0.6 -- standard contrast?

[Stephen Benskin]

I have heard that "standard contrast" for negative film is 0.6, on the straight-line portion of the h&d curve. I have a couple of questions about this.

1. is this true?
2. Why 0.6 not 1?
3. Do manufacturers development data uniformly target this standard contrast value?
4. If I have papers from grade 0 to grade 5 at my disposal, how much tolerance does that give me with regards to the contrast of my negatives?

I am trying to determine if there is a development recipe that I can use with all the different films I use.

1. The standard is based on the statistical average conditions and the average paper grade. Loyd Jones determined the statistical average luminance range is 2.20 logs or 7.33 stops. Now this doesn't include the entire range from accent black to a pure specular reflections. This is all based on Tone Reproduction Theory which is a fascinating subject. Kodak had 0.56 as their normal of a couple of decades, but changed it to 0.58 because of the increased use of 35mm cameras which have a slightly higher average flare than larger formats. Flare went from 0.34 to 0.40 or 1 1/4 to 1 1/3 stops. According to the below chart published in Photographic Materials and Processing, the LER that falls in the center of the range for a grade 2 paper printed on a diffusion enlarger is 1.05. This also doesn't include the entire range of the paper.

1.05 / (2.20 - 0.40) = 0.58



2. Why not 1? The negative is not the final step in the process. Paper just so happens to be slight under 2.00 (~1.70) so the negative needs to be about half that in order the reproduction to equal 1.00. negative gradient x paper gradient = 1.00.

3. Manufacturers tend to follow the statistical averages.

4. Please see the chart above and the table I've uploaded. While matching the NDR to the Paper LER helps match the film to the paper, it doesn't account for the local contrast and how the print tones relate to the original values.

Other people have already pointed out the connection between 0.60 sounding familiar and the ISO black and white film standard. The number; however, can be misleading. While it is technically processing the film higher than normal, it doesn't make a change in the film speed when you consider the standard uses the Delta-X criterion for speed determination. Delta-X is a way to mathematically determine the placement of the fractional gradient point using an easy to find fix density point. The contrast parameters of the standard are part of the math. If the negative has a density of 0.80 at a Δ1.30 log-H from the 0.10 density point, the difference between the 0.10 and fractional gradient point is always 0.296. Only when these conditions are made can you determine accurate film speed. At other film gradients, the Delta-X equation must be used. As the film speed doesn't change very quickly with changes in film gradient, the slightly higher gradient for ISO testing does not give a false higher speed value.

 

[Stephen Benskin]

The Kodak Xtol datasheet says that if you develop TMY for 9.25 minutes, that the contrast will be 0.56 at an exposure index of 400, but 0.62 at EI 800. Why would exposure index make any difference in contrast, at the same development time?

It's about "pushing for speed." Which you can't do and that's why the phrase is in parentheses. Same thing as Kodak does with TMZ.

 

[Photo Engineer]

The final paper print averages 1.5 - 1.7 in contrast (delta density / delta log E). This is done so that the eye can integrate the overall contrast as a comfortable 1.0 in "average" contrast with enlarger flare factored in. Thus, the film is about 0.6 - 0.7 with lens flare factored in. The pro films are lower in contrast than the consumer films in this range.

As you increase development time, the contrast of the film goes up and the apparent speed point goes up. You are trading contrast and grain for an apparent speed increase.

When we start from scratch to design a film, we have an aim curve drawn for us by our standards committee and which guarantees the correct contrast and speed point for a given product. The release process for B&W in my time was seasoned D76, straight, no dilution, and speed, contrast, sharpness and grain were all evaluated in this process.

The low contrast value for film was chosen to give the longest latitude possible, and so that printers could select the point on the straight line for optimum printing. Or, they could use split grade or a host of alternative processes. A low contrast film would limit option. Besides which, paper has its own limiting latitude imposed by the lack of ability to go much over 2.0 in density in an enlarging paper. So, the film supplies the latitude and the paper supplies the best image possible imposed by the laws of physics.

