N.B. Death of the Zone System is a series about common misinterpretations of Ansel Adam's and Fred Archer's famous Zone System as photographers apply it to digital technology. This particular post is a continuation of Part III and should be read in preparation for the ideas of this piece.
Previously I addressed the oversimplification of Zone System by correlating the Zones directly to the dynamic range of the camera, as if this is the only part of the photographic process worth considering. Mostly I spoke to the myth that film has a smaller dynamic range than digital sensors and therefore why ZS does not need any alteration to the number of Zones and their definitions. In this post I will continue to explain where the number of Zones arises and upon what photographic or visual art materials it is based.
An important part of teaching ZS is emphasizing the whole system nature of how it works from the subject as perceived by the photographer all the way to the tonality they desire in the final print, projection, or electronically displayed image. Since Adams was creating the system for the reflective print he created the system around the visual limits of photographic paper. (These visual limits are not exclusive to photographic paper but are the same for any visual art involving pigment on or in a surface and made visible by a light reflecting off its surface.) Whether he knew it or not, by defining the limits around a reflected print he was actually giving us information about the ultimate end of the photographic chain: the human eye and the brain processing the information.
I personally was unaware of the limits of our visual system largely because most textbooks on photography and cinematography skimp on research about our visual system and often only speak to the dynamic range of the eye through its entire adaptation from night vision (scotopic) to daytime vision (photopic). Since our entire visual range throughout adaptation is enormous I was completely surprised when I read this statement in C.L. Hardin's Color for Philosophers (1988, 24-25).
It was earlier remarked that the eye is capable of operating over a huge dynamic range of light intensities, on the order of ten trillion to one. But for any single scene with which the eye can deal, the intensity ratio of the brightest to the dimmest object rarely exceeds 200 to 1, and this is about the limit of differences that can be signaled by a visual neuron (Haber and Hershenson 1980, 50-51; Evan 1974, 195; Barlow and Mollen 1982, 102).
After some quick calculation to translate this ratio into stops I was shocked to discover that our visual system has a DR of about 7 2/3 of a stop when stabilized at a specific light level. My initial reaction was incredulity and a knee-jerk rationalization that our eyes must be adapting through a wide range when viewing an image. And yet, I quickly came to realize this is far from the case because we do not view prints and projected images under changing lighting so they are providing a nearly stable quantity of light to our eye.
I have continually put to trial this 8 stop quantity in the research for my class only to continually find a great deal of scientific and practical support. (At some point in the future I hope to write a further discussion of the limits of our visual system with some more in-depth research, since there is some nuance here worth exploring.) Adams probably was not aware of what little research existed at the time, but he ultimately based his definition of Zones on the appearance of the final image. I encourage those who go on to read or re-read The Negative from these posts to pay close attention to where he separates discussions of the appearance of Zones from the materials capturing or displaying the image.
The Zone System allows us to relate various luminances of a subject with the gray values from black to white that we visualize to represent each one in the final image (1981, 47).
in this definition that starts Chapter 4 Adams is discussing Zones as previsualized or observed luminances and how they appear on the print but not in a manner that addresses camera technology. This reveals how much we should look to our visual system as what first defines the limits of our medium before considering next the display technology, then the capture medium, and finally the subject brightness range. This has always been the case in sensitometry that analysis of a photographic chain is performed backwards from display to capture medium. If you ever find a photographer making claims about Zones, Zone definitions, how many stops are in a Zone you must be wary if their discussion revolves purely around the capture medium because this is a mere part of a greater chain. Both film and digital sensors (recording in RAW or Log mode) do not render tonality in any way resembling how it should appear to the eye so why try and tie them directly to Zones? This short circuit approach to ZS is what causes conceptual trouble and leads to the pitfalls I described in part III.
So leveraging this knowledge that the eye at a given level of adaptation has a dynamic range of about 8 stops one can really understand how Zone System works, tuning our chosen photographic process to our visual system. Take at this diagram on page 52 of The Negative.
Note the 8 stop dynamic range from Zones I to Zones IX. An important point I should make is that the surrounding text upon inattentive reading can make it appear that the dynamic range or textural range depicted is for the negative. What he is discussing is exposures that give the appearance of this dynamic range or textural range.
Notice that the dynamic range depicted is again 8 stops. I tried to locate in the text surrounding this chart exactly what this dynamic range is? Camera, print? It's not entirely clear because he is addressing the visual appearance of Zones in context of the entire process. Whether he explicitly understood this he created the Zones based on the limits of human vision and this is reflected in scientific research performed much later. This is nicely illustrated in this diagram I found in the 4th edition of the Theory of the Photographic Process (James 1977, 545).
