Tone Reproduction

Zone System and sensitometry combined

We have gained a basic understanding of the Zone System and sensitometry in the previous two chapters. Combining the two will allow us to get control over image contrast and understand how zones are represented throughout the image reproduction cycle. Ultimately, they show us the importance of tonal gradation and how to manipulate it through exposure, development, and critical material selection. Studies conducted by several authors, including my own, indicate that most viewers prefer prints with a full tonal scale from pure white to solid black and an abundance of gray tones in between. Therefore, a widely accepted print is likely to have maximum contrast and full gradation. Im not saying that a fine print requires all tones all the time, but most people like it that way, most of the time. Apart from the creative license of the artist, we generally like our printed images to be a reasonably true representation of the scene captured. Unfortunately, in many cases this is impossible, since the average subject brightness ratio is far greater than the maximum print brightness ratio. As a consequence, the resulting print is often a disappointing record that fails to satisfy our memorys expectations. The most efficient man-made reflectors cast back as much as 98% of the light that reaches them, but natural objects are limited to about 90% reflection. On the other hand, natures best absorbers reflect as little as 1.5% of the light that they receive. If uniformly illuminated, the maximum reflectance ratio of an outdoor scene is, therefore, about 60:1 (90/1.5) or 6 stops. Although these extremes are not unlikely, average outdoor scenes have more moderate reflection ratios of around 30:1 or about 5 stops. So far, we have assumed perfectly diffuse lighting, but the various objects in an outdoor scene rarely receive the same illumination. The subject lighting ratio between direct light and shadow illumination

can be controlled in a studio environment, but in natural daylight, it can reach values of up to 12:1 on a clear sunny day. This can significantly alter the subject brightness ratio. Lets assume that we have a very dark and a very bright object placed in a sunlit scene. If the bright object is moved into the sunlight and the

Tone Reproduction

113

fig.1 The measurements taken by Jones and Condit in 1941 serve as a starting point for an objective tonereproduction analysis. They obtained data on the subject brightness ratio of 126 outdoor scenes. The lowest-contrast scene had a value of 27:1, and the highest-contrast scene had a value of 760:1. For the whole group of scenes, the average subject brightness ratio was 160:1.

fig.2 During tone reproduction zones are transferred from the subject, through the negative, to the paper. The even zone spacing of the subject zone scale is altered throughout the cycle. The extreme zones are typically compressed while medium zones are often expanded. In the darkroom, the negative is projected onto the paper, and the textural negative density range becomes the textural paper log exposure range. The paper curve is turned sideways to accommodate this fact in this example of a film with normal development printed onto normal graded paper. The gray reference scale will be used in the rest of the book to identify tonal alterations due to material and processing modifications.

Comparing this with the field test, it is evident that the paper is unable to realistically represent an averfrequency of 12 age outdoor scene, by about a factor of 2 or roughly subject brightness ratio occurrence a stop. The next example can help us visualize how in outdoor scenes the paper is falling short of our expectations. We 8 know from the Introduction to the Zone System that Zone V has a reflectance of 18%. Zone VI is twice 4 as bright in the subject zone scale, and therefore, it must have a reflectance of 36%. Consequently, Zone VII has 72% reflectance and Zone VIII must have 0 10 20 30 50 70 100 200 300 500 700 1,000 twice that. But wait a minute! We said that the paper subject brightness ratio cannot reflect more than 90% and this calculation would get us above 100%. Besides, we are still missing Zone IX and X. The answer can only be that we do dark object into the shade, then the subject brightness not have a linear relationship between subject zone ratio is maximized to 720:1 (60x12) or almost 10 stops. brightness and print zone reflections. Photographic These numbers were verified in a field study published paper cannot handle the brightness ratio of an average in November 1941 by Loyd A. Jones and H. R. Condit, outdoor scene. Zone V can be represented realistically who analyzed data from 126 different outdoor scenes in at about 18% reflection, but the extreme zones must which the subject brightness ratios ranged from a low be compressed to fit into the print zone scale. This is 27:1 (about 5 stops) to a high 760:1 (almost 10 stops), why it is so challenging to capture the sparkle we reaveraging at 160:1 or a little more than 7 stops (fig.1). member from the original scene. Many printers labor The greatest possible print brightness ratio on the with dodging and burning tools to bring some of the other hand, is limited by the reflection density ratio sparkle back. The Zone System does not eliminate of the paper. The white base of unexposed but fully these techniques, but it maximizes image contrast fixed photographic paper is capable of reflecting about control and, therefore, actually promotes dodging 90% of the light. A fully exposed and developed paper and burning from being a poor rescue attempt to a with a glossy finish can be so dark that less than 1% of powerful tool of creative print manipulation. the incident light is reflected. However, the extreme toe and shoulder regions of the characteristic curve Film and Paper Are Setting the Tone are of little use to pictorial photography, due to the Tone reproduction is one of the most important factonal compression they cause, and hence, the useful tors in print quality. Fig.2 shows how the tonal values print brightness ratio is reduced to about 80:1. transfer through the image reproduction cycle from
average ratio 160:1
(126 measurements)