Neither the toe or the shoulder of a negative should be used except in emergencies.

Now, onward.... Reversal films have a gamma of about 1.5 - 1.7 and thus to duplicate this curve in a print, the print material must have a contrast of about 1.0. This means that there is basically no latitude for exposure error in either the film or the print process which also means that reversal printing is much harder and requires masking to fix up the overlay of the toe to toe and the shoulder to shoulder problem which crops up in reversal printing and which makes generational prints look worse and worse.

You can make about 12 generation of neg-pos, neg-pos dupes with existing films, but with reversal it is just about 1 pos-pos iteration and this is also why Hollywood has spurned the use of pos-pos systems. They are not and at the current time cannot duplicate neg-pos systems.

Now, at one time, Kodak did design a pos-pos system with an intermediate that was 0.6 in gamma and was even masked. It basically went nowhere due to lack of interest.

PE

 

[BetterSense]

The low contrast value for film was chosen to give the longest latitude possible, and so that printers could select the point on the straight line for optimum printing. Or, they could use split grade or a host of alternative processes. A low contrast film would limit option. Besides which, paper has its own limiting latitude imposed by the lack of ability to go much over 2.0 in density in an enlarging paper. So, the film supplies the latitude and the paper supplies the best image possible imposed by the laws of physics.

This paragraph answers my most pressing questions perfectly.

 

[Bill Burk]

The Kodak Xtol datasheet says that if you develop TMY for 9.25 minutes, that the contrast will be 0.56 at an exposure index of 400, but 0.62 at EI 800. Why would exposure index make any difference in contrast, at the same development time?

I've changed my mind about it being a typo regarding the times.

If you cross-reference the TMY2 datasheet there is a lot of emphasis in words that you do NOT increase development time for 1 stop push.

So my opinion now: Times given in the XTOL table for TMY2 correctly reflect the discussion in the TMY2 datasheet that you do not have to develop longer for a one stop push to EI 800.

This could save you money if you don't have to pay labs for Push Processing surcharges. It could be convenient if you batch develop film together that was shot under different conditions. So it's a good thing you don't have to develop longer for a 1 stop push.

But as for the cell that shows 0.62 Contrast... I think it is still a good contrast aim if you have the flexibility to develop for different times. So even though 0.62 doesn't match up with the recommended times, 0.62 is not WRONG.

 

[Photo Engineer]

I wish to change the quote above from my post.

A film of too low contrast would limit options just as one with too high contrast. Too low and you cannot fix any problems really, and too high and you have no latitude to work with at all.

It does not change the sense of what I said, but merely puts proper limits on the paragraph.

PE

 

[Stephen Benskin]

The ISO method of determining speed does not directly use the straight-line contrast, but rather seems to define contrast in the toe region. If I understand correctly, different films, when developed to the ISO criteria for shadow speed, may then have varying straight-line contrast. Actual photographers, on the other hand, seem obliged to develop to a target contrast that looks correct. I would be interested to know what this contrast ends up being for different films when they are developed according to the ISO standard.

I just looked at some Tri-X 120 and 4x5 both with long toes. The average gradient of the films that fit in the ISO parameters was 0.67. TMX fit at just over 0.61.

 

[ic-racer]

1. The standard is based on the statistical average conditions and the average paper grade. Loyd Jones determined the statistical average luminance range is 2.20 logs or 7.33 stops. Now this doesn't include the entire range from accent black to a pure specular reflections. This is all based on Tone Reproduction Theory which is a fascinating subject. Kodak had 0.56 as their normal of a couple of decades, but changed it to 0.58 because of the increased use of 35mm cameras which have a slightly higher average flare than larger formats. Flare went from 0.34 to 0.40 or 1 1/4 to 1 1/3 stops. According to the below chart published in Photographic Materials and Processing, the LER that falls in the center of the range for a grade 2 paper printed on a diffusion enlarger is 1.05. This also doesn't include the entire range of the paper.

