Screen Uniformity Testing

What uniformity means

A screen is uniform when the same input produces the same light across the whole surface. If you show a flat white image, every part should look equally bright and the same shade. If you show flat gray, the same applies.

Real panels rarely achieve this perfectly. A corner may be a bit dimmer, a stripe may look slightly warmer, or the middle may be brighter than the edges. Uniformity testing measures how much, and whether it is acceptable for the intended use.

There are two aspects. Luminance uniformity is about evenness of brightness. Color uniformity is about evenness of tint. A screen can be good at one and weaker at the other.

Why panels are uneven

The causes depend on the technology, and the article on LCD backlights and the article on organic light emitters describe the underlying parts.

• Backlight distribution. On an LCD, light has to be spread evenly from the edge or from an array of sources behind the panel. Small differences in how the light spreads can leave brighter or dimmer zones.
• Pressure and mechanical stress. If the layers are squeezed by the frame or by a tight bezel, light can leak in places. Backlight bleed and clouding are the best-known results, and they are most visible on dark images.
• Variation in the panel itself. Liquid crystal thickness and pixel properties vary slightly, which can create faint patches or banding. Organic emitters can differ slightly in brightness or wear from pixel to pixel, so they too can show unevenness.
• Driving and compensation. Manufacturers sometimes apply a correction map to smooth these differences. Its quality varies between models.
• Aging. Parts of a screen can age at different rates. The article on panel aging covers how brightness and color shift over time, and image retention can leave patterns that look like unevenness.
• Temperature. Heat from electronics can warm part of a screen and change how it behaves, a topic related to display heat.

Checking by eye

You can learn a great deal about uniformity without any instrument.

1. Warm up the screen for a while, and disable automatic brightness and dynamic contrast.
2. Dim the room so that reflections do not hide faint patches.
3. Show solid colors one after another. Full-screen mid-gray is especially good, since it makes cloudiness and tints stand out. Also try dark gray, white, and each primary color.
4. Sit centered and at a normal distance, then move slightly from side to side. Note whether the unevenness moves with your head, which suggests a viewing angle effect, or stays put, which suggests the panel itself.
5. Look at black in a dim room for glow near the edges and corners.
6. Photograph carefully if you want a record. Cameras adjust exposure automatically, so a photo can exaggerate or hide patches. Use fixed exposure if you can.

Since your eye adapts, it can be fooled. A patch that looks dark may only be darker than its surroundings. Comparing to a second screen or turning the screen off and on helps you judge.

Checking with an instrument

For measured results, the general method is a grid.

• Divide the screen into a grid of zones, for example a small number of rows and columns. A commonly used scheme is nine points, three by three, but the count depends on the method and on how demanding the check is.
• Show a flat field, such as white or a mid-gray.
• Measure the luminance and color at each point with a suitable instrument.
• Compare the readings with the center or with the average.

A hypothetical example with simple numbers: suppose the center of a white field measures 200 nits, and the dimmest of the grid points measures 170 nits. Relative to the center, that point reads 170 divided by 200, which is 0.85, or 85 percent. In other words it is about 15 percent dimmer. Different organizations summarize this in different ways, such as comparing the dimmest point with the brightest, so a stated uniformity figure only makes sense when you know the method. The numbers here are invented for illustration.

For color, the readings are compared as tint differences. Small tint differences are hard to see at low levels, and most people notice a difference more easily when two regions sit side by side.

More thorough methods use a camera-based system that captures the whole screen at once and produces a map of brightness and color. These are faster for detail, but they need careful setup and are usually found in manufacturing and laboratories.

Which images reveal what

Different test images reveal different faults:

• Mid-gray is often the most revealing for general unevenness, because both brighter and darker departures are visible.
• Black shows leakage and glow.
• White shows tints and dirt or marks.
• Primary colors show whether one color channel varies.
• Low-level grays reveal banding or dither pattern differences that vanish at higher brightness.

The article on test patterns shows how simple images expose such issues, and color and brightness checks are discussed in the articles on color accuracy and on brightness and contrast.

How much is acceptable

There is no single answer. Almost every panel has some variation, and a small amount is normal. How much matters depends on the use:

• For video watched in a dim room, uneven black and clouding matter more.
• For photo or design work, tint differences across the screen matter, since an edited color may look different in one corner.
• For general office use, modest variation is usually not noticeable.
• For safety-critical or clinical uses, requirements are set by regulators and specialists, and this article provides only general information.

Manufacturers may state tolerances, and buyers who need tighter control sometimes specify them when ordering. Whether a particular screen counts as defective is usually a matter of the seller’s or maker’s stated policy.

Improving matters

Some options exist, but none is a guarantee. Reseating a loose bezel can ease pressure-related leakage in some cases. Lowering brightness reduces how visible bleed is. Some professional screens use stored compensation to even out the panel, while others rely on the user to calibrate. Calibration can improve the center of the screen, but it generally cannot repair an uneven panel. If unevenness is severe or new, a warranty or return option may be worth exploring.