16

number of scenes

1.8
pictorial range

0.0

0.6

0.9

0.3

1.2

2.1

1.8

1.5

1.5
1.29

textural range

0.09

IX

textural negative density range

1.2 0.9 0.6 0.3 0.0 0 I

VIII
e pm nt

textural paper log exposure range

VII VI V IV III II
I 0

textural print density range

no

rm

al

de

ve

lo

gra

de

2
1.89

0.24

VII

VIII IX

II

III

IV

VI

VII

VIII

IX

X
normal

III

IV

VI

II

0 I

Subject Zone Scale

Negative Zone Scale

Print Zone Scale

114

Way Beyond Monochrome

average outdoor scene brightness ratio = 160 7 zones log subject brightness range = 2.10 6 zones log subject brightness range = 1.80
1/2 1 2 4 8 16 32 64 128 256

pictorial range textural range

512

light units

II

III

IV

VI

VII

VIII

IX

Subject
diffused highlight specular highlight

loss of tonality
Optical Image

fig.3 This diagram, based on a Kodak original, shows how the zones are compressed and expanded in the tone-reproduction cycle from the scene to the final print. It also shows the influence of the optical images on tonal values, and it helps to standardize a few key terms and values for negative and print density.

camera lens with moderate are

pictorial negative density range = 1.20 textural negative density range = 1.05 III IV V VI VII
1.37 1.29

0.24

0.10

base+fog
standard paper Dmax limit

II

VIII

IX

Negative
developed to avgGrad = 0.57

textural print log exposure range

Optical Image

enlarger lens with moderate are

VIII

paper Dmax
2.10 2.15

I
1.97

II
1.89

III

IV

VI

VII

Print
diffusion enlarger grade 2 standard paper Dmin

6 zones textural print density range = 1.80 7 zones pictorial print density range = 1.90 maximum print density range

useful print brightness ratio = 80

the subject, through the negative, to the print, while scale. After exposure and development of the film, they creating their individual zone scales. This example are represented in the negative zone scale. Most modexplores the influence of film and paper, but it ignores ern films have a relatively straight characteristic curve any optical or physical equipment influence from except for the toe and the extreme areas of the shoulcamera or enlarger. Assumed is a negative, which was der. Therefore, severe tonal exposed and developed as normal and then printed compression is restrained to the onto a grade-2 paper. Ansel Adams supposedly aban- extreme shadow and highlight Throughout the book, the term subject doned the terminology of zone scales for negative zones. All other zones in the brightness range is used to describe a and print in his later years, because he felt that zone negative are also compressed, range of measurable light intensities. scales are only applicable to the subject brightness but more or less evenly spaced. Strictly speaking, this is not correct, range, turning into tonal scales for negative and In the darkroom, the negative because brightness only refers to the print. Nevertheless, we will maintain the zone scales is projected onto the paper, and human perception of luminance and not throughout the reproduction cycle in this book. It is the different negative densithe measurable quantity of it. The term consistent, and it helps to understand how zones are ties correlate to different print subject luminance range is technically represented in negative and print, and how different exposures. In other words, the more accurate, but it is not frequently materials influence them. textural negative density range used in photographic literature. To avoid Zones II, VIII and V are highlighted, because they becomes the textural paper log confusion, we have chosen to use the are considered to be the boundaries and the center of exposure range. In fig.2, the incorrect, but more generally understood, the pictorial range, respectively. All zones start out paper curve is turned sideways term and its abbreviation, SBR, instead. evenly spaced, exactly one stop apart in the subject zone to accommodate this fact.

