Film curve plotting and fitting

[Ray Rogers]

Photons, despite being massless, actually do have momentum. So, I think film speed would be faster if you ran rapidly toward your subject.

BetterSense



Can't argue with the Laws of Physics!
Just don't forget to open the pin hole!



Stephen,
Can I too, get a copy of the paper you just sent Ralph?

Thanks to both of you!
Thanks to BS for making me laugh... and
Thanks to SB for making me think!

Ray

(well, i will eventually! )

 

[tlitody]

Can't argue with the Laws of Physics!

Want a bet?

if E = mc² and m = 0 then E = 0 too, so it follows that c² = E/m and if E and m are zero then c² = 0 too, so light has no speed, mass or energy according to some.

 

[Robert Budding]

Ralph,

I do now. I've already emailed it to you.

Robert,

I was hoping it would show more than that.

I'm with Einstein:

Make everything as simple as possible, but not simpler.
~ Albert Einstein

Any zone that I expect to produce detail is on the 'straight' portion of the film curve.

 

[Photo Engineer]

And C = (M/E)^1/2 which implies that both M and E can be negative or that the entity M/E can be imaginary.

Imagine that! That is the basis BTW of Tachyon theory.

PE

 

[Robert Budding]

Now if PE could just unify Relativity with Quantum Mechanics . . .

 

[Photo Engineer]

Morgan Freeman is doing that! Or was it Edward Witten? Or Michio Taku? IDK.

PE

 

[Stephen Benskin]

In the category of it's never as simple as you think, the following paragraph is from Factors Affecting the Quality of Black-and-White Reflection Prints and is referring to the camera exposure vs quality graph in post #168.

"Both curves reach the maximum quality level of 100 per cent. It can be seen, however, that the long-toe Film B is capable of producing prints of highest quality over a greater range of camera exposures than the short-toe Film A. To achieve optimum quality with the short-toe film requires a critical choice of slight under-exposure and subsequent printing on a high-contrast grade of print material. With extreme over-exposure, however, the long-toe film results in negatives with an excessively high density range which available print materials cannot accommodate; hence, a loss in print quality results. The short-toe film shows a superiority in this over-exposure region."

 

[RalphLambrecht]

In the category of it's never as simple as you think, the following paragraph is from Factors Affecting the Quality of Black-and-White Reflection Prints and is referring to the camera exposure vs quality graph in post #168.

"... With extreme over-exposure, however, the long-toe film results in negatives with an excessively high density range which available print materials cannot accommodate. ..."

Steve

I don't now what they mean by 'extreme', but it makes sense that tonality not compressed by a long toe increases the density range. However, who wants to lose shadow detail in order to fit the density range onto the paper? We've all seen these pictures, nice contrast in the midtones but dead shadows. That's not the answer. I rather overexpose and underdevelop and then take the hardest paper I can afford. Whatever I lose in shadows and highlights that way, I get back with dodging and burning. This way, I get contrasty midtones and don't lose any shadow or highlight detail.

 

[Stephen Benskin]

Steve

I don't now what they mean by 'extreme', but it makes sense that tonality not compressed by a long toe increases the density range. However, who wants to lose shadow detail in order to fit the density range onto the paper?

I don't think we should get stuck on semantics. Simonds also says that short toes curves should be printed on high contrast papers when it is obvious he just means higher contrast papers. He could also be referring to the reason why the quality of long toed films drops off before short toed.

The graphs aren't talking about the best way to photograph a scene. This was a case of camera exposure vs quality and how closely it can come to 100% (as it is defined). Of course, as exposure increases with long toed films, there is an increase in NDR because of the upsweep and it's a simple case to reduce development slightly or work the highlights in printing, but this isn't the subject of the graph or paragraph. Other graphs in the paper deal with other factors such as fig 6 & 7 that are about negative gamma and paper grade respectively.

For me, what I take away from the paragraph is one size doesn't fit all nor are there any universal hard and fast rules.

