My first Zone System test [Archive] - Large Format Photography Forum

venchka

12-Sep-2008, 11:06

Hmmmmmmmmmmm...

Here are my thoughts and bear in mind:
1. I know nothing. Very possible.
2. I'm about to make the same test myself with Ilford HP5+.

1. You won't be photogrpahing light boxes. Set your camera up to photograph a real subject refelcting light. Use the light box to evenly light a grey card?
2. Stop down 4 stops to Zone 1. Your test is supposed to find the minimum exposure to produce Zone 1 (1 stop above base + fog) on the negative. It seems to me, and I am probably wrong, that closing down 5 stops to Zone 0 would be the same as base + fog.
3. Pull the darkslide out halfway. That way you will have an exposed half and an unexposed half of the negative. Cheaper and easier than drilling holes in dark slides. 8x10 dark slides are expensive.
4. Start the test at Box Speed (400) + 2/3 stop. Looking at my Weston Master V the progression would be:
650
500
400
320
250
200
Open +1/3 stop for each exposure, 6 in all. That should be enough to find "your" correct film speed for the film-developer-time-temperature used in the test.
5. Develop all sheets at the same time in the developer you will be using after the test. Use either the film makers time or the developer makers time for the test.
6. Evaluate to find Zone 1.

That's what I'm getting ready to do. Probably this afternoon before Hurricane Ike gets here.

1. Pick an evenly llluminated surface.
2. Six exposures in 1/3 stop increments from ASA 650 to ASA 200. OOPS! ISO.
3. Develop in Xtol 1:3 in a Jobo 3010 tank on a one way Beseler motor base at 30+ RPM. Time of 12 minutes 15 seconds at 68F or 20C from the German Xtol data sheet for rotary processing.
4. I'm planning to contact Richard Ritter and send the negatives to hime for evaluation.

Good luck!

Darren Kruger

12-Sep-2008, 12:03

3. Pull the darkslide out halfway. That way you will have an exposed half and an unexposed half of the negative. Cheaper and easier than drilling holes in dark slides. 8x10 dark slides are expensive.

that will work for the first exposure. but what about the second? How would you mask the part that you just exposed to light?

For testing 4x5 film in the past, I have used a darkslide that is mostly cut in half. Looks like a big L. It would produce two panoramic images on the same 4x5, with around 8mm of blank film between the image parts. It allows for two tests (images) per sheet plus an area to read the base fog in between.

Another option for testing film that doesn't involve cutting up a darkslide is using a Stouffer Step Wedge. There is a thread on APUG (http://www.apug.org/forums/forum216/34420-simple-step-wedge-testing.html) that describes one method for using it.

-Darren

kev curry

12-Sep-2008, 12:47

Jack there's a very simple and easy way of obtaining your Personal Film Speed and Development Time's in the free section of ''View Camera Magazine''. Its a great way of testing. It might be worth a read. It doesn't have to be complicated:-)

http://www.viewcamera.com/archives.html

kev

venchka

12-Sep-2008, 13:07

From the article quoted above:

Now, pull the darkslide out only half
way. You want to expose only half of the film and leave
the other have unexposed. After partially exposing this
sheet, do the samewith another sheet,

3 holders. 6 sheets of film. 6 half negatives exposed.

I based my plan of attack on Steve's article. However, I think my method of starting at ISO 650 through ISO 200 is easier for my old brain to understand than using percentages of the box speed.

After partially exposing this
sheet, do the samewith another sheet, but at a film speed
of 75% of the recommended exposure index. Now, do a
third sheet at a film speed of 50% of the manufacturer’s
recommendation. Now, do a fourth sheet with a speed
of125% of the manufacturer’s recommendation, and
then a fifth sheet with a speed of 150% of the manufacturer’s
recommendation. The sixth sheet can remain unexposed.

400 x 0.75 = 300 Not on my meter. Using 320 & 250 instead.
400 x 1.5 = 600 Not on my meter. Using 650 instead.

jack_hui

12-Sep-2008, 17:27

Dear all,

Thanks you very much for your kind comment, I will re-construct my test.

Jack

jetcode

13-Sep-2008, 01:03

I use a special card for calibrating exposure and development times or working ISO. I use something similar to the following:

http://www.colorhq.com/product-p/50103.htm

1) setup in desired light scenario.
2) meter the subject and determine target exposure
3) shoot 3 sheets at 3 different ISO's
4) develop each individually noting ISO and development time used
5) tweak development time according to exposure and intended dynamic range

the result is exposure/process time that is tuned to the light source and intended dynamic range for a particular film

ElrodCod

20-Sep-2008, 07:41

Jack,
Make it easy on yourself & get a copy of Fred Picker's book The Zone VI Workshop and follow the directions.

