In every workshop I teach, I make it a point to ask each student about their experiences in other workshops. I do this to improve my own workshops by learning from the successes and mistakes of others. One of the most common complaints I hear from workshop students concerning approaches to the Zone System they have learned elsewhere is that once they have completed their Zone System calibration efforts and begun shooting negatives in earnest, they have been disappointed with their results. “My negatives still weren’t any better,” is a refrain I hear quite often.

Though this singular complaint usually has a multitude of underlying causes, ONE cause stands out above all the others — The method taught for determining film speed.

Disclaimer

Before I go any further with this, I ought to clarify something I feel is very important:

The fact that I am writing an article about film speed determination may lead some readers to mistakenly assume that I am lending support to a widely held misconception about the Zone System — this misconception being that the Zone System is primarily concerned with finding “your film speed” and “your development time,” as is often implied in so many Zone System books and articles. This is not the case.

The Zone System is about informed visualization^{footnote 1} and the tools necessary to carry that visualization to completion. Determining your film speed and development time (also known as calibration), though certainly important, is such a minor pursuit when compared to the whole of the Zone System, that it should be downplayed as much as humanly possible. In fact, compared to the totality of the Zone System, calibration becomes almost (not quite) insignificant. Within limits, one could indeed apply the Zone System very effectively without ever calibrating at all!^{footnote 2}

Having put calibration into proper perspective, let’s get back to the subject of film speed.

Approaches taken to determining film speed…

There are a wide variety of methods for determining film speed for Zone System purposes. Some involve using stepwedges, others shooting pictures of gray cards, still some others recommend you take photographs of the side of a building. Most of the current Zone System calibration methods however have one thing in common — they recommend basing film speed on a fixed Zone I density. Most often, the Zone I density sought is .10 over base + fog.

On the surface, this approach seems logical. After all, the generally accepted rule is that .10 over b+f is the first consistently printable negative density. That is, it is the first negative density that can be visually distinguished in a print as being somewhat lighter than the black produced from a Zone 0 negative density. And in fact this is true.^{footnote 3} A Zone I density lower than .10 over b+f is unlikely to print as a distinguishably different tone from that of a Zone 0 negative density, if that Zone 0 negative density is indeed printed as a Zone 0, black print tone.^{footnote 4} However, it is not logical to jump from this observation to the conclusion that .10 over b+f is a valid reference point for testing film speed.

Despite the fact that it is more or less reasonable to define a Zone I negative density as .10 over b+f, it is not reasonable to base our film speed determinations on this. Only under certain circumstances, such as when the subject is classed as Normal and will therefore receive Normal development, is this approach valid. Even then, it is only valid occasionally, and not for all Normal subjects. To base film speed on a fixed Zone I density is to completely ignore what is happening in Zones 0, II, III and IV, which is to court disaster.

Huh?

OK, perhaps a little elaboration and clarification are in order.

The idea behind the traditional way we approach film speed calibration (.10 over b+f) is to base our speeds on the minimum densities that will just render the very darkest tones, Zone 0 and I, properly in a print. This is all well and good, except for the fact that, paradoxically…

In the field we generally base our exposures on Zone III, sometimes on IV, seldom on Zone II and virtually never on Zone I! (There’s nothing wrong or misguided with basing exposure on any Zone you desire. But although basing an exposure on Zone I, may at times be rational, it is very uncommon practice indeed.)

The logical flaw involved here should now be apparent. We base our film speed tests on Zone I, but then go into the field and base most of our exposures on Zone III. Then when we make a print, we look to Zone III shadows to determine whether or not exposure was adequate. Never, ever do we look at Zone I shadows for this purpose! It makes no sense to base all of our film speed tests on one Zone (I) and all of our actual exposures on another (III).

If we base our calibration tests on Zone I, then we should also base our exposures on Zone I in the field. Unfortunately, this is impractical for a number of reasons, not the least of which being that Zone I areas of the subject are usually quite small and difficult to meter adequately. In addition, basing exposures on Zone I could very often lead to inadequate Zone II, III and IV detail because — if we base exposure on a Zone I tone, another, more important tone we wish to have on Zone III, may not fall there.