One of my concerns is always has to do with whether the results we think we have are actually connected to the factors we attribute them to. As Simonds concludes, "the data from many such computations strikingly point out an important consideration when direct visual comparisons are being made of the quality of prints obtained from different negative materials small, but subjectively significant, quality difference can be realized by changing the D-log E characteristic of the negative material; but these small quality differences can only be appreciated when the levels of all the other variables of the reproduction system are carefully controlled. The importance of optimizing the printing conditions cannot be emphasized too strongly when comparisons are being made of prints made from different negative materials. Even a slight error in print exposure can obscure a potential quality improvement offered by a negative material with an improved D-log E curve shape. Examination of the data of these studies has emphasized the fact that a valid comparison of the relative merits of two films requires extreme care and accuracy on the part of the technician preparing the demonstration prints."

 

[RalphLambrecht]

... For me, what I take away from the paragraph is one size doesn't fit all nor are there any universal hard and fast rules. ...[/I]

Stephen

I must admit that I'm very surprised to see you coming to this conclusion. Anyway, you must have changed your mind, because you highlighted something completely different in the text:

To achieve optimum quality with the short-toe film requires a critical choice of slight under-exposure and subsequent printing on a high-contrast grade of print material.

By the way, I cannot agree with either of these statements. There are some (very good) universal rules when it comes to film exposure and development, and underexposure is not on the path to optimum quality!

 

[Stephen Benskin]

Ralph,

I've always been that way. I just tend to post perspectives that are missing from the conversations. My philosophy is you need to understand how something works in order to move beyond it. My photography isn't about reproducing reality. I tend to print for a heighten sense of reality emphasizing extended shadow and mid-tones. In fact, I've had people confuse some of my work for drawings. When I shoot, even though I set the EI at my tested EFS, I place the exposure where I have usable tones and that placement is dependent on the subject. For me, the negative is there to capture the information, not define it.

I don't understand the slight underexposure comment either. Plus, I find it counter-intuitive that the short toed film needs a higher contrast paper than a long toed film or that a long toed film reaches a higher level of quality than a short toed film. Still, there it is.

There are a number of counter-intuitive aspects of photography that I find interesting. For instance, as the illumination on a print dims, there's a point where the darker tones look lighter and the lighter tones look darker. Another counter-intuitive example comes from The Theory of the Photographic Process, "When the film characteristics, camera exposure, and negative development are accurately controlled and held at what might me called the "normal" level, there is a distinct tendency for observers to prefer prints made on the "normal" grade of paper regardless of the luminance range of the scene." A flat scene processed normally will look better on a grade 2 paper?

Let's consider some possibilities of the underexposure statement. First, it is slight underexposure and what is his definition of underexposure as there is a safety factor built into film speed. Simonds never said anything about losing detail, and since short toed films have a good toe gradient, there shouldn't be a problem with loss of shadow separation. It would probably be comparable to normal exposure with a long toed curve.

Second, he said the negatives were made within a studio setting where flare is lower. "The effect (of flare in outdoor conditions) has been to convert the short-toe film into a long toe film...These considerations indicate the desirability of a long-toe film for studio photography and a short toe film for outdoor photography in which normal amounts of camera flare is present." Perhaps if the tests were done outdoors, the results would have been different.

Third, one of the images Simonds used was a portrait. From Theory of the Photographic Process, "For the portrait subject, a higher gradient is evidently desired in the middletone and highlight regions of the reproduction, where the skin tone usually lies. It is obtained at the expense of subdued contrast in the shadow regions." Maybe that is a contributing reason why the "slight underexposure" and higher contrast for a short toed film as the long toed film will naturally produce this effect. Kodak used to advertise TXP as a studio film and portrait papers were said not to have a straight-line.

Finally, maybe it's just something new to us that we need to consider.

Since this information is scientifically obtained, I find it's worth pondering, and at the very least, to realize that there are always caveats. As the ultimate goal of photography is to produce something that is subjective, art, I can bet that there are caveats even to those universal rules of film exposure and development. Of course, I'm not talking about the chemical process of development or the physics involved in exposure. It might be fun to take a look at some of them. If it were all so cut and dried, there would only be one system and we wouldn't have need of magazines and forums such as this.

 

[Photo Engineer]

In post 185 Steve, I can disagree with the part from Mees. That book was originally written when the old style films were all that were available and was updated just as the new straight line films with sharper toes were being introduced. Neither Mees nor James were known as photographers, but rather as engineers.

Therefore, I suspect that quote and the following information is merely an opinion being expressed based on the films they had, and not on scientific proof.

