Critical Thinking 2 - Normal Negative Density Range - ZS vs Tone Reproduction

[Stephen Benskin]

You are confusing negative density range with gradient. This is the fallacy of the zone system. The contrast (gradient) of the paper has nothing to do with the negative density range, which can vary quite a lot on a single roll of film, whereas the negatives' gradient remains unchanged in that roll. The paper contrast should be changed in cases of accidental over-development or underdevelopment (which affects the negatives' gradient), but not the density range of the negative. Assume you want to photograph a brick wall in even light. There will be a very narrow density range, right?

Let me start off by clarifying once again that The Zone System is a simplified version of Tone Reproduction and that I use Tone Reproduction. A good source on Tone Reproduction is Photographic Materials and Processes by Stroebel, Compton, Current, and Zakia (not Basic Photographic Materials and Process). It was, and may still be, a text book used at RIT. Another is The Theory of the Photographic Process, 3rd and 4th Editions. The 3rd edition tends to have more on black and white. It has chapters on The Interpretation of Sensitometric Results and The Theory of Tone Reproduction. The 4th edition combined General Sensitometry and The Interpretation of Sensitometric Results in to a single chapter Sensitometry of Black-and-White Materials. Tone Reproduction is a subchapter of Tone and Color Reproduction.

It's also good practice when making declarative statements to include support, like a reference. How am I confusing NDR with gradient? How is it a fallacy of the Zone System? What reference supports your statement that the gradient of the paper has nothing to do with the negative density range? Hitchens' razor states "what can be asserted without evidence can also be dismissed without evidence."

Slope or Gradient = Rise / Run

With the photographic negative, Run is Δlog-H, which represents the camera illuminance range of the subject luminance range. Rise is the difference in the negative density range between the two points of the run range. There are a number of methods used to determine a film curve gradient, Gamma, Contrast Index, and Ilford's G-bar. Each method measures different parts of the curve and can produce different results. Placing the name of the method in front of the resulting gradient value indicates the conditions of the method were used to determine the gradient.

The gradient for a "Normal" negative comes from the statistical average for the subject luminance range or Log Subject Luminance Range (LSLR) which is 2.22 or 7 1/3 stops, average flare factor, and the desired aim value for Rise.

The aim CI for Normal.

Desired Negative Density Range / (LSLR - Flare)
OR
1.05 / (2.2 - 0.4) = 0.58

To find the resulting NDR for a given CI is NDR = CI * (LSLR-Flare)

This was given to me by Dick Dickerson, who was the director of Kodak's R&D department and was responsible for T-Max films and Xtol. Bill uploaded a version earlier but in this version I've emphasized the statistical a average part.

This is from the processing tables on the Xtol datasheet. The aim CI for normal processing is 0.58.

So, knowing the Subject Luminance Range, average flare, and film gradient, the resulting negative density range (NDR) for that LSLR can be determined. You can also determine the aim film gradient using the aim NDR. This comes from the printing conditions. It doesn't need to only apply to silver paper but that is what I'll restrict it to. The NDR becomes the paper's log Exposure Range (LER). LER is determined using the following method as described in ISO 6846:1992 - Photography — Black-and-white continuous-tone papers — Determination of ISO speed and ISO range for printing.

The introduction states, "This International Standard establishes a method for measuring those photographic characteristics of papers used for printing from negative images which will aid users to select the appropriate products for their applications. ISO speed and ISO range are two measurements considered important for this purpose. Studies have shown that highly acceptable prints are generally obtained if the log exposure range (LER) of a photographic paper is equal to the effective density range of the negative.1) Therefore, ISO range, which is directly related to LER provides a useful criterion for grading papers." The standard's bibliography includes the paper, JONES, L.A. and NELSON, C.N. Control of Photographic Printing: Improvements in Technology and further Analysis of Results. Journal of the Optical Society of America, 38 (11 ): 1948.

The standard uses ISOR to depict the LER. It's simply the LER * 100. Here is an excerpt from Ilford's Multigrade FB Classic datasheet.

Part of the conclusion reached by the Jones and Nelson paper, "A careful study has been made in these Laboratories of the requirements which should be met by such a specification primarily from the standpoint of the user of these materials . The following conclusions have been reached:

1. The sensitometric exposure scale is the most suitable basis for the derivation of grade numbers because it bears a significant and useful relation to the density scales of the negatives which in practice give the best prints in each grade of paper.
2. The contrast of the paper is not satisfactory as a basis for deriving grade numbers. For papers differing in D-log E curve shape of in density scale, it gives equal ratings to papers which are unsuitable for printing the same negatives and unequal rating to papers which are suitable for printing the same negatives. The term contrast relates to the subjective appearance of prints made on the paper. While the sensitometric exposure scale is an important factor in determining the contrast of the paper, other essential factors are the density scale of the paper and the shape or gradient characteristics of the D-log E curve. If these other factors are included in the derivation of a grade number, the number will no longer be indicative of the density scale of the negatives which give the best prints, and the practical significance and utility of the grade numbers will be seriously impaired."

