What ever method you decide on, the key is to think it through and understand the purpose, concepts, and variables.

Take a look at Ralph's Nomograph. Here are three questions to ask yourself.

Why is the aim negative density range for a "typical diffusion enlarger" 1.20 when the ISO standard has it as 1.05?

If you draw a line from paper CI 0.58 through 7 SBR to 0.58 adjusted aveGradient SBR (middle scale), through Normal Camera Lens Flare and you get 0.58 for the approximate final avgGradient. How is it possible to have the same CI from beginning to end when normal flare is at least a stop?

How can 0.58 be the aim average CI when flare is zero, yet CI 0.58 is also normal using a 1 1/3 stop flare factor: 1.05 / 2.2 - .40 = 0.58?

Last edited by Stephen Benskin; 07-16-2011 at 11:07 AM. Click to view previous post history.

The attached example might help to explain the difference between sensitometric exposure and photographic exposure and hopefully convey the importance of contacting the step tablet in order to produce a characteristic film curve which excludes many variables such as flare.

I've uploaded two more examples to help illustrate interpreting photographic exposure using sensitometric exposure. The first example shows how the discrepancy between the Zone System's aim negative density range and the ISO's aim negative density range is less a difference in system parameters and more about how the curve is interpreted. The Zone System doesn't factor in flare in the interpretation and instead uses a higher aim NDR. The ISOR method factors in flare, effectively reducing the illuminance range from the subject and consequently producing a smaller aim negative density range. Both methods will produce a negative with the same contrast index, but only one represents "reality" and only one works in relation with the ISO paper testing method. But I believe this is an excellent example about importance between sensitometric exposure and photographic exposure and understanding the difference when interpreting the data.

The other example shows the theoretical matching of the negative to the paper from a various subject luminance ranges. It's theoretical because it uses a fixed flare model yet flare tends to increase with higher luminance ranges and decrease with lower luminance ranges. However, it does illustrate the application of photographic exposure with sensitometric data (exposure). Determining the appropriate curve CI to luminance range that matched the paper LER was done using the equation CI = Negative Density Range / (Subject Luminance Range - Flare).

Hi Stephen, sorry for the late reply. Some new film arrived today so I'm ready to repeat my testing and I now have time to think about this again.

Originally Posted by Stephen Benskin

Take a look at Ralph's Nomograph. Here are three questions to ask yourself.
Why is the aim negative density range for a "typical diffusion enlarger" 1.20 when the ISO standard has it as 1.05?

Actually when I look at his nomograph and read page 1 of the pdf Ralph gave us in this thread, the ISO target density range is 1.30, not 1.20. He goes on to explain why we don't target that. One reason is that "In general, advertised ISO film speeds are too optimistic and suggested development times are too long." also "Not knowing the exact combination of products we use for our photographic intent, they (the film manufacturers) have had to make a few assumptions."

Originally Posted by Stephen Benskin

If you draw a line from paper CI 0.58 through 7 SBR to 0.58 adjusted aveGradient SBR (middle scale), through Normal Camera Lens Flare and you get 0.58 for the approximate final avgGradient. How is it possible to have the same CI from beginning to end when normal flare is at least a stop?
How can 0.58 be the aim average CI when flare is zero, yet CI 0.58 is also normal using a 1 1/3 stop flare factor: 1.05 / 2.2 - .40 = 0.58?

To start with this nomograph something Kodak created and Ralph reproduced, so I'd prefer both of them justified it before I did.
It is possible the nomograph is in error since as you point out, 'normal' flare results in no adjustment in the CI. Alternatively there may be something 1/we don't understand about using it. I'd need Ralph to chime in here, or to see the original Kodak publication with any accompanying text.

As for you calculating 0.58 as normal when using a 1 1/3 stop flare factor and a SBR or 7 stops, you have used a target CI of 1.05 and I don't know where you got that from. If instead we assume a target CI of 1.30 as I mention above, and convert the 7 stops to the correct units by multiplying by 0.3 to then the adjusted CI =(1.3/(2.1-0.4))=0.76, not 0.58.

You posted 2 other replies, coincidentally one only 1 hour ago which I haven't had time to digest, so I will have to reply to those at a later time.

I never said anything was incorrect (although the speed part is). There are always different ways to approach a problem. I just wanted to emphasis the importance of understanding the variables and how they work together. I have to double check the Kodak version, but Ralph's works fine as long as you accept that Normal has already factored in flare and accept the parameters for which Ralph defines Normal.