The options are either to continue with the clearly irrational approach of basing tests on Zone I and exposures on Zone III, or to alter our approach to film speed testing and begin basing our speed tests on the same Zone on which we usually base exposures — Zone III! This is the approach I will recommend here. But before we get too far ahead of ourselves, lets take a look at some other reasons why basing film speed on Zone I is illogical…

As you can see from the varying densities for Zone III in each curve, basing film speed on Zone I results in a significant difference in Zone III density and contrast for the three films — and therefore in the appearance of the shadows in the eventual final print. Each negative will produce shadow density and contrast significantly different from the others. Obviously, we are likely to prefer one over the other two, and therefore the .10 over b+f speed determination would provide less than optimal results with at least two of the three films.

In Figure 2, I have shifted the curves from Figure 1 so that all three intersect at the same Zone III density. This is, after all, the desired result if we base negative exposure on Zone III — to have a Zone III exposure consistently produce the same density in each negative.^{footnote 5} As you can plainly see, in order for the three films to produce the same consistent density for Zone III, their speeds at Zone I are no longer the same. One becomes in effect, a slower film, another faster. Therefore, if we desire a consistent density for the Zone on which we actually base our exposures (Zone III), then the .10 over b+f concept for film speed is obviously useless.

Figure 3 illustrates a lower CI curve such as is produced with contraction negatives. With contractions, an increase in exposure is of course required. But if you provide only enough increase in exposure to maintain Zone I at .10 over b+f, the result is a Zone III density that is much lower than may be logically desired for Zone III. Instead, even more exposure is needed to move a Zone III subject tone up to a negative density that is adequate for printing. Obviously, a film speed based on a Zone I density of .10 over b+f is not applicable to contraction negatives.

Figure 4, a higher CI curve such as might be produced by expansion, shows that with expansions, reducing exposure only enough to maintain Zone I at .10 over b+f (the traditional recommendation) results in a Zone III density that may be too high, providing shadows that are perhaps too contrasty. A still greater decrease in exposure may be required here to bring Zone III in line. You may think this would make Zone I unprintable, since its density might then be lower than .10 over b+f. However, due to the increased contrast resulting from expansion in this part of the characteristic curve, it is easier to separate Zones 0 and I in a print and therefore, a Zone I density lower than .10 over b+f can often be effectively printed from expansion negatives. From this it is plain that a film speed based on a Zone I density of .10 over b+f doesn’t work for expansions either.

“Correct” film speeds are often wrong! Or…

Reasons why an “average” optimal film speed is best.

In a later section, I’m going to tell you how to successfully test for an “average” optimal Zone III film speed. But first, you may wish to know — Why do I refer to an “average” optimal speed, and not just an “optimal” speed?...

Because the optimal speed of a film varies, depending on the subject!

(Did I hear someone out there mutter, “Get a rope?”) Yes, you did understand correctly — optimal exposure is as much a function of the subject being photographed as it is of the chosen film and development scheme. Because of this, there is no optimal speed that fits every Normal subject. Neither is there an optimal speed for every N+1 subject, or for any other N group. Optimal film speed can be, and usually is, different from one N subject to the next, or one N+2 subject to the next, etc. There are five basic ways in which optimal film speed for Zone III shadows^{footnote 6} can be shifted by the subject itself.