I've noticed you frequently refer to the 1.0 contrast aim in Negative Making for Professional Photographers. "It has been found through a series of comprehensive tests that for the great majority of scenes the middle tones should be reproduced at a gradient of 1.0 on a tone reproduction curve." The tone reproduction curve part isn't referring to the paper curve. The tone reproduction curve is a construct that compares the values of the original subject to the resulting reflection densities on the printed photograph. In a four quadrant reproductive diagram, it's found in the upper right in quadrant 4. The desired reproduction curve falls below a reference curve that represents a 1:1 reproduction and the mid-tone gradient should fall at 1.10 or higher.

 

[Stephen Benskin]

Something caught my eye while reading through the 1952 Kodak publication Negative Making for Professional Photographers. It could just be a coincidence, a bit of confirmation bias, a rationalization, or it could be serendipity. Even so, it's been a fun discovery.

There's a frequently used rule of thumb suggesting taking the manufacturer's published development time and reducing the time of development by 15-20%. Most, if not all, manufacturer's published development times for normal are based on printing with diffusion enlargers. Right off the bat it didn't make sense to apply the rule of thumb to negatives intended to be printed on diffusion enlargers as the times were a result of testing films printed with diffusion enlargers. I've long suspected that the rule of thumb came from people who either printed using a condenser enlarger and neglected to include that fact with the recommendation, printed with condenser enlargers and didn't know the published times were for diffusion enlargers, or they printed with a condenser enlarger and weren't aware there was a difference between the two types of enlargers.

What has always struck me as strange, though, was how consistently the values 15-20% are used. The rate of development can differ depending on the film / developer combination. A 15-20% change in development time for one film and developer can have very different results compared to another. The 15-20% range seems overly consistent when compared to the varying development rates from all the film / developer combinations. It almost appears to be an agree upon value but is that possible coming organically from thousands of individual observations. Maybe the range had a source.

When reading through the publication, I ran across this excerpt that confirms my suspicions. "Kodak processing recommendations for film are generally based on the use of diffusion-type enlargers, or on contact printing which results in prints of approximately the same contrast, everything else being equal. Obviously, these same processing recommendations should be modified by a reduction of 15 to 20 percent in gamma to suit condenser-type enlargers if prints of the same contrast are to be obtained." This publication probably isn't the original source of the recommendation. The rule of thumb does seem to have originated from Kodak and Kodak probably used it in many different publications over the years.

The 15-20% is a reduction in contrast and not time. 15-20% less contrast will produce consistent results no matter the film / developer combination. This makes sense.

The 15-20% reduction part appears to have survived but the original concept of applying it to contrast has been usurped and repurposed as time.

 

[Augustus Caesar]

Something caught my eye while reading through the 1952 Kodak publication Negative Making for Professional Photographers. It could just be a coincidence, a bit of confirmation bias, a rationalization, or it could be serendipity. Even so, it's been a fun discovery.

There's a frequently used rule of thumb suggesting taking the manufacturer's published development time and reducing the time of development by 15-20%. Most, if not all, manufacturer's published development times for normal are based on printing with diffusion enlargers. Right off the bat it didn't make sense to apply the rule of thumb to negatives intended to be printed on diffusion enlargers as the times were a result of testing films printed with diffusion enlargers. I've long suspected that the rule of thumb came from people who either printed using a condenser enlarger and neglected to include that fact with the recommendation, printed with condenser enlargers and didn't know the published times were for diffusion enlargers, or they printed with a condenser enlarger and weren't aware there was a difference between the two types of enlargers.

What has always struck me as strange, though, was how consistently the values 15-20% are used. The rate of development can differ depending on the film / developer combination. A 15-20% change in development time for one film and developer can have very different results compared to another. The 15-20% range seems overly consistent when compared to the varying development rates from all the film / developer combinations. It almost appears to be an agree upon value but is that possible coming organically from thousands of individual observations. Maybe the range had a source.

When reading through the publication, I ran across this excerpt that confirms my suspicions. "Kodak processing recommendations for film are generally based on the use of diffusion-type enlargers, or on contact printing which results in prints of approximately the same contrast, everything else being equal. Obviously, these same processing recommendations should be modified by a reduction of 15 to 20 percent in gamma to suit condenser-type enlargers if prints of the same contrast are to be obtained." This publication probably isn't the original source of the recommendation. It does seem to come from Kodak and Kodak probably used it in many different publications over the years.

The 15-20% is a reduction in contrast and not time. 15-20% less contrast will produce consistent results no matter the film / developer combination. This makes sense.

The 15-20% reduction part appears to have survived but the original concept of it applying to contrast has been usurped and repurposed as time.

