A Cinema Luminance Range by the People, for the People: Viewer Preferences on Luminance Limits for a Large-Screen Environment, by Suzanne Farrell, Timo Kunkel, and Scott Daly

A number of websites have published the findings of Dolby’s study of luminance preferences for the small screen, yet they all but ignore the fact that Dolby also conducted a similar study with a large screen. Many assume that the larger screen in a movie theater only needs a fraction of the luminance of a consumer television in order to appear to have the same lightness. However, if the screen in a theater only needs to be around 70% as bright as televisions or computer displays as some believe, then they’re still only achieving a few percent of their potential maximum luminance. In Dolby’s study of viewer preferences for the small screen, it was found that a black level of 0.005 cd/m2 and a highlight level of 20,000 cd/m2 was necessary to satisfy 90% of viewers. Some have argued that such a large dynamic range is unnecessary, as the steady state dynamic range of the human visual system (HVS) is in the neighborhood of 3.6 log units, but this is only true for a single observer and for brief durations. Predicting an individual’s steady state while seated in a theater is next to impossible, and in a venue with multiple viewers, each can be viewing different parts of a scene and be in a different state of adaptation. Participants in the study sat three picture heights (3 PH) from a 13 ft. screen with all the lights out except for the exit sign. The same test images as used in the small screen study were used for this one. It was found that viewers preferred a dynamic range of 22 f-stops regardless of screen size.

Photo: Dolby

2 thoughts on “A Cinema Luminance Range by the People, for the People: Viewer Preferences on Luminance Limits for a Large-Screen Environment, by Suzanne Farrell, Timo Kunkel, and Scott Daly

Add yours

  1. Using data from ground based telescopes, space shuttle missions, satellites, high altitude aircraft and computer models, scientists have standardized on a value for average spectral irradiance for Earth. This quantity is known as the solar constant, is 1366.1 watts per square meter at the distance of Earth’s orbit. However, this is for the full spectrum which includes visible and invisible solar radiations. Sticking to just the visible spectrum for our sun or any star with a black body temperature of 5778K is 93 lumens per watt or 127,000 lumens per square meter. Converting to candelas we divide by 12.57 and get 10,103 candelas per square meter, or in other words, 10,103 nits, the brightness of our sunlight.

    If you included the IR radiation along with the visible spectrum, the notion that twice this amount or double the Sun’s visible irradiance was necessary to satisfy 90% of the viewer’s and we begin to see a problem. No longer a dark theater, this is daylight viewing under a giant magnifying glass. Beyond sunburn, this is spontaneous combustion. What light source projects this magnitude of brightness on a 13 foot screen? And you can still see the exit sign?

    To be fair, you did give that 20,000 nits was for small screen. But how small? A pinpoint specular highlight from a laser? To be honest this study from Dolby if it is that, obfuscates behind an assortment of ratio metrics, PH to seating ratio, log, DR, theater to computer or television ratio and f-stop, HVS without giving a single hard quantity of anything except 20,000 nits and 13 feet which I can assure didn’t happen that way if the exit sign was something readable in the dark. Because anything other than flashing lasers into your eyes, and they may well have done that for all we know, but to somehow arrive at a conclusion that 90% weren’t satisfied with anything less, I can’t buy that.

    Could it have been a mistake? 2,000 nits perhaps? Would be much more believable.

    1. The large screen highlight preference was 7,000 nits, black level preference 0.002 nits, lower than the small screen study. The Christie projector could only output 2,100 nits so the figures were extrapolated. Naturally, the participants would tone down large bright areas to avoid discomfort. The DCI test also showed viewers prefer much higher luminance levels on the big screen than are currently available. To put things into perspective, 10,000 nits is the brightness of an ordinary fluorescent lightbulb. A small specular highlight or emissive light of that intensity for a brief duration on the large screen would not irreparably damage anyone’s eyes.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Blog at WordPress.com.

Up ↑