1. Local contrast in the subject’s shadows can be too low, making them appear somewhat muddy in the print, as though they had originally been underexposed in the negative. (See "The Primacy of Local Contrast")
2. Local contrast in the shadows may be too high. This is a less common problem than above. Shadows with high local contrast can sometimes appear too harsh, as though overexposure or overdevelopment of the negative had occurred.
3. The reflectance range of the subject’s shadows can be too short. (There may not be any subject tones falling on Zones 0 and I, after you have placed an important tone on Zone III.) This produces an effect visually similar to low-local contrast (see above), and may look in a print as though the negative had been underexposed.
4. The reflectance range of the subject’s shadows can be too long (Once you have placed an important subject reflectance on Zone III, tones you would prefer on Zones I and II may fall lower on the Zone scale.), resulting in prints that look as though the negative may have been underexposed and overdeveloped, or has high local contrast.
5. Finally, optimal film speed is affected by the color of light illuminating the shadows of the subject, and by the color of the subject itself. Color affects both film speed and contrast. (There is enough on this topic for an entire article of its own, but for our purposes here, it is enough to know that it happens.)

Numbers 3 and 4 above require a little explanation…

One of the reasons for the existence of the Zone System is the observation that a conventional, non-Zone System exposure, based on an average reflectance reading of the overall scene, can leave shadows and highlights with too little or too much density. An average reading of a subject with a too-long reflectance range for example, can leave important shadow tones exposed too low on the Zone scale (and highlight tones exposed too high).

In Zone System practice there is a similar price to be paid for basing exposure on Zone III. When we place a subject reflectance on Zone III, we are often making the same mistake mentioned above, on a smaller scale. Very seldom do we take the time to measure subject reflectances lower than the tone which we have placed on Zone III, to see where they fall.^{footnote 7}

With some subjects, there may not be any tones that fall lower than Zone II or I. This produces short-reflectance-range shadows, which may appear flat and muddy in a print. With other scenes, there may be one or two important subject reflectances that fall lower on the Zone scale than we might like. This results in long-reflectance-range shadows, which may appear contrasty in a print. The subject of short and long reflectance range shadows is another topic that could occupy an entire article of its own, but again, it is enough at this point that we know it happens.

There are quite a number of other factors, not related to the subject itself, that can also alter your optimal film speed from one photograph to the next… such as the contrast of your lens, flare, filters and any number of other things. The point of this section is primarily to illustrate just how much of a moving target optimal exposure is, and how important the need is for a film speed determination method that gives best overall results for the way you work and the kind of subject matter you photograph, in light of all these difficult-to-control variables.

What are we really after?

Many of us who teach and write about the Zone System have gotten overly enamored of our densitometers. We have made the mistaken assumption that because we can measure densities and plot curves (things I encourage students not to waste time on), a carved-in-stone density for Zone I (or II or III) can be declared. This is at best an unfortunate illusion (or should that be delusion). Whether or not the density for any Zone is adequate can be judged only in light of the relationship between it and the densities of adjacent Zones, and more importantly, how it actually looks when printed!

We should never be interested in establishing a set density for a given Zone. That is counterproductive. Instead we must ask ourselves, “As photographers, what do we really want?” Do we want Zone I exposures to produce a density of .10 over b+f, or Zone III exposures a density of .35 over b+f? Or, do we really just want negatives that produce a specific level of tone and contrast in the shadows of our prints? Obviously, we want the latter. (Remember, we’re talking about exposure only. Tone and contrast for higher Zones can wait for another article). We don’t really care what the density for a Zone III shadow is, as long as it prints the way we want it to. This is all that genuinely matters.

By now you should be in the market for a new approach to speed testing. In looking for a new method, you might have been tempted to retain some of the “old school” approach by simply declaring a “standard” density for Zone III, such as .35 over b+f, and testing for that rather than for a standard Zone I density of .10 over b+f. But, as you learned in the previous section, just because we’ve achieved a predetermined density, doesn’t mean it will look good in a print. Whether or not a Zone III density is “correct” hinges to a great extent on the relative densities of Zones 0, I, II and IV. If the densities of Zones adjacent to III, (II and IV) are too similar, such as is the case with most contractions, low-local-contrast shadows and short-reflectance-range shadows—shadows will be flat in the print. If they are too divergent, as with some expansions, high-local-contrast shadows and shadows with a too-long reflectance range—shadows may be too contrasty in a print.