PE

 

[Stephen Benskin]

The Theory of the Photographic Process 3rd edition 1966 was edited by James. The chapter from which the quotes were acquired is "The Theory of Tone Reproduction" written by C.N. Nelson. The theory books were a compilation of current photographic knowledge and research originally created as a reference for the employees of Kodak. Chapter 22 contained many references to the series of papers by Simonds published in the late 50s and earily 60s as well as many of the graphs from his papers.

The quotes used in post #158 were to help propose a possible explanation to a statement made by Simonds in a 1962 paper. Simonds suggested that long toed films were preferable in studio portrait work. Until this year, Kodak sold a classic long toed film - TXP.

For those unfamiliar with C.E. Kenneth Mees. He established the R&D department at Kodak. George Eastman want Mees so much, he bought the company Wratten, in which Mees worked. There is a very nice biography available from the George Eastman House, C.E. Kenneth Mees - Pioneer of Industrial Research. Among the books Mees wrote is a history of photographic research, From Dry Plates to Ektachrome Film: A Study of Photographic Research.

 

[Photo Engineer]

Steven;

Thanks for clarifying the actual author of that statement.

I am quite familiar with the work of Mees and James, and in fact, I knew Howard as we worked on and off together on several small items. He also taught several of the courses that I attended. So, I am familiar with the book.

Pursuant to this discussion, I draw your attention to figure 22.4 in that chapter in which Nelson shows the different print scales needed for optimum reproduction depending on the viewing conditions. This, of course, is carried through the chapter showing how the different tones in that figure essentially require different starting curves.

Now, to be sure, Nelson shows tons of curves, many of which you have reproduced here in one form or another such as figure 22.10 which we discussed previously in another thread and which you added in an earlier post here.

My point being that when we did this work (and I have done most everything in that chapter personally from start to finish), I have to say that my supervisor reminded me that we sold pictures, not graphs, curves or test charts.

So, having seen all of your many many references and graphs and charts, may we see some of your pictures to illustrate these? I don't seem to have seen many of them to illustrate all of the data.

Thanks.

PE

 

[Bill Burk]

Thanks guys, let me see if I understand the counterintuitive. Here's how I interpret...

Long toe films, developed to ASA, will reach 0.1 while there is still quite a bit of toe left. So the straight line portion has to be a higher pitch to make up the difference. The higher pitch will continue eventually blowing out the highlights. But straight toe film will have gone up most the toe by 0.1 so it won't have to make up time and it can have a gentler slope. That gentler slope will continue into the highlights. So you will need higher contrast paper to print since the negative will be flatter.

I also get the impression the tests were done with standardized development, which I believe was a very reasonable scientific decision to make. But since many of us vary the development time or chem, to accommodate style or subject, we can (and do) arrive at different conclusions for what is optimal.

Bill

p.s. In an earlier post I said I cannot get good results with a long subject range. I meant that as a very specific comment about my results because my film is mottled at 4-5 minutes development time. The mottling will make my picture ugly. You are all welcome to take pictures of unlit cave interiors and beaches and develop in Pyro. I'm trying to stick to one camera, film, developer, paper.

 

[Photo Engineer]

The toe does not determine the slope (or pitch?). It only determines the total length of the straight line portion of the curve and therefore the effective latitude. Some early films had no straight line portion and therefore contrast varied from zone to zone (I hate that term (zone) ) .

PE

 

[Photo Engineer]

Yours either dude.

Steve;

Please view my gallery then, and read my fully documented articles with curves and photos in the Emulsion Making and Coating forum!

PE

 

[Stephen Benskin]

I don't have any research photographs. I do have curves that I've generated in my research. But if anyone is interested in my photography, I have images on my Facebook page, however, I can't guarantee the quality of the digital reproduction.

 

[Bill Burk]

Hi PE,

One day there will be prints to illustrate these stories. Sorry, I got tired of the terms Contrast Index, Average Gradient and gamma. (I'll avoid "zone" too for it is as confusing a term as "level" is in Advanced Dungeons & Dragons.)

The toe does not affect the average at all that is true. But I think the toe affects the immediate slope at the 0.9 density when you develop for ASA.

Pictorially, this would only affect highlights of a long range subject or an accidental or deliberate gross overexposure - and when you develop the film to ASA.

