April 29, 2021

Doesn't brick or colored paint "ruin" accurate bias lights?

We get this question a lot, and I want to provide some perspective.

First, let me just say that if you are color grading video, you absolutely want to have the most control over the environment that you can have. This includes spectrally-flat paint and light control - i.e. no light contamination from windows, glowing LED displays on devices, etc.

Now, with that out of that way, there are absolutely times where this is not possible, and many colorists have told me about working out of hotel rooms or, more recently due to the pandemic, from home.

I'd like to point out a few things that many of us intuitively know:

We don’t calibrate a TV for the color of the paint in the room. We calibrate it for D65, which is what the white point of the light should be.
The color of the paint doesn't impact the color of the light very much but the color of the light does impact how accurate the paint looks to us.

Think of a night club or party with colored lights. There is a huge difference between being in a white room with red light and a red-painted room with white light. The walls may appear to look a similar color, but everything else in the room looks drastically different.

Simply put, under red lights, everything in the room will appear to be red. Your skin will look red, your clothing will look red, and everything else under the red lights will look red.

On the other hand, if we are in a room with red paint and a white light source, this will not be the case (unless the walls have a very high specular reflectance - think a red-tinted mirror or even glossy red paint, like a sports car).

You can even stand right next to the red wall and have the white light bounce onto you and you will still not look red (unless you have a really bad sunburn).

I'm going to discuss two different things. The first is called chromatic adaptation and the second is opponent-process color theory.

We adapt to the color of light around us pretty quickly through a process called chromatic adaptation and that's a different process from opponent-process color (color wheel) theory. Both of these things are going on, but chromatic adaptation has outsized role when watching a transmissive display, like a TV or monitor.

Basically, we stare a TV without changing our angle very frequently, so opponent-process doesn't really impact the image because if you adapt to the blue wall, it's mostly impacting your vision around the screen and not the screen itself.

More than the color of the paint, you'll be adapting to the color of the light in the room from the bias lights as the sole light source.

Think about this: How much does the paint impact the TV with other lights on? This is really no different. Ideal bias lighting should be nothing more than a light source of the right white point in the best possible location.

There are different things going on when we watch TV in a room with ambient light.

Opponent process color theory - Example: Marketers put green labels on tomato sauce to make the sauce look more red/ripe. Stare at an image of the American flag for 30 seconds and look away and we see the inverse afterimage:

Chromatic adaptation - We adapt to our lighting. If I look at my phone under 3000K incandescent bulbs or candlelight, the screen looks bluish under warm light and it looks magenta under low quality, greenish light. If you have a newer apple iOS device, turn truetone on and off to see how the phone (and you) adapts to lighting, not to the color of textiles or paint in the room.

Metamerism index / Low CRI (color rendering index) light sources - We see poorly in low CRI light. We can see better under a dim, higher CRI light than a brighter low-CRI light. Think of mismatching blue and black socks under a bad light.

Look at how white light bounces off of your blue wall onto the white ceiling. You don’t see a blue reflection on the ceiling. This is very different than if you reflected blue light off a blue or white wall onto a white ceiling.

The color of the paint has less impact than the color of light. This makes sense. We don’t calibrate a TV for the color of the paint in the room. We calibrate it for D65, which is what the white point of the light should be.

If we try to "correct" for the color of the wall by bouncing red light off a blue wall, we don't truly get gray (a red surface would not reflect blue light. Instead, you'd get darkness). However, paints are not purely red or blue. They contain a mixture of pigments. If we try to correct the wall color with an opposing light color, we will end up bathed in inaccurate light and end up adapting to it, making the display look wrong.

All of this is a long way to say that if you have beige, powder yellow, light green or blue walls, they have surprisingly little impact on the white point of light in the room. And, if you have colored walls, as so many people do, accurate lights are going to still measure very close to D65 from where you'd be sitting.

However, when you can paint the walls gray, it really lets your display shine, and if you are professional colorist, you obviously want maximum control over your environment, which depends on the situation. Colorists spend a lot of time scrutinizing a single frame of a scene while most of us at home don't really press pause and stare at something for very long.

The gray paint provides an extra level of scrutiny that a colorist needs. This also explains why the recommended brightness for professionals and consumers is different.

The recommended brightness of bias lighting can vary based on the user. While production professionals usually prefer a dim surround with a lower brightness (4.5-5 cd/m^2) because it helps them see more acutely than with higher light levels, consumers often enjoy higher-brightness settings (10% of the maximum brightness of the display) when watching their favorite series at home because this makes colors really stand out and improves perceived black levels.

Previous article Which length of bias lighting do I need for my TV?

Next article Eye Strain and OLED: The Truth is that It's Worse