In the final analysis, the only true test of film speed is not a densitometer, a gray card or a stepwedge…it is the answer to this question: “How do the shadows look in a print?” If they look good, then your film speed is correct. If they do not, then no matter how precisely determined, your speed is wrong!

Summary

Despite all the technical gibberish and pseudo-science that occurs in photography in general and in the name of the Zone System in particular, just about everyone can agree on two things…

• The only thing that really matters in photography is the print.
• The final true test of the effectiveness of any film speed determination method is the appearance of shadows in the print.

  In addition to the above two general points of agreement, I feel that in this article I’ve solidly established the following two truisms…

• Film speeds ought to be based on the same Zone on which we usually base exposure — Zone III. Not on Zone I.
• The only logical way to test for film speed, for Zone System purposes, is to base speed on our judgement of how well Zone III shadows look in a print, not on a specific negative density.

How to do it.

For the sake of brevity, what follows is an abridged description of only the film speed testing half of the calibration process I normally teach to students. In actual practice, I teach students how to test for development time and film speed, simultaneously, and with the same negatives. If you are interested in learning about the whole process, see "Zone System Calibration."

Since the only true test of film speed is the appearance of shadows in the print, obviously we must make real prints in order to test for film speed.

And because optimal film speed varies from one subject to the next, we must make prints from real negatives, of real subjects, in order to see which film speed gives us the best looking shadows, most frequently. Stepwedges and artificial test targets, regardless of type, simply won’t do.

As with any other testing method, speeds for different N numbers (N-2, N+1, etc.) must be tested separately. You will (obviously) get a different optimal speed for N+1 subjects than you do for N subjects, etc.^{footnote 8} Be sure to test for only one N # at a time.

The goal with this approach to testing is to find an average film speed for each class of subject (N+, N-, etc.) that gives optimal results for the majority of subjects in that class. That is, for subjects you categorize as Normal, you will have one optimal, average film speed. For each N+1, N-2, etc. class of subject, you will have a distinct optimal, average film speed.

It is important to keep in mind that this approach to calibration is designed not to interfere with, or interrupt your normal work flow. Simply follow your usual working habits, shooting only real subjects you would have photographed anyway — the only difference being that you will shoot some extra negatives for testing purposes.

If you are testing film speed for N subjects, then for each N subject for which you base exposure on Zone III^{footnote 9} shoot a bracketed series of negatives one and a half to two stops, in half stop steps, on either side of your starting film speed. For this approach to be effective, you will need at least a dozen sets of bracketed negatives from as many different scenes, so that you can average your results later. The reason you need to average results for a number of different scenes is, as previously mentioned, because optimal film speed is going to vary from one subject to the next, and because you want a large enough sampling so that errors with individual subjects do not have a significant influence or your results. If this sounds a little tedious, remember that you should not be shooting test negatives, but rather, real negatives of real subjects you would have photographed anyway.

If you are testing an ISO 400 film, then as discussed earlier (see Footnote #2), you should start with a presumed EI of 200 and therefore bracket shots at EI’s of: 75, 100, 150, 200, 300, 400 and 600.^{footnote 10} This puts your most likely film speed in the middle of the bracketed range. If you are uncertain, add another bracketed shot at either end, at EI’s of 50 and 800. (Experienced Zone System photographers can judge film speed quite effectively just by looking at bracketed sets of negatives, thereby bypassing many of the steps here. For the beginning and intermediate Zone System photographer however, this method is very effective and at the same time is a tremendous learning experience.)

Make a small print from each of your bracketed negatives on grade 2 paper.^{footnote 11} (The reason for grade 2 paper is simply to have a consistent set of results and to target negatives at a medium contrast paper — not because there is anything sacred about grade 2 paper.)

Print each negative in the bracketed set for best shadow detail, ignoring the appearance of highlights. Start with the least exposed negative and work your way up to the most exposed. Try to approximately match the density of the shadow areas in each print and remember that you are trying to get a Zone III tone. The first time or two that you do this, you may have to go back and reprint one or two of the negatives so that all of the Zone III prints approximately match in tonality.