0.09 0.07 0.05

Tone Reproduction

115

Zone 0

Negative 0.00
0.03 0.07

Print 2.10
2.09 2.06

Monitor

0.10
0.14 0.19

2.04
2.00 1.95

II

0.24
0.28 0.33

1.89
1.81 1.72

III

0.38
0.43 0.49

1.61
1.48 1.34

IV

0.54
0.60 0.66

1.19
1.04 0.89

0.72
0.78 0.84

0.75
0.62 0.50

VI

0.90
0.97 1.03

0.40
0.32 0.25

VII

1.10
1.16 1.22

0.19
0.15 0.12

VIII

1.29
1.35 1.42

0.09
0.08 0.07

IX

1.48
1.55 1.61

0.06
0.05

1.67
1.73 1.79

XI

1.85
2000 - 2006

fig.4 This table shows standard Zone System values for relative negative transmission and absolute print reflection density in 1/3-stop increments. To illustrate the relationship between the analog and the digital Zone System, typical grayscale values for computer monitors are also shown in K%.

more interest, and much simpler, for us to follow the 7 stops of the pictorial range, from the beginning of Zone II to the end of Zone VIII, through the tone 100% reproduction. We will, therefore, ignore the remaining 1/6 stops on either end of the average outdoor scene. They fall within Zone I and IX and have little tonal 99 98 value anyway. In addition to the density values of the 97 pictorial range, it will prove useful to also track the 96 textural range, which reaches from the center of Zone 95 II to the center of Zone VIII. But for now, lets keep 93 in mind that we are starting with seven zones, or a 90 subject brightness range of log 2.1. 86 Zone compression starts as soon as the lens has 82 formed the optical image in the camera. Minute lens 77 and camera flare, caused mainly by the higher zones, 71 bring non-image forming light to the lower zones. 64 Consequently, the zone scale moves to the right, com56 pressing the lower zones and leaving the higher zones 48 40 unaffected. The resulting image is projected onto the 32 film and is turned into negative densities through de25 velopment and processing. A typical film characteristic 19 curve has a relatively low gradient, which compresses 14 all zones evenly, except for the toe and shoulder, where 10 the low and high zones are compressed more severely. 6 A quality negative, suitable for a diffusion enlarger, 4 has a negative density range of about log 1.2 to cover 3 the seven zones. This is little more than half of the 2 original subject brightness range. 1 Zone compression continues in the enlarger. Simi0 lar to the light distribution in cameras, enlargers and their lenses suffer from flare too. However, since image tones are reversed in the negative, the more luminous Tone Reproduction and the lower zones now scatter some non-image forming light Zone System into the higher zones. As a consequence, the zone scale Ralph W. Lambrecht Fig.3 also illustrates how the zones are moves to the left, this time compressing the higher compressed and expanded in the photo- zones and leaving the lower zones unaffected. After graphic reproduction cycle from the scene to the final image projection and paper processing, we can evaluprint. The difference to fig.2 is that the characteristic ate the final print. Photographic paper, by design, has curves are not shown anymore, but in addition to their a much larger density range than the negative, which contribution, the influence of the optical images from provides an opportunity to expand the zones again. camera and enlarger are explored. Moreover, a few key Nevertheless, toe and shoulder of the paper characvalues for negative and print density are labeled so we teristic curve compress the already compacted shadow can start to create a personal density standard, which and highlight zones one more time, but midtones are will function as a reference for your own values. expanded again and often exaggerated. Fig.3 presents the evenly spaced subject zones In the end, the compression from the subject with their doubling light units on top. The average brightness range to the negative density range, and the outdoor scene, with a subject brightness ratio of subsequent compression and expansion to the print 160:1, is equivalent to about 7 1/3 stops. It will be of density range, have shifted the evenly spaced zones of