I don't have time at the moment to go through you response, but I should be able to take a pass at it this evening.

Actually when I look at his nomograph and read page 1 of the pdf Ralph gave us in this thread, the ISO target density range is 1.30, not 1.20. He goes on to explain why we don't target that. One reason is that "In general, advertised ISO film speeds are too optimistic and suggested development times are too long." also "Not knowing the exact combination of products we use for our photographic intent, they (the film manufacturers) have had to make a few assumptions."

Peter,

Okay, first we have to make sure we are referencing the same material. I believe you are using the pdf file from the Beyond Monochrome's website entitled "Testing Film Speed and Development." If you are then you also must be referring to fig 1 on page one of the document. And if that is the case, you are mistaking the log-H range for negative density range. Your use of the term "nomograph" if referring to figure 1 also confused me because I didn't think the term applies to that type of graph. It may, but in my earlier response, I was referencing the fig on page 140 of the first edition of Ralph's book. So I hope we are now talking about the same thing. BTW, figure 1 is from the ISO standard for determining black and white negative film speeds: ISO 6 - 1993.

The figure 1 defines the contrast parameters under the ISO standard for which the film has to be developed before film speed can be determined. The reason for this isn't to reflect "normal" processing conditions but because under this set of conditions there's an agreement between the fractional gradient method of film speed determination with a fixed density method. I've attached a rather technical paper that explains it all. I've also attached something I've written that isn't as technical. Another paper, "Simple Methods for Approximating the Fractional Gradient Speed of Photographic Materials" is too big to upload, but I am willing to email it to anybody interested.

Let's take a look at the quote, "In general, advertised ISO film speeds are too optimistic and suggested development times are too long." Why are the ISO film speeds too optimistic? Based on what? What quality standard is the ISO speed being compared to? Ralph mentions the Zone System. The problem is the testing parameters and assumptions between the two methods are different, so it's like comparing apples to oranges. The ISO film speeds might be considered optimistic compare to the Zone System speeds, but to conclude the ISO speeds are wrong, you'd have assume that in some way the Zone System method produces more accurate results. In reality, there is only one method to determine film speeds and that is the ISO method (see the safety factors paper). All else is more about preferred exposure. You can call it EI if you want.

Part of the purpose of film speed is to define the exposure boundaries that will produce a quality image. In determining color reversal film speed, the high and low points are defined first. The speed point is then the mean of the two. Why? Because with transparencies, quality is determined by how the middle tones are reproduced. With black and white negative film, it's the shadows that are critical, so the minimum gradient (not density) needs to be found. Once this point is found, any exposure increase, within limits, over this point will produce a quality print. This point is known as the fractional gradient point and is found 0.29 log-H units below the 0.10 fixed density point when using the ISO contrast parameters.

What about the second part of the quote? Are the suggested development times too long? They are if you assume the film speed is determined simply using a fixed density method of 0.10 over Fb+f, except it's not. There is a hidden equation behind the contrast parameters of the ISO standard. Only under these conditions is the 0.28 log-H relationship between 0.10 and the fractional gradient point certain. And if you are assuming the speed point of 0.10 also necessarily defines the point where the shadow exposure is supposed to fall, you'd be wrong. The speed point may or may not define where the shadow exposure is supposed to fall, but that's not it's main purpose. It's purpose is to calculate a film speed. According to the scientific papers, if the contrast of the film is less than or greater than the the ISO contrast parameters, the use of a fixed density method isn't recommended as it will give inaccurate results. A different method must be used.

As for you calculating 0.58 as normal when using a 1 1/3 stop flare factor and a SBR or 7 stops, you have used a target CI of 1.05 and I don't know where you got that from. If instead we assume a target CI of 1.30 as I mention above, and convert the 7 stops to the correct units by multiplying by 0.3 to then the adjusted CI =(1.3/(2.1-0.4))=0.76, not 0.58.

The negative is the intermediary step between the subject and the print. It's purpose is to take the luminance range of the subject and produce it in a reasonably accurate way on the print. In order to determine the necessary contrast of the negative required to achieve this goal, you need to know the log exposure range of the paper (LER) and the luminance range of the subject (LSLR). The LER of a paper is determine from the density points 0.04 and 90% of the paper's D-max. The mean LER associated with a grade 2 paper is 1.05.