When all the negatives from a bracketed series have been printed, look at the prints in order, from the print made from the least exposed negative to that made from the most exposed negative. You should see a steady, yet usually subtle improvement in Zone III image quality from print to print, until you reach the print made from the negative that received optimal exposure.

Once you have printed the negative that provides optimum exposure, the remaining negatives in your bracketed set, those having had even more exposure, will usually not provide prints with shadow detail that is significantly any better. That is, if the best speed for your subject with an ISO 400 film is EI 200, then the EI 300, 400 and 600 negatives will not produce shadows that look as good. But the EI 150, 100 and 75 negatives should all produce shadows that look largely just as good as the EI 200 negative. The idea is to find the fastest speed that will produce shadows that are not significantly improved upon by additional exposure. Make a note of the optimal speed for this particular subject.

Now go on to the next Normal scene for which you have a bracketed series of negatives and print those the same way. Don’t be surprised that you get different optimal film speeds for each Normal subject. This is to be expected and is the reason a number of subjects must be tested and their results averaged, before any meaningful information is derived.

Here are some of the things you will notice that may surprise you:

• The differences in shadow image quality from one print in a series to the next are often very subtle. It will frequently be difficult to decide at which point a “best” negative has been reached. This further illustrates the futility of proclaiming one universally applicable film speed method, based on a specific density. When in doubt, always opt for the fastest of the negatives you are considering. If it is just impossible to tell, guess. This approach to calibration takes that uncertainty into consideration when you average the results from a number of different scenes.
  
  A useful aid in judging these prints is to compare the print made from the least exposed negative to that made from the most exposed negative, then progressively compare each of the other prints in the series in turn, to the print made from the most exposed negative.
• After you have printed several bracketed sets of negatives from different Normal scenes, you will notice that with some subjects it was much easier to decide which bracketed negative received optimal exposure. This is because some of your subjects will have had higher local contrast in the shadows than others, causing more noticeable changes in shadow quality from one print to the next.
• With some sets of bracketed negatives, print image quality seems to continue to improve right through the entire bracketed series. You probably did not miscalculate exposure. Because continuing increases in exposure steadily move the shadows of your subject higher up the characteristic curve of the negative, contrast will continue to increase in the shadows and therefore continue to improve overall shadow detail and contrast throughout your bracketed set. However, in most cases there will be a point where the improvement gained between negatives becomes smaller. You will have to judge for yourself where improvement starts to fall off and select the appropriate negative. Do not let these kinds of confusing results discourage you. You will get such results with some subjects. This only serves to prove one of the main points of this article — that correct exposure is very much a moving target and it is therefore impossible to declare any static “givens” about film speed.

When you have printed the bracketed negatives from a number of different Normal (or N+ or N-) subjects and are ready to determine your average, optimal film speed for that N#, disregard one each, of the highest and lowest results, then average the rest. If necessary, round the result to the next slowest third or half stop speed, as you prefer. For example:

Number of Subjects	Optimum Speed Noted
1 set	EI 400 (disregard because it's the highest)
2 sets	EI 300
6 sets	EI 200
2 sets	EI 150
1 set	EI 100 (disregard because it's the lowest)
2100÷10=210 — round to 200 for your average, optimal speed.footnote 12

You now have an average film speed for N, or whatever N# for which you were testing, which you know to be the optimal average speed for the kind of subject matter you photograph and for the way you work, because you arrived at this conclusion by judging the appearance of shadows in real prints, from real negatives of real subjects.

Now you can move on to the next N# class of subject you wish to test for an optimal, average film speed. Do not however, waste time testing for contractions greater than N-2 or for expansions greater than N+2. It is impossible to calibrate meaningfully for expansions and contractions beyond this range. Why? Well…

That’s the subject of another article.