Zone spacing changes again as soon as the negative is printed, which is represented through the paper characteristic curve. The already compressed highlights are hitting the pronounced toe of the paper curve, where they are compressed even further, making highlight separation difficult. Quite the opposite is happening at the center of the zone scale. Zone V is falling onto the steep portion of the paper curve, and it is being expanded as a result, causing increased tonal separation in the midtones. At Zone II, the process is reversed into tonal compression again, due to the influence of the film toe and paper shoulder. The actual tonal representation depends on the film, developer, paper, and to some extent even on the equipment used. It may also differ somewhat from the simulation in fig.2. Nevertheless, the fact that zones are being compressed and expanded, while moving through the reproduction cycle, is realistically presented. Approximating gray tones were added to fig.2 in order to provide a visual relationship for the print zones. This normal tonal scale will be used as a reference throughout the rest of the book, showing how any material or processing change will alter the tonal representation.

116

Way Beyond Monochrome

From the field study conducted by Jones and Condit, we know that the subject brightness range of the average outdoor scene is wider than the density range of photographic paper. We have accepted the zone compression at the highlight and shadow end of the tonal scale to get a full-scale print, but now, we want to know how this compares to an exact reproduction of scene luminance. The theory of tone reproduction is divided into an objective and a subjective, or psychological, method. The study of subjective tone reproduction compares the visual sensation of the human eye, also called brightness, as a response to the subject luminance of the original scene, to the sensation created when viewStandard Values and Their Manipulation ing the final photograph and its surrounding areas. We The table in fig.4 shows a collection of standard Zone will be discussing some of these effects as they influSystem values for relative negative transmission and ence our choice of print mounting and illumination absolute print reflection density in 1/3-stop increments. in coming chapters, but a more detailed discussion is To illustrate the relationship between analog and digi- better left to more specialized scientific literature. tal Zone System, typical grayscale values for computer monitors are also shown in K%. The development of these numbers was based on a few material and equipment assumptions, and they may not be completely valid for all photographers and their material choices. Therefore, I do not claim absolute validity for these numbers, but I am confident that they apply in most situations where a film has been exposed and developed normally (N), and a diffusion enlarger was used to print onto a bright white photographic paper having a pearl or glossy finish. The standard densities for negatives and prints change as soon as the film development is altered to control a more or less demanding subject brightness range. When the subject brightness range is larger than normal, a reduced development time is chosen to compensate. The inclusion of additional zones will push all other zones up in density and they will become darker than normal. When the subject brightness range is smaller than normal, an increased development is chosen to compensate. The development time is extended to increase the negative densities of the middle zones, avoiding an otherwise dull negative and print. The exclusion of some zones will pull all other zones down in density and they will become lighter than normal. Development and Film Processing provides more detail on this subject. The next chapter will show how different materials can influence the tonal scale.

the subject significantly. As a rule of thumb, low and high zones are usually compressed and middle zones are typically expanded. There is little similarity left between the original subject zone scale and the final print zone scale. With modern papers, a pictorial print density range of log 1.9 is typically available to cover the original seven subject zones. This almost brings back the original subject brightness range of log 2.1. However, we must realize that the original brightness ratio is significantly reduced, from 160:1 in the subject down to about 80:1 in the print. We have to accept this material behavior to some extent, but throughout this book, we will discover ways to work around it or manipulate some of it to our advantage.