1.05 / (2.20 - 0.40) = 0.58

View attachment 101336

2. Why not 1? The negative is not the final step in the process. Paper just so happens to be slight under 2.00 (~1.70) so the negative needs to be about half that in order the reproduction to equal 1.00. negative gradient x paper gradient = 1.00.

3. Manufacturers tend to follow the statistical averages.

4. Please see the chart above and the table I've uploaded. While matching the NDR to the Paper LER helps match the film to the paper, it doesn't account for the local contrast and how the print tones relate to the original values.

Other people have already pointed out the connection between 0.60 sounding familiar and the ISO black and white film standard. The number; however, can be misleading. While it is technically processing the film higher than normal, it doesn't make a change in the film speed when you consider the standard uses the Delta-X criterion for speed determination. Delta-X is a way to mathematically determine the placement of the fractional gradient point using an easy to find fix density point. The contrast parameters of the standard are part of the math. If the negative has a density of 0.80 at a Δ1.30 log-H from the 0.10 density point, the difference between the 0.10 and fractional gradient point is always 0.296. Only when these conditions are made can you determine accurate film speed. At other film gradients, the Delta-X equation must be used. As the film speed doesn't change very quickly with changes in film gradient, the slightly higher gradient for ISO testing does not give a false higher speed value.

I don't think that answers the question because it leaves the question of what is so special about #2 paper.

My research and personal experience has me understanding that rollfilm negatives have higher quality if they are processed to gammas of less than 1. So my premise is papers are manufactured to support those lower gammas. Not the other way around.

I don't know of a research paper that specifically addresses these issues (that is why I am engaging you further into the conversation) but I have inferred my above premise on tangential remarks present in some of the papers I reviewed (many of which you sent to me).

For example my premise is supported in the Nelson paper Safety Factors in Camera Exposure, 1960. There is a graphic indication of the adverse effects of extended development of enlarged rollfilms.

 

[BetterSense]

Illustrative curves for two hypothetical films (zero B+F density) with the same ISO speed but different toe shapes.

Thanks for the explanation. I don't think this explains the TMAX datasheet thing though. I don't see how the curve shape could be said to change if the film is developed the same. Kodak must be using a different definition of contrast index here.

 

[Thomas Bertilsson]

Thanks for the explanation. I don't think this explains the TMAX datasheet thing though. I don't see how the curve shape could be said to change if the film is developed the same. Kodak must be using a different definition of contrast index here.

Actually it does. The curve is a component of three things:
1. The film characteristics
2. Film exposure
3. Developing regime

1 and 3 stay the same, while film exposure changes. That should put some of the shadow values below the threshold of the toe of the curve, resulting in increased overall contrast.

 

[Photo Engineer]

Well, first, lets leave out contact printing. It does not address flare. All enlarging papers have to address this issue in gauging contrast for normal negatives.

That said, an average grade 2 paper has a mid scale contrast of 2.5 measured using the straight line and centered between a density of 0.8 and 1.0. A paper of 3.0 however, is suggested as being best used as a normal contrast paper if one has a diffusion enlarger. This comes from some of the earliest texts I have.

Almost all films show exposure in the toe and shoulder to some extent in average scenes, but the paper cannot handle it. So, we use what we can which is in the straight line portion. The "first acceptable print" method has determined that this is the print that has little or no curvature from the film in it.

And having conducted many many tests for this compiling data at the tutelage of Daan Zwick, I can assure you that it does work. However, examining all of the data is tedious and time consuming, especially when it is an "extra duty". So, if you look up Daan, you may find some work. Also try Jim Bartelson and Leroy DeMarsh.

PE

 

[ic-racer]

The low contrast value for film was chosen to give the longest latitude possible,...

PE

Good explanation, thanks for your input. The graph I posted above from Nelson supports this.

 

[Photo Engineer]

Interestingly enough, many early films were used to make lantern slides. They often used the same emulsion for taking and projecting and thus a contrast of 1.0 was the most useful (actually a bit higher than that) and thus they made the original exposure and the dupe on the same batch of plates.

Exact exposure was always a problem for both "products".