The LSLR used for a "normal" negative is based on the statistically average scene which is 7 1/3 stops or 2.20 logs. While this may not seem like a big difference from the often used 7 stops, it can make a difference with making sense of the calculations: 1.05 / (2.2 - .4) = 0.58. Anybody remember when Kodak used CI 0.56 as normal? That was because coated large format lenses have slightly less flare than 35mm lenses with the greater number of elements: 1.05 / (2.2 - .34) = 0.56.

Ralph likes to use a negative density range of 1.20 for normal. His normal CI (average gradient) is also 0.58. As both aim Contrast Indexes are identical, that means that any scene photographed and developed to a CI of 0.58 is going to be the same on the negative. How is this possible with different aim negative density ranges?

Ralph's model - 1.20 / 2.1 = 0.57

Flare model - 1.05 / (2.2 - 0.40) = 0.58

Last edited by Stephen Benskin; 07-26-2011 at 12:20 AM. Click to view previous post history.

Hi Stephen, thanks for your comprehensive reply. I would really love for Ralph to contribute to this discussion as he would have much to say on this topic he has devoted so much of his time, his book and this thread to it.

Your replies directly challenge his approach and the method from his book (WBM) which I had put plenty of trust in and was in the process of carrying out when your replies threw a spanner in my works.
Also, because Ralph has plenty of respect here and elsewhere, I find myself still aligned to his methods but am being forced to understand this more than I expected and come to my own conclusion if I am to break allegiance with what he does and recommends.

In his absence and prior to me digesting the pdfs you attached I don't expect to add much value to this thread. I might have more questions than answers I think. For now I will just add a few comments and go away and read your pdfs, and I will PM you my email address so you can send me "Simple Methods for Approximating the Fractional Gradient Speed of Photographic Materials"

Originally Posted by Stephen Benskin

Peter,
Okay, first we have to make sure we are referencing the same material. I believe you are using the pdf file from the Beyond Monochrome's website entitled "Testing Film Speed and Development." If you are then you also must be referring to fig 1 on page one of the document. And if that is the case, you are mistaking the log-H range for negative density range. Your use of the term "nomograph" if referring to figure 1 also confused me because I didn't think the term applies to that type of graph. It may, but in my earlier response, I was referencing the fig on page 140 of the first edition of Ralph's book.

Actually I was referencing the nomograph on p. 140 of WBM. There is a similar version of it on p. 66 of Stroebel's book "Basic photographic materials and processes" (p. 66 here) (which in turn was taken from a Kodak Publication from 1976) but the neg density ranges that nomograph lists differ from the one in WBM by quite a bit. e.g. the ISO std CI of 0.57 (in 1976) matches a neg dens range of 1.05, but in Relph's nomograh the ISO std of 0.58 matches a neg. dens. range of 1.29 (which I rounded up to 1.30 and used that in the formula I gave above.

Now why do the two nomographs differ so much ? I think it is because Ralph made a different assumption than that in the footnote (*) of p. 66 which reads "these are the typical negative density ranges that result when normal luminance range subjects (7 stops range) are exposed with moderate flare level lenses and developed to the contrast index shown in the left scale".

For now I don't understand enough to comment on the rest of your post except the last part where you write:

Originally Posted by Stephen Benskin

Ralph likes to use a negative density range of 1.20 for normal. His normal CI (average gradient) is also 0.58. As both aim Contrast Indexes are identical, that means that any scene photographed and developed to a CI of 0.58 is going to be the same on the negative. How is this possible with different aim negative density ranges?

Ralph's model - 1.20 / 2.1 = 0.57

Flare model - 1.05 / (2.2 - 0.40) = 0.58

Basically my only comment here is to Ralph: Help me out here please !!

Your replies directly challenge his approach and the method from his book (WBM) which I had put plenty of trust in and was in the process of carrying out when your replies threw a spanner in my works.
Also, because Ralph has plenty of respect here and elsewhere, I find myself still aligned to his methods but am being forced to understand this more than I expected and come to my own conclusion if I am to break allegiance with what he does and recommends.

Peter,

I'm not really directly challenging Ralph's approach. In fact, Ralph and I am in close agreement with most of the goals and conclusions. We just tend to disagree on the theory. I have a lot of respect for Ralph and there isn't anything I'm saying here that we haven't already discussed. What I am attempting to do is what you yourself have admitted. I wish to convey the importance of understanding the material to the point so you can come to your own informed conclusions. Approaching any material with a good amount of informed skepticism is just good methodology and it will help avoid the risk of dogma. My advice is to always ask yourself "why" and follow where it leads.

