quoted for outdoor displays, and it is the least useful number on its own. A bright panel behind a glossy front surface can be harder to read in sunlight than a dimmer panel with an effective anti-glare treatment. What determines readability is the contrast the viewer actually sees, and that depends on reflected light as much as emitted light.

This article sets out a measurement approach that produces a defensible readability statement rather than a brightness figure.

Why brightness alone does not prove readability

Readability depends on the difference between the bright state and the dark state as the eye receives them. Ambient light adds reflected luminance to both, but it hurts the dark state far more because the dark state has little emitted light to compete with.

This is why two panels with identical luminance can behave completely differently outdoors. The one with lower reflectance keeps more of its contrast; the one with higher reflectance washes out even though its output is the same.

Reflected luminance and contrast under light

Under illumination, the perceived contrast is the sum of emitted and reflected luminance in the bright state divided by the reflected luminance in the dark state. Two consequences follow directly.

First, raising brightness helps, but with diminishing returns: each additional nit competes against a reflected floor that does not change. Second, reducing reflectance helps proportionally in the dark state, which is why surface treatment and bonding often deliver more perceived improvement than another few hundred nits.

Light sources and illumination levels

Time of day matters as much as the light source. A display tested at noon in summer is being measured in conditions it will only experience for part of the year; the same unit tested in winter at the same clock time sees a different sun angle and a different reflection. Record the date and time alongside the illumination, and repeat the measurement in the season that matters most for the product’s market.

Time of day matters as much as the light source. A display tested at noon in summer is being measured in conditions it will only experience for part of the year; the same unit tested in winter at the same clock time sees a different sun angle and a different reflection. Record the date and time alongside the illumination, and repeat the measurement in the season that matters most for the product’s market.

For products that live outdoors permanently, the design-side treatment of the same problem is covered in the guide to brightness and contrast for outdoor displays. Outdoor conditions are not a single number. An overcast day delivers diffuse light from the whole sky, which is comparatively easy to tolerate. Direct sun delivers a high-intensity source in a narrow angular range, which produces a specular reflection when the geometry lines up. Shade with a bright background creates a high-contrast field where the eye adapts to the surroundings rather than the screen.

A test that simulates only one of these will not describe the others. Record which condition the measurement represents: diffuse illuminance at a defined level, or a directional source at a defined angle and intensity.

Measurement geometry and simulated sunlight

Geometry determines what you measure. A directional light source aligned with the instrument’s line of sight produces the worst-case reflection and the lowest perceived contrast. A source at 45¡ã measures the diffusion behaviour of the surface instead.

Both are legitimate tests, and they answer different questions: “how bad is it at the worst moment of the day” versus “how does this surface behave under general daylight”. State which one you ran, and at what angle, because the numbers are not interchangeable.

How bonding and surface treatment change results

Measure the haze trade-off explicitly if the product has a wide range of ambient conditions. An anti-glare surface that reads comfortably at midday may look slightly veiled in a dark room, and the acceptance decision depends on which condition matters more. Measuring both, with the real content and the real viewing distance, gives a defensible answer; choosing on the basis of one condition and discovering the other after installation does not.

Optical bonding removes the internal air gap, which eliminates a set of internal reflections that otherwise lift the dark state. That change alone improves outdoor contrast measurably, before any change to the panel or the front surface.

Surface treatment then addresses the external reflection. An anti-glare surface scatters the reflection over a wider angle, reducing its intensity at any single point at the cost of a slight haze. An anti-reflective coating reduces reflectance directly but is more sensitive to handling. Where both are applied, they should be measured separately so the contribution of each is known.

Legibility tests versus contrast ratios

Observer-based legibility tests need a defined protocol, otherwise they produce opinions. Use a fixed viewing order, a fixed adaptation period before each observation, and a fixed set of content – the actual interface, not a test pattern. Ask the observer to read specific values rather than to rate the screen, because a reading task produces a measurable outcome and a rating produces a preference.

Observer-based legibility tests need a defined protocol, otherwise they produce opinions. Use a fixed viewing order, a fixed adaptation period before each observation, and a fixed set of content – the actual interface, not a test pattern. Ask the observer to read specific values rather than to rate the screen, because a reading task produces a measurable outcome and a rating produces a preference.

A contrast ratio is a measurement; legibility is the outcome the user cares about. Both belong in a readability evaluation, because a numerically adequate ratio can still produce unreadable text if the font is too thin or the content is dense.

A practical legibility test uses the content the product will actually display: the real font, the real size, the real colours, viewed at the intended distance under the defined illumination. The pass criterion is whether an observer can read the critical values, not whether the number exceeds a threshold.

Angle dependence of readability

Reflections are angle-dependent, and so is contrast. A display that reads well for an operator standing directly in front may be unusable for someone at 45¡ã, because the reflection moves into their line of sight.

If the product has more than one viewing position – a kiosk, a machine with a seated and a standing operator, a vehicle with a passenger screen – measure at each. This is also where anti-glare surfaces earn their place: they trade a small amount of clarity for a much wider angle of usable readability.

Documenting a readability claim

Keep the completed protocol with the product documentation rather than in a laboratory folder. When the panel, the bonding or the surface treatment is changed later – for example during an end-of-life replacement – the original protocol is what allows the change to be assessed against the same conditions instead of being re-argued from the beginning.

A defensible claim includes five elements: the panel and surface treatment, the illumination level and geometry, the measurement instrument and distance, the content used for the legibility check, and the viewing angles tested.

Element	Example of what to record
Configuration	Panel, bonding, surface treatment, brightness setting
Illumination	Level, source type, angle relative to the display normal
Instrument	Model, calibration status, distance and aperture
Content	The actual interface, font and size used in the product
Angles	Views tested, including the worst-case position
Result	Measured values plus the pass or fail decision

Field verification after installation

Record the field result in the same format as the laboratory result, so the two can be compared rather than merely admired. When the numbers disagree, the difference is usually one of four things: a different illumination level, a different viewing angle, a different brightness setting, or content that differs from the test content. Naming which one accounts for the gap is what turns a site visit into useful data.

Record the field result in the same format as the laboratory result, so the two can be compared rather than merely admired. When the numbers disagree, the difference is usually one of four things: a different illumination level, a different viewing angle, a different brightness setting, or content that differs from the test content. Naming which one accounts for the gap is what turns a site visit into useful data.

Laboratory conditions are a model of the installation, not the installation itself. After the first units are deployed, verify at the site: at midday, at the worst angle, with the real content and the real surrounding surfaces.

This is where unexpected contributors appear – a nearby white wall, a polished floor, a canopy edge that produces a reflection the model did not include. Field verification is cheap at the prototype stage and expensive after a fleet has been installed.

Measurement protocol

The protocol that works in practice: fix the configuration, set the illumination and geometry, measure the bright and dark states with content representative of real use, repeat at each viewing angle, run the legibility check with an observer, and record everything in one document.

For the design side of the same problem – how brightness, bonding and surface treatment are chosen rather than measured – the guide to sunlight readability by design covers the trade-offs. This article is the verification half of that pair.

Frequently asked questions

How many nits do we need for outdoor readability?

There is no universal figure, because reflectance and viewing angle change the answer. Measure the intended configuration under the intended illumination rather than adopting a number from a specification sheet.

Does anti-glare treatment reduce image quality?

It slightly reduces clarity because it scatters light, and that trade is usually worth it in bright environments and questionable in dark ones. Test with the real content at the real viewing distance.

Can we compare two suppliers’ readability numbers directly?

Only if the illumination, geometry and content match. A number without conditions is not comparable with anything.