I will grant you that the difference is insignificant if your careful exposure keeps you off this part of the curve. And anyone here who deliberately exposes on this part of the curve would very likely develop less than ASA.

 

[ic-racer]

Well, 20 pages of exposure discussion. Now how about some thoughts on determining gamma or contrast index or slope or "any useful information about development" from a film curve. Again, I'm referring to some computer generated information based on a dataset rather than plotting the dataset by hand.

This is how I do it based on the software I have.
I have the software plot out the curve and I make sure it is not some crazy thing. Then I have the software give me a linear regression of first eleven steps above 0.1. (based on a 21 step wedge sensitometer exposure).

I don't have a name for the number because all the names have been taken. I can't call it "Gamma" or "Contrast Index" because those terms are already in use and are measured differently on the curve.

What would I like to see in software??

Just as I'd like software that will give me the 0.3G point from a dataset, I'd also like a computerized determination of the "Contrast Index" like one would get from the good old fashioned clear plastic "Contrast Index Gauge" http://www.tpub.com/content/photography/14208/css/14208_56.htm

 

[RalphLambrecht]

... I'd also like a computerized determination of the "Contrast Index" like one would get from the good old fashioned clear plastic "Contrast Index Gauge" http://www.tpub.com/content/photography/14208/css/14208_56.htm

Why? Is it really worth it to make, buy and run software for that? The plastic gage is quicker than even starting a program. How many curves are you planning on measuring?

 

[Kirk Keyes]

So, having seen all of your many many references and graphs and charts, may we see some of your pictures to illustrate these? I don't seem to have seen many of them to illustrate all of the data.

PE - I have one of Stephen's prints - the one that is his avatar. It is a very fine print. In fact, it's quality ranks right up there with prints I have by the Adams Trust (printed by Alan Ross), Alan Ross, and John Wimberley.

Stephen had a portfolio printed in Photo Techniques about 4 or 5 years ago if you would like to see some reproduced there.

 

[ic-racer]

Why? Is it really worth it to make, buy and run software for that? The plastic gage is quicker than even starting a program. How many curves are you planning on measuring?

But wait, for 20 pages on a thread on "film curves" you think we should be plotting them on graph paper

To do it on graph paper, you have to plot all the points, which takes longer than typing in the numbers in the column. Then line up the thing.

Then you have wasted a nice piece of paper. Remember the 'paperless' office...

This is a serious pet peeve for me in that I have essentially supercomputing power to run some 'Flash Player' to display ads, but I can't have the computer make some straight forward 'computations'

 

[tlitody]

But wait, for 20 pages on a thread on "film curves" you think we should be plotting them on graph paper

To do it on graph paper, you have to plot all the points, which takes longer than typing in the numbers in the column. Then line up the thing.

Yeah and you would have finished it a lot quicker than this thread has been running

Then you have wasted a nice piece of paper. Remember the 'paperless' office...

Haven't you heard of the bitless office for people who aren't operated by computers?

This is a serious pet peeve for me in that I have essentially supercomputing power to run some 'Flash Player' to display ads, but I can't have the computer make some straight forward 'computations'

Operator error in my book. Suggest you look at post #83 where you claim to be able to write C++. So instead of bellyaching that no one has done it for you, get on and DIY.

 

[Stephen Benskin]

Well, 20 pages of exposure discussion. Now how about some thoughts on determining gamma or contrast index or slope or "any useful information about development" from a film curve. Again, I'm referring to some computer generated information based on a dataset rather than plotting the dataset by hand.

This is how I do it based on the software I have.
I have the software plot out the curve and I make sure it is not some crazy thing. Then I have the software give me a linear regression of first eleven steps above 0.1. (based on a 21 step wedge sensitometer exposure).

I don't have a name for the number because all the names have been taken. I can't call it "Gamma" or "Contrast Index" because those terms are already in use and are measured differently on the curve.

What would I like to see in software??

Just as I'd like software that will give me the 0.3G point from a dataset, I'd also like a computerized determination of the "Contrast Index" like one would get from the good old fashioned clear plastic "Contrast Index Gauge" http://www.tpub.com/content/photography/14208/css/14208_56.htm

Delta-X Criterion gives you the 0.3G point and for the rest of the curve plotting, I'll send you a paper which I think covers it.