Actually I was referencing the nomograph on p. 140 of WBM. There is a similar version of it on p. 66 of Stroebel's book "Basic photographic materials and processes" (p. 66 here) (which in turn was taken from a Kodak Publication from 1976) but the neg density ranges that nomograph lists differ from the one in WBM by quite a bit. e.g. the ISO std CI of 0.57 (in 1976) matches a neg dens range of 1.05, but in Ralph's nomograh the ISO std of 0.58 matches a neg. dens. range of 1.29 (which I rounded up to 1.30 and used that in the formula I gave above.

Now why do the two nomographs differ so much ? I think it is because Ralph made a different assumption than that in the footnote (*) of p. 66 which reads "these are the typical negative density ranges that result when normal luminance range subjects (7 stops range) are exposed with moderate flare level lenses and developed to the contrast index shown in the left scale".

I've attached three nomographs. One is from Ralph's Way Beyond Monochrome. One is from Basic Materials and Processes. And one from Materials and Processes.

Here's my perspective on the three aims for a negative under normal conditions intended to be printed on a diffusion enlarger. There is no difference between them. That being said, I admit that it wouldn’t be very hard to come to the conclusion that they represent very different models, but once we take a closer look and break down the variables, they are virtually identical.

First, I do need to make a small correction in your interpretation of the negative density range that matches 0.58 or the 0.57 that is used as the aim avgGradient in the WBM monograph. I’ve drawn a line through the average conditions to the point that is labeled “Typical Diffusion Enlarger“. As you can see it has 1.20 as the negative density range for an avgGradient of 0.57. So, we need to adjust your 1.30 down to 1.20.

Let’s take a look at the variables (they are also in the attached word document):

WBM - NDR = 1.20, LSLR = 2.1 (7 stops), Starting CI = 0.57, Required CI = 0.57
Kodak - NDR = 1.05, LSLR = 2.1 (7 stops), Starting CI = 0.57, Required CI = 0.57
M&P - NDR = 1.05, LSLR = 2.2 (7 1/3 stops), Starting CI = NA, Required CI = 0.58

The biggest difference is between the aim NDR of WBM and the other two.

Let’s take a look at how the variables work together. The equation to determine gradient is Rise / Run. Or the Output (NDR) divided by the Input (log-H as it relates to the LSLR). (This can also be viewed in the word document.)

Since the Kodak and M&P numbers don’t equal the aim gradient, there can only be two possible conclusions. They are either wrong or there’s a variable missing. If you look at the bottom of the M&P monograph, it reads “corrected for average flare conditions.” Flare has been incorporated into the M&P monograph, but is missing from the above equation. We can also assume the Kodak monograph has also corrected for average flare. The WBM monograph works as is. Does this mean it hasn’t been corrected for average flare?

Equation incorporating flare: NDR / (LSLR – Flare) = CI (average gradient)
To find the value of flare: (CI*LSLR – NDR) / CI = Flare

As you can see with the Kodak monograph, if you start on the left at 0.57 and move through 7 stops luminance range to the middle scale at 0.57. Then if you draw a line through moderate flare, the final CI is at 0.57. If you repeat all the steps except draw the line through the high flare (box), the final CI needed to process the film to is 0.61. It’s exactly the same with the WBM monograph, but there is a small difference.

With the Kodak monograph and the M&P monograph, the scales have been “zeroed” out for normal flare although flare remains as part of the equation in determining the average gradient. With the WBM monograph, flare appears to have been eliminated from the “normal” 7 stop conditions. This is the only way to have identical values for the LSLR and CI variables while having a different NDR value. The WBM monograph works just as well as Kodak’s with determining the beginning and ending CI (avgGradient). The only difference is a question of the aim negative density range.

Can the different values just represent different preferences for the negative density range? That would make sense except it wouldn’t balance the equation. If the preference is for a negative density range of 1.20 with a luminance range of 7 stops that included flare, the final CI would have to be different than 0.57. For a one stop flare factor, the projected CI would have be 0.67 for the equation to work with a 1.20 NDR. If you have the same CI, luminance range, and flare, you can’t have different resulting negative density ranges, so the difference in the NDR can’t be attributed to NDR preferences.

No matter the assumption for the normal luminance range or the assumption for the amount of flare, if the film is developed to the same CI, the negative density range for a given scene will be the same. The M&P uses a 2.2 log luminance range, but it uses a higher flare factor, so the CI is basically the same as Kodak’s with a 2.1 log luminance range and a smaller flare value. I can argue that 2.2 is statistically more accurate as is the flare and Kodak only used 2.1 because it’s a nice round number and by doing so they had to fudge the value of flare in order to make the CI conform with reality. But no matter the reason, as long as the film has the same CI value, the results will be the same whether the luminance range is 2.2 with .40 flare or 2.1 with .30 flare. The end results are the same even though conceptually and theoretically one is more correct.