PE

 

[Stephen Benskin]

The "first acceptable print" method has determined that this is the print that has little or no curvature from the film in it.

Real quick. The first excellent print test lead to the fractional gradient method, which by it's very nature having the speed point at 0.3 time the average gradient places exposure in an area of curvature.

 

[Stephen Benskin]

The Kodak Xtol datasheet says that if you develop TMY for 9.25 minutes, that the contrast will be 0.56 at an exposure index of 400, but 0.62 at EI 800. Why would exposure index make any difference in contrast, at the same development time?

This is how it works. The first example has a CI of 0.56 and the shadow exposure is placed on 0.10 above Fb+f. The NDR is 1.05.



The next example has the same curve but with one stop underexposure which is the same as rating the film at twice the indicated speed. The exposure shifting 0.30 to the left leaves a NDR of 0.93. Underexposure basically reduces the effective log-H range. Instead of a log-H of 1.85 like the normal exposure, the effective log-H range is 1.55.



Extending the development to create a film with a CI 0.61 shifts 0.10 slightly to the left and increased the gradient of the film, so while the shadows remain thinner than a normally exposure negative, the density range is more or less the same but with a log-H range still effectively shorter than with the normal exposure. This is what is known as "pushing for speed."



Some might have noticed that the CI change is not as much as a +1 when compensating for a scene with a stop shorter subject Luminance range. When I was working in a lab, I tried to make a distinction between pushing for speed or contrast, but the customers didn't understand the concept, became suspicious, and refused to cooperate.

 

[Stephen Benskin]

I don't think that answers the question because it leaves the question of what is so special about #2 paper.

My research and personal experience has me understanding that rollfilm negatives have higher quality if they are processed to gammas of less than 1. So my premise is papers are manufactured to support those lower gammas. Not the other way around.

I don't know of a research paper that specifically addresses these issues (that is why I am engaging you further into the conversation) but I have inferred my above premise on tangential remarks present in some of the papers I reviewed (many of which you sent to me).

For example my premise is supported in the Nelson paper Safety Factors in Camera Exposure, 1960. There is a graphic indication of the adverse effects of extended development of enlarged rollfilms.

View attachment 101392

I don't think there's any thing special about Grade 2 except it's in the middle.

You will notice in the graph how the degree of enlargement plays a large factor in how quickly image quality falls off with exposure. That is one of the main reasons why the 1960 standard dropped the safety factor effectively increasing film speed, and consequently reducing negative density. Image quality with large format isn't as affected by exposure increases, but the use of smaller formats was on the increase, so film speed was adjusted accordingly.

There are a number of papers by Jones. Two major ones that come to mind and ones I haven't read for awhile are:

Jones, L.A., and Nelson, C.N., The Control of Photographic Printing by Measured Characteristics of the Negative, Journal of the Optical Society of America, V. 32, 1942.

Jones, L.A., and Nelson, C.N., Control of Photographic Printing: Improvement in Terminology and Further Analysis of Results, Journal of the Optical Society of America, V. 38, No. 11, 1948.

 

[Bill Burk]

Here's the dilemma or paradox BetterSense is talking about.

My proofreader's opinion is that the number marked with an asterisk should be 0.56

 

[Stephen Benskin]

Here's the dilemma or paradox BetterSense is talking about.

My proofreader's opinion is that the number marked with an asterisk should be 0.56

Looked at the sheet. Typo. Apparently for some of the films Kodak doesn't recommend adjusting development for one stop underexposure. Looks like with the TMY, they missed combining the two EIs into a single row.

 

[Photo Engineer]

Real quick. The first excellent print test lead to the fractional gradient method, which by it's very nature having the speed point at 0.3 time the average gradient places exposure in an area of curvature.

Stephen, these were all evolving at the same time as everyone realized that what was needed was a long latitude film and this could not be done with reversal. Everyone in design knew this and knew how to test for it to prove its reality.

The qualification needed was the ability to coat and test films and papers with different contrast gradients. I am one of the lucky few who have been able to do this (within limits - the cost burden to the project at hand), and to prove these suppositions against actual results and customer comments.