Yes, I was aware that Kodak's times were for diffusion. I prefer to vary dilution rather than time. I work with these variables until I get at least 9 minutes for a proper negative.

 

[Stephen Benskin]

Yes, I was aware that Kodak's times were for diffusion. I prefer to vary dilution rather than time. I work with these variables until I get at least 9 minutes for a proper negative.

This is just sharing a discovery. A fun fact. Totally separate topic from the one we were discussing and it's not intended to be directed toward or about anyone. I thought about creating a new thread. My apologies for not being clear.

 

[Keith Tapscott.]

The 15-20% is a reduction in contrast and not time. 15-20% less contrast will produce consistent results no matter the film / developer combination. This makes sense.

The 15-20% reduction part appears to have survived but the original concept of applying it to contrast has been usurped and repurposed as time.

Thanks for pointing this out. I have often read about the reduction of development times without any mention of the light source of the enlarger type used.

Ilford used to publish times for a G-bar value of 0.55 for condenser enlargers and a G-bar of 0.70 for cold cathode enlargers. These days they publish a G-bar value for 0.62 .

If I remember correctly, Kodak gave times for a contrast index of 0.56 (0.58?)
for diffuser enlargers and 0.45 for condenser enlargers. These are of course starting points and can be adjusted for personal preference.

 

[Stephen Benskin]

Thanks for pointing this out. I have often read about the reduction of development times without any mention of the light source of the enlarger type used.

Ilford used to publish times for a G-bar value of 0.55 for condenser enlargers and a G-bar of 0.70 for cold cathode enlargers. These days they publish a G-bar value for 0.62 .

If I remember correctly, Kodak gave times for a contrast index of 0.56 (0.58?)
for diffuser enlargers and 0.45 for condenser enlargers. These are of course starting points and can be adjusted for personal preference.

I think the difference between 0.56 and 0.58 has to do with flare. 0.56 was used up until the late 80s and I'm assuming it changed because of the greater use of 35mm which has a higher flare value because 35mm tends to have a higher number of lens elements. The difference is less than 1/3 stop though. I've noticed something strange with the updated Xtol datasheet, Kodak film is back to 0.56 but non-Kodak film is still 0.58. The only way this can happen that I can think of is Kodak didn't want to spend the money to retest films from other manufacturers.

 

[Bill Burk]

In this case TMY2 going 15% down in contrast (CI) from 0.62 to 0.53 takes you from 13 minutes to 11 minutes

Or 15%

http://beefalobill.com/imgs/tmy2timeci.jpg

 

[john_s]

....... The only way this can happen that I can think of is Kodak didn't want to spend the money to retest films from other manufacturers.

They could still have guessed a most likely adjustment for those films. Maybe they just forgot about it.

On a slightly related topic, Agfa used to set standard contrast at 0.65 which many of us found too contrasty. I remember Lloyd Erlick writing years ago that Agfa films would have been more popular if their recommended development times had been reduced (and maybe film speed reduced also, which might have made the films look not so good against the competition.

 

[Stephen Benskin]

They could still have guessed a most likely adjustment for those films. Maybe they just forgot about it.

On a slightly related topic, Agfa used to set standard contrast at 0.65 which many of us found too contrasty. I remember Lloyd Erlick writing years ago that Agfa films would have been more popular if their recommended development times had been reduced (and maybe film speed reduced also, which might have made the films look not so good against the competition.

It's sad but not surprising as Kodak probably lost most of their best people when things went south. What would Mees and Jones think?

If I correctly recall, Agfa uses or used a modified form of Gamma. A Gamma of 0.65 - 0.70 is around a normal range equivalent to a CI of around 0.56 - 0.60. The idea of a modified Gamma throws a bit of a wrench into the works as the actual testing method isn't know and can't be evaluated. The method of testing Gamma is different enough from Contrast Index or G-bar that direct comparisons may not be reliable. The paper on Contrast Index compares the two methods. Ironically, the creation of Contrast Index and the publishing of the paper was Kodak's attempt to standardize contrast determination. It certain would have made comparisons easier and reduced confusion.

From Agfa Black-and-White Chemicals Film processing datasheet ---- Agfa APX 100 from Agfa Range of Films datasheet

If Agfa uses the ISO standard, they are adhering to the parameters of the ISO standard. It's apples to apples then.

From C.J. Niederpruem, C.N. Nelson, and J.A.C. Yule, Contrast Index, Photographic Science and Engineering, Vol 10, Number 1, Jan-Feb 1966. See the link for the paper above.

 

[Bill Burk]

But nobody’s published development times are the same as the lab development time to meet parameters. And Agfa’s still using “Gamma” instead of “Kodak’s” CI, right?

Sometimes the development time is associated to an expected contrast. But there are a lot of development times that don’t specify contrast. And that gets my goat. I almost wanted to create a new site to compete with massive dev chart. But I don’t want to put in all the effort. How hard would it have been for people to have learned to evaluate and share contrast?