David Kachel

1. Informed visualization is a term of my own invention. Even the beginning snapshooter with no knowledge of photography at all, visualizes the final photograph when taking a picture. The difference between the snapshooter and the professional is that the snapshooter’s visualization usually has little to do with the reality of what is in front of the camera and what is physically possible. The Informed Visualization of the skilled Zone System photographer on the other hand, is visualization based on a thorough understanding of one’s tools and materials, and of what is or is not possible to produce.


2. I strongly encourage readers not to let anyone make a mystery of, or create a mantra from “your film speed,” or “your development time.” It is hardly a secret that the vast majority of expert Zone System photographers end up using a film speed that is approximately half the manufacturer’s recommended speed. If you start testing at this point (EI 50 for ISO 100 films, EI 200 for ISO 400 films, etc.) you will already be very close to your most likely results. Likewise, most Zone System photographers end up using a development time that is roughly 70% (for non-T-grain films) to 80% (for T-grain films) the manufacturer’s recommended development time. Starting with such a time will put you very much in the ballpark at the outset.


3. Densities smaller than .10 can be measured and are easy to distinguish visually in a negative. However, just because you can see a density in a negative, doesn’t mean you can print it. For a Zone I density to be printable, it must produce a black distinguishable from that black which is produced by an area of the negative representing one full stop less exposure (ie., Zone 0). Densities lower than .10 over b+f can be seen and measured, but not easily (or at all) printed.


4. It is important to keep in mind that the negative density for Zone 0 is not – 0.00 over b+f, but rather something marginally less than whatever the density may be for Zone I. eg., if the density for Zone I is .10 over b+f, then the actual density for Zone 0 is likely to be something on the order of .06 or .07 over b+f. Likewise, a Zone 00 density will be roughly .03 to .04 over b+f. There is in fact an entire section of characteristic curve beneath Zone I that can provide a lot of very useful information, not the least significant of which is the fact that correct Zone I density depends on the relationship between the densities of Zones 0, I & II and the contrast in that part of the characteristic curve. For some negatives, a Zone I density of .10 over b+f may print distinguishably from Zone 0, for others it may not. In still some others, a Zone I density lower than .10 over b+f may yet be printable.


5. Over the years I’ve learned never to underestimate my readers. Some of you have no doubt already spotted the minor flaw in this logic, and you are correct — it is not sufficient to just have the same density. Other factors which will be covered further on in this article, also come into play.


6. I define shadows in this context as, whatever tones you are exposing on Zones 0 through IV, regardless of whether or not they are actually in “shade.”


7. The logical conclusion from this would of course be to base exposures on Zone I. But, as mentioned previously, this is impractical for a variety of reasons.


8. Zone System photographers have always gotten different speeds for different N numbers. But we have always based the new speeds on the same Zone I, .10 over b+f assumption. Now we are basing altered film speed on the appearance of Zone III shadow detail and contrast in the print. You will find significant differences in the speeds you obtain for expansions and contractions by this method as compared to the old .10 over b+f approach. For example, using one of my SLIMT bleaches with T-Max 400 and processing to N-2, I achieved a speed loss of zero when I based speed on a Zone I density of .10 over b+f. Using the same film and processing method, but basing my EI only on the appearance of Zone III shadows in the print, I instead got a speed loss between one-half and one stop.


9. For purposes of these tests, do not include subjects for which you based exposure on some other Zone… photograph them of course, simply don’t include them in your tests.


10. When testing for expansions, use a starting speed a half-stop faster. For contractions, use a starting speed a full stop slower.


11. When testing speeds for contraction negatives, test on grade 3 paper, rather than grade 2. For reasons too involved to discuss here, contraction negatives should all be aimed at grade 3 papers to compensate for the diminished local contrast produced by contraction.


12. Strictly speaking, this approach is mathematically flawed. Obviously, the correct average film speed for the example shown is 200, not 210! However, the inherent mathematical error is too tiny, in all cases, to be meaningful and therefore it makes more sense to use this simplified approach, rather than a more “purist” math.