The scene curve on the right shows the eye adapted to a daylight with an average log luminance of 10,000 candelas/meters squared and how the visual system has a dynamic range of 7 1/2 stops at this light level. (Log Luminance of 3.5 - 1.25 = 2.25/0.3 = 7.5 stops) The curve of the print on the left is extremely similar in order to emulate these conditions. Discussions about the curve of the transparency will be addressed in a future post since the characteristics of our eyes in dark surround circumstances can change based on levels of adaptation. I must admit, there is some nuance here and I would like to further explore the level of adaptation of the eye for a monitor versus theatrical projection to further refine my claims.
This chart displays the curve of the human eye on the right versus the curves of a print and projected transparency. Notice how the limits of the human eye are quite narrow - 7 1/2 stops, and how the print matches this so that it appears the same as the original scene.
Another of my favorite diagrams I have found in my research is shown below and depicts the sensitometry curve of the eye at not just the photoreceptor stage, but at each stage of neural processing subsequent to the cones and rods.
From Optical Radiation Measurements Volume 5: Visual Measurements (Bartleson and Grum 1984, 252). For those unaccustomed to Log Luminance scales the farthest right curves (photopic adaptation and bright surround) for each transfer function has the following range in stops: Cones - 10 stops, Bipolar Cells - 6 1/2 stops, Ganglion Cells - 6 1/2 stops. I am currently looking for the original test data to make sure there is no information in the shoulder of the curve as per my recommendation in the previous post.
By now I hope it is clear why the Zone System ties together our eyes and the photographic process and that the definitions of 11 Zones will remain the same no matter what the technology and its dynamic range. (This may sound heretical, but will probably be the case even including HDR monitors and projectors). Simply, the majority of image displays whether reflective or transmissive look correct to our visual system so long as they produce 8 stops of tonality framed by an extra stop or stop and a half both sides to give us the visual sensation of a rich black and a pure white.
My hope in dispelling the mistaken idea that Zones directly correspond to sensor DR goes beyond just technical clarification. The problem also involves the dissemination of mistaken info around the internet and how these ideas are spread with a veneer of authority. The lack of research and unquestioned assumptions of online "experts" is irresponsible. In my previous post I used a quote from Martin Bailey's blog which is an excellent source of information on photography and which I recommend to amateur photographers. My reason for using his blog is simply because he handed me the best instance of this mistake. He too equates Zones to sensor DR only to find the same numerical mismatch of 11 Zones to the 12 stops of DR of his Canon DSLR. However, just after this he claims:
"... so it’s a bit wider than Adams’ definition, but in practice I’ve found that even now, thinking of each zone as a stop of exposure works fine."
This last comment is an offhanded acknowledgement that he has found ZS to work just fine despite it having a smaller DR than his camera. That's because ZS works when you base your workflow around your visual system and not the entirely arbitrary relation of numbers you find in books and discussions. Martin Bailey gets good results because he has a strong visual memory of the Zones, a well-calibrated monitor to look at his photographs, a particular curve he applies to the RAW image from his camera, and takes photographs assuming the limits of the ZS which are based on the human eye. Which is why analysis from the final display backwards through the capture medium to the subject is so important; if one blindly makes adjustments through each step of the photographic chain then the photographer will not fully understand their process. Ideally, they should know the real photometric values of the light reflected or emitted from the display and how this relates to the light intensity through each step of the process all the way to the initial reflected or emitted light from a subject in order to have a comprehensive understanding.
Now, one very important clarification I should make about the 8 stop dynamic range of human vision. This fact is generally true but can change somewhat based on viewing circumstances such as whether our eye is observing a bright image in dark room, or in a brightly lit location (which in scientific parlance is referred to as bright surround and dark surround.) I would like to address this in a piece further down the road as I want to obtain the full data in order to extrapolate some important lessons.
Finally, my hope after parts III and IV is that anyone pursuing the photographic arts begins to put themselves back into the picture; all puns intended. That is, to begin to learn and pay attention to how they perceive the world and how display technologies work. The stress on camera dynamic range is really more of a marketing tool, a number that grows over time to impress gear heads, but is not necessarily important to ZS. (I will demonstrate this in Part V of this series). Be careful to avoid simplistic numerical associations and first learn how to see.