In image editing, the room illumination matters

The day before yesterday I wrote about the importance of the image surround for image editing.

In image editing, the surround matters

The reason the surround is important is that it affects the state of adaptation of you, the image editor and therefor the way that tones and colors appear.

Let’s imagine that the image under consideration has no surround: it fills the whole screen. In effect, the room that’s visible beyond the screen becomes the surround for the image, and the brightness and the color of the objects in the editor’s visual field affect adaptation, and thus the way the image appears.

Now make the image a bit smaller, and fill the part of the screen that’s not the image with a solid color. That affects adaptation. Not very much if the band of solid color around the image is small, and a lot if it is large. The adaptation of the viewer is affected by both the surround and the objects in the visual field but which is more important depends on the size (and tone, but I’m not going into that) of the surround.

It is usually the case that the objects around the monitor are not self-luminous: they reflect room light incident upon them. It is further the case, absent conscious efforts in setting up the room, that the reflectance of the objects averages about 18% and the chromaticity is approximately neutral. If that’s the case, the state of adaptation of the viewer in the no-surround case is entirely determined by the brightness and spectrum of the room light.

If the on-screen surround is of neutral chromaticity, it has the same white point as the monitor. You can avoid adaptation chromaticity shifts by making the room illumination have the same white point as the monitor. That is the basis of the once-universally-recommended D50 monitor white Point, D50 room illumination (Macbeth used to make D50 fluorescent tubes; maybe they still do), and D50 illumination in your print viewing station. That works, and is the gold standard even now, but is unnecessarily strict. With proper adaptation, you can edit with a D65 white point, and view your prints with a D50 white point, and things will look just fine.

As an aside, one of the reasons for not setting the monitor white point to D50 doesn’t apply any more. In the CRT era, and in the LCD with fluorescent backlight era, D50 was often too dim. With LED backlights, that is no longer the case.

If you’re going to edit on a D65 monitor, it’s not going to be easy to find D65 room illumination, so you want to turn the room illumination down to the point where it is not significant in your adaptation. With a black surround, that’s really dark – I’ll guess 4 or 8 lux, assuming your monitor is set to 80 to 100 candelas per square meter. You probably want enough light so that you’re not bumping into things, so you probably will opt for somewhat brighter illumination than that. If so, consider at least a gray surround, and do use a white one for soft proofing.

There’s another reason to turn down the lights when editing. Monitor calibration takes place with a bop over the sensor that obscures room light. That means that the sensor sees the deepest blacks the monitor can produce. When you view that monitor in anything but a dark room, the room light bounces off the screen, making the dark tones lighter than they were when the screen was calibrated. Some calibration systems have ways to compensate for that, but, in my experience, they don’t work very well.

In the old days – in digital photography that’s the 1990’s – many high end monitors came with hoods to partially ameliorate the effect of stray room light falling on the monitor. Sadly, that is not the case anymore. However, both NEC and Eizo make accessory hoods for their monitors.

If you want to get really picky, consider that the reflection of the image and the surround on the monitor will reflect off your shirt and affect the dark tones. If you are looking for a reason to adopt the Steve Jobs black turtleneck look, you’ve got one now.