To do this, Daan would send out bundles of prints to customers for comments. Leroy would read the densities in a MacBeth viewing cabinet, and the results of comments and readings would be tabulated. And, BTW, the densities actually viewed NEVER matched those of a reflection densitometer due to angle of view (goniophotometry). Thus, all data had to be reconciled.

This is not a math problem for the head, it is a combined human - math - coating problem which must be done in its entirety in order to comprehend the actual "print space". Grant and I had the opportunity to discuss this at length during our tenure at KRL together, and we agreed (I think) that this was not just about graphs. The piles of Zwixpix that I had in my office bore testimony to that.

PE

 

[Stephen Benskin]

Stephen, these were all evolving at the same time as everyone realized that what was needed was a long latitude film and this could not be done with reversal. Everyone in design knew this and knew how to test for it to prove its reality.

The qualification needed was the ability to coat and test films and papers with different contrast gradients. I am one of the lucky few who have been able to do this (within limits - the cost burden to the project at hand), and to prove these suppositions against actual results and customer comments.

To do this, Daan would send out bundles of prints to customers for comments. Leroy would read the densities in a MacBeth viewing cabinet, and the results of comments and readings would be tabulated. And, BTW, the densities actually viewed NEVER matched those of a reflection densitometer due to angle of view (goniophotometry). Thus, all data had to be reconciled.

This is not a math problem for the head, it is a combined human - math - coating problem which must be done in its entirety in order to comprehend the actual "print space". Grant and I had the opportunity to discuss this at length during our tenure at KRL together, and we agreed (I think) that this was not just about graphs. The piles of Zwixpix that I had in my office bore testimony to that.

PE

Ron, you misrepresented the conclusions from the "first excellent print." Your statement was about a specific test. It was by Jones. It was psychophysical. It lead to the fractional gradient that places the speed point in the toe. It did not conclude negative exposure needed to avoid the toe. Subsequent tests by other people isn't what your statement was about. My response was only about your statement concerning the first excellent print and nothing else.

Jones' test was called the FIRST excellent print, not the Absolute best excellent print. Quality increased a little with increased exposure and then leveled out. We all make misstatements from time to time. It's only natural with the speed in which these posts are created. I said something about transparencies earlier that wasn't what I meant to say. It's no biggie.

 

[Stephen Benskin]

Illustrative curves for two hypothetical films (zero B+F density) with the same ISO speed but different toe shapes.

Excellent example.

 

[Photo Engineer]

Ron, you misrepresented the conclusions from the "first excellent print." Your statement was about a specific test.

No, I did not. I know what "first excellent print" means and in fact delineate this and related latitude in portions of my book. The first figure shows the first excellent print, but the majority of excellent prints are further up the curve or slightly over exposed from one POV. The second two figures from Haist show the effects of exposure on latitude or the effects on getting an acceptable print. These three figures are all that one needs to understand more fully, neg-pos systems.

PE

 

[Stephen Benskin]

No, I did not. I know what "first excellent print" means and in fact delineate this and related latitude in portions of my book. The first figure shows the first excellent print, but the majority of excellent prints are further up the curve or slightly over exposed from one POV. The second two figures from Haist show the effects of exposure on latitude or the effects on getting an acceptable print. These three figures are all that one needs to understand more fully, neg-pos systems.

PE

Ron, The first graph comes directly from one of the Jones papers connected with the first excellent print test. It shouldn't take me too long to track down the paper and page. Haist must have reprinted it. The other two are familiar but I'm not sure where they are from.

 

[Photo Engineer]

The first of those I cite comes from the work of J. L. Tupper and was published in the '40s IIRC. Both Mees and Haist published variations on the second two from their work in the '40s and '50s. The work published by Mees was also originally done by Tupper and others on Mees staff.

PE

 

[Bill Burk]

In your diagram, M of "First Excellent"... That's the famous 0.3 times gradient point, right? So "a little bit" on the toe.