Tone-Reproduction Theory

fig.5 This illustration shows how the photographer combines imagination and knowledge to bring the tonereproduction cycle full circle. Applied with experience, this can be done if the final print is meant to be a close reproduction of the original scene or an artistic expression thereof.
(illustration 1976 by White, Zakia and Lorenz, 'The New Zone System Manual', Morgan & Morgan, Inc. Dobbs Ferry, NY, USA, ISBN 0-87100-100-4)

Tone Reproduction

117

fig.6 This is an example of a detailed tonereproduction cycle for normal film development and normal paper contrast. Quadrant 4 shows the resulting objective tone-reproduction curve. In the preferred print, highlight and shadow detail is sacrificed for a higher than objective contrast in the midtones.

The study of objective tone reproduction, on the other hand, compares the densities of the photographic print with the log luminance of the original scene. This provides information on how closely the photographic process has come to represent an exact tone reproduction of the subject luminance. The illustration in fig.5 is a simplified view of how the combination of imagination and skill brings the tone-reproduction cycle full circle. Before the actual picture is taken, the scene is viewed with the final photograph in mind. The photographer takes a look at the scene and forms a mental representation of the intended reproduction. The brightest highlight cannot be brighter than the papers white, and the darkest shadow cannot be darker than the papers black. Armed with the necessary experience and knowledge

1.8 1.5 1.2 0.9 0.6 0.3 0.0 0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 2.7 3.0

Photographic Film
analog negative

Q2

X IX VIII

Photographic Paper
analog print

Q3

m or

al

de

ve

m lop

en

VII

VI V IV III II
I 0 VII
VIII IX

nor

mal

con

tras

Exposure Values
log exposure values

Print Values
absolute log reection density

II

III

IV

VI

VII

VIII

IX

X 0

no

re
no rm al re a

ex a

Camera & Lens

ct t

Q1

on

Tone Reproduction

118

Way Beyond Monochrome

re pr od

e n tiv tio t jec uc rin ob prod ce p re hoi e c ton rst for

Q4

about the equipment and materials used, film and paper are then exposed and developed to create the visualized print. Depending on the photographer's intent, this is done to either obtain a literal recording of the scene or a creative departure from reality. Fig.6 illustrates a more detailed example of an objective tone-reproduction study for normal film development and normal paper contrast, creating a normal tone-reproduction cycle. Quadrant 1 shows the subject values as they appear in the scene to be photographed, and how they are influenced by camera and lens flare into the film exposure values. These are projected into quadrant 2, and together with their developed negative transmission densities, they build the film characteristic curve. In quadrant 3, the negative values and the equivalent print reflection densities create the paper characteristic curve and the resulting print values. These are projected into quadrant 4, and when combined with the original subject values from quadrant 1, they build the objective tone-reproduction curve. For clarity, only the projection lines for the two endpoints of the pictorial range are shown throughout this tone-reproduction cycle. If film and paper had straight line characteristic curves, or in other words, if the densities of film and paper were to increase by a consistent amount for every consistent exposure increase, then the tone reproduction would be represented by a straight line in quadrant 4. This exact tone-reproduction line is shown as a reference and can be used to quantify the objective tone reproduction. As you will see in Fine-Tuning Print Exposure & Contrast, material characteristics and practical photographic experience demand optimized lighting conditions for satisfactory print viewing. Print illumination of around 1,000 lux (100 foot-candles) is about ideal. This viewing condition will require an objective tone-reproduction curve similar to the one in fig.6 to meet the standards of subjective excellence. This standard is the result of another study by Loyd A. Jones, C. N. Nelson and H. R. Condit, in which thousands of prints were made from more than a hundred outdoor scenes. They differed in exposure, contrast, density and tone-reproduction curve shape. The reference line was arbitrarily placed so that it intersected with the curve at the highlight point, because it cannot be below the minimum density of the paper. The preferred prints had a curve laying, in average, about 0.3 below