How the WBM NDR works depends on how you look at it. And this answer also answers how the Zone System can have an aim NDR of 1.25, while the WBM can have an aim NDR of 1.20, while (lack of a better term) the sensitometric aim NDR can be at 1.05 and all have almost identical aim CIs.

It’s all about how the data is interpreted. The Zone System uses a camera to make the test and the goal is to produce two points of density representing the end points for a seven stop exposure range: 0.10 over Fb+f and 1.35 over Fb+f. With this data, we can determine the aim CI or gradient: 1.25 / 2.1 = 0.59. What most people don’t realize is that while the test is shot with a camera, there is almost no flare using the single tone subject and the point where the exposure is made on the film curve (you are really just under and over exposing the mid-tone exposure). As not flare is involved, the aim NDR has to be made artificially higher in order to balance the equation and produce the necessary CI.

1.25 / 2.1 = 0.59 is really the same as 1.05 / (2.1 - .32) = 0.59. One calculation incorporates the flare factor with a realistic NDR and one has a NDR that doesn’t factor in flare. The concept becomes even clearer working with film curves. The proper way to produce a film curve is to contact the step tablet. This forms a non flare curve. If you have a curve with a CI of 0.59 and you measure a 2.1 log-H range from 0.10, the resulting negative density range would be 1.25. Since we know all scenes contain flare, the effective log-H range isn't measured at 2.1 but 2.1 minus the flare value. Both methods will indicate a negative processed to the same CI, so the results are identical. You can think of this as the difference between the sensitometric exposure or the non flare contacted step tablet exposure and the photographic exposure or the exposure that represents the variables encountered when shooting.

With the variance of flare in shooting conditions, the small difference in aim CI values between the three methods is inconsequential.

Therefore, there is no difference between the results of these three systems as there really is no real difference between the three nomographs; however, there is a difference conceptually. The Zone System and WBM NDR values are derived from interpreting results from the film curve with out incorporating flare to the subject luminance range. While all three methods will produce identical CIs, conceptually only one is technically correct (what is “correct” is a whole different post).

By understanding the “whys” of the variables, you are in control of the process and not the other way around. Don’t forget to keep this in mind when using the Film Speed and Development Test spreadsheet.

Hi Stephen,
well after your lengthy treatment of this I think I have come full circle. I'm sure I initially interpreted you correctly when I thought you were encouraging me to strongly consider contact printing the step wedge rather than doing the in-camera, low flare option I last described. Now I'm convinced you're saying it doesn't matter which method I use as long as I understand the theory sufficiently to properly interpret and apply the results.
I also thought you were indirectly challenging Ralph's method by asking "How is this possible": however in your last post you then finally managed to explain how it is possible without invalidating Ralph's method.

(You wrote:"Ralph likes to use a negative density range of 1.20 for normal. His normal CI (average gradient) is also 0.58. As both aim Contrast Indexes are identical, that means that any scene photographed and developed to a CI of 0.58 is going to be the same on the negative. How is this possible with different aim negative density ranges")

So with my new ultra low flare setup I intend on repeating my tests, hopefully this weekend.

Originally Posted by Stephen Benskin

First, I do need to make a small correction in your interpretation of the negative density range that matches 0.58 or the 0.57 that is used as the aim avgGradient in the WBM monograph. I’ve drawn a line through the average conditions to the point that is labeled “Typical Diffusion Enlarger“. As you can see it has 1.20 as the negative density range for an avgGradient of 0.57. So, we need to adjust your 1.30 down to 1.20..

I did make a typo, but not the one you corrected ! I actually meant to write
"in Ralph's nomograh the ISO std of 0.615 (not 0.58) matches a neg. dens. range of 1.29 (which I rounded up to 1.30)" you can see this is where the ISO Standard arrow is pointing on the far LHS of WBM's nomograph you attached.

To me there really is only one testing method for film and that is contacting, but however you decide to go, I'm glad to see that it is with an open mind and from an informed position. I'm sure you will find the testing process to be an incredibly valuable learning experience. Best of luck.

Stephen

Last edited by Stephen Benskin; 07-28-2011 at 08:16 PM. Click to view previous post history.