A car showroom LED wall that made the paint look wrong

The dealer installed a 6 by 3 metre wall behind the cars. The silver car turned blue on camera. The problem was not the wall, it was the 6500 K lighting in the room.

All insights →

The complaint

The showroom wall was installed and looked fine in person. Then the dealer's marketing team shot a video for social media and the silver car in front of the wall looked baby blue. They blamed the screen.

What was actually happening

The wall was set to 6500 K, which is a cool daylight white. The showroom lighting was 4000 K warm white. The car between them was reflecting both. On camera, the white balance was locked to the warm room light, so the wall went blue and the car followed.

The measurement

We put a colour meter on the wall showing a white field: 6480 K. On the car bonnet under the showroom lights: 4100 K. On the floor: 4250 K. A 2400 K mismatch across a 3-metre gap.

What we changed

Set the wall to 4200 K to match the room, not to the broadcast standard. Slight loss of blue saturation on the screen, but the cars now read correctly on camera and in person. Then we added a scene preset at 5500 K for when the dealer shoots with their own lighting kit.

Why the default was wrong

6500 K is the default because it is the standard for broadcast and for most LED walls in commercial spaces. In a showroom, the product is the car, not the screen. The screen has to match the lighting that sells the car.

The general principle

In any retail space where the screen sits next to a physical product, match the screen white point to the ambient lighting, not to the factory default. The screen is part of the lighting design, whether the lighting designer planned for it or not.

One more detail

The wall was also too close to the cars. At 1.2 metres behind the bonnet, the reflection of the screen showed on the paint. Moving it to 2.4 metres and adding a slight downward tilt removed the reflection without changing the content.

Related products

Let’s find the display your project needs.

Discuss your project ↗

Language