0.0

0.6

0.9

0.3

1.2

2.1

1.8

1.5

Negative Values

uc

tio

relative log transmission density

III

VI

IV

II

0 0

I II III Subject Values IV V VI VII VIII IX X

the reference line in density (lighter than exact tonal This results in a print with natural and convincing reproduction). The highlight gradients were very low smooth tones, just as our eyes perceive them. The (less contrast), but the midtone gradients were always same is not necessarily the case with digital imaging, high, 1.15 and above (more contrast), blending into since the digital recording of distinct gray levels is lower shadow gradients (less contrast again). limited to a finite amount. This can cause abrupt Consequently, a preferred first-choice print unsightly tonal changes, also called posterization sacrifices tonal separation in highlights and to some or banding, whenever low-bit recordings (8 bit) are extent in shadows, to magnify separation of the heavily manipulated. However, this is easily avoided midtones. Whenever the midtone gradient is below by sensibly manipulating only the high-bit recordings 1.10, the print will be judged as being dull or too flat. (10, 12 or 16 bit) of raw camera or scanner data files, In addition, the preferred print must be lighter than which is explored a bit further in the next chapter. the exact tone-reproduction curve, since a print with a curve density approaching the reference line was IX 0 judged as being too dark. This consistent failure of the VIII preferred print to match the exact tone-reproduction 10 VII curve is thought to be a consequence of normal huQ2 20 man eye functionality and its compensation for large Digital File luminance ranges in the natural environment. The 30 VI eye has a definite preference for fine midtone detail 40 and compensates for it with compressed highlights, if a wide scene brightness range should require it. We 50 will use this knowledge to prepare, present and display V 60 our photographs accordingly.
ge le
a

fig.7 This is an example of a digital tone-reproduction cycle for a digital camera and a calibrated print. Quadrant 1 and 4 are identical to the analog process, since analog and digital camera have similar flare characteristics (Q1), and the preference for a 'first-choice' print does not change with the process of image creation (Q4).

0.0 0 0 10 20 30 40

0.6 6

0.9 9

0.3 3

Digital Print

2 1.2

2.1 1

1.8 8

Q3

5 1.5

Digital Values

ca

dat

lib

ima

rat

era

grayscale [K%]

50

ed

sted

cam

pri

adju

nt

raw

60

70

70

Fig.7 shows an example of a digital tone-reproduction cycle for a digital camera and a calibrated print. Quadrants 1 and 4 are identical to the analog process, since analog and digital camera have similar flare characteristics (Q1), and the preference for a 'first-choice' print does not change with the process of image creation (Q4). However, Quadrants 2 and 3 differ from their analog counterparts, since their vertical scales have been replaced by a digital grayscale ranging from 0-100% (K%). By adjusting the raw camera data, the image file is manipulated until an aesthetically pleasing image is created. This typically happens as a combination of automatic camera adjustment and manual fine-tuning, using image software and a calibrated monitor. The result is stored as a digital file (Q2) and then printed on a calibrated printer to create a digital print (Q3), which satisfies the firstchoice print requirements. The resulting zone scales are indeed very similar to their analog cousins. For all practical purposes, negatives and paper, as used in analog photography, are continuous-tone materials. Even the smallest increase in exposure causes a slight density change in the light sensitive emulsions.

Digital Tone Reproduction

80 90 100 0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 2.7 3.0

IV III
II I 0

80 90 100

VII

III

VIII

VI

IV

II

IX

Exposure Values
log exposure values

Print Values
absolute log reection density

II

III

IV

VI

VII

VIII

IX

X 0 I

no

le

ns

II

re

Digital Camera & Lens

Q1

ex a

ct

to

Tone Reproduction

Tone Reproduction

ne

re pr od uc

e n tiv tio t jec uc rin ob prod ce p e hoi er c ton rst for

III

Q4

tio

r no m al le ns a re

Subject Values

IV V VI VII VIII IX X

log exposure values

119