Ken Lee

20-Sep-2008, 08:46

<begin controversy>
I'm in that book, and in my very humble opinion, if your thermometer / water / chemistry / agitation / shutters / light meter / lens etc. are all within normal parameters, you can shoot whatever film you like, at 1/2 the stated speed, and you will be very close to what you will discover after all the testing is done.

One can state it in reverse: If you find that your personal film speed is at variance with 1/2 the manufacturer's number, this suggests that there is something unusual about one of those factors.
<end controversy>

Kevin Crisp

20-Sep-2008, 11:09

I don't think it ends the controversy. I use a Pentax digital meter calibrated by both Calumet and Mr. Ritter. In the 19 years I've been shooting sheet film, I have never had a test on any film give me a result less than the manufacturer's number. My first test of Tri-X mailed to Fred Picker for a densitometer reading, came back with a note of 320. Tmax 100 tests at 125 for me. I follow the Picker method as improved in his newsletter where he said, essentially, "if I did revise the book the film test would read like this..."

So do the test right, don't try to reinvent it or improve upon it, and go take pictures. Your results may well be half of the manufacturer's number. If you cut the manufacturer's rating in half and then nail down a development time appropriate for a Zone VIII density you can go take pictures and move on from what should be a very preliminary thing to get out of the way. Good luck and keep at it.

Ken Lee

20-Sep-2008, 11:30

Please overlook my lame sense of humor. I didn't mean that my silly pronouncement would end the controversy. I was pointing out, in a kind of geeky way, that the controversy will continue.

Aside from minor variations from batch to batch, the speed of the film is the only constant in the equation. It is we who introduce all the variables: our water and chemistry, our thermometer, the color of our daylight or light source, our sense of what constitutes normal agitation, our lenses, etc.

If we all test the same film, what we are really uncovering, are the peculiarities of our methodologies.

As a chemist, if I determine that water boils at 500 degrees Fahrenheit, any other chemist would suggest that there is either something contaminating my water sample, or perhaps I am not at sea level, or perhaps my thermometer needs calibration. Water is water, after all. My imperfect technique, can be compared against the perfect constancy of Water.

If my "personal" speed for TMax is 1600, then perhaps my grey card has faded, and needs replacing, if you know what I mean. It's not a judgement, it's just Chemistry. Or as they said in The Godfather: "It's nothing personal: it's just business". :rolleyes:

Hector.Navarro

20-Sep-2008, 11:36

Since you have received no real answers, I will weigh in with my observation that your approach should get you the data you require, but that you will probably find that film is much less expensive than all of those dark slides that you plan to mutilate.

Hello Jack,

totally agree with this comment, but I would add this:

have you considered instead of mutilating your dark slides, just pulling them half way out of the holder so that you can get one half film exposure and the other half unexposed?

Hector

ElrodCod

21-Sep-2008, 16:25

Aside from minor variations from batch to batch, the speed of the film is the only constant in the equation. It is we who introduce all the variables: our water and chemistry, our thermometer, the color of our daylight or light source, our sense of what constitutes normal agitation, our lenses, etc.

If we all test the same film, what we are really uncovering, are the peculiarities of our methodologies.

If my "personal" speed for TMax is 1600, then perhaps my grey card has faded, and needs replacing, if you know what I mean. It's not a judgement, it's just Chemistry. Or as they said in The Godfather: "It's nothing personal: it's just business". :rolleyes:

One of the variables that you haven't mentioned is shutter speeds. Have you tested yours? Don't be surprised if you find them plus or minus 25% from the marked speed.

jack_hui

22-Sep-2008, 19:37

Hi all,

I didn't check with this post for awhile. Well, whatever, here is my test result, done
completely by myself (but I set the test ISO to 400).

http://www.pbase.com/jack_hui/image/103523914.jpg

I found that my ND would be ISO320 and 5:15min (HC110E), N+1 would be ISO500 and 9:00 min (HC110E), N-1 would be ISO250 and 4:00 (HC110E).

Is the result reasonable for you guys?? I have shot some real pictures by using this
setting, by looking at the negative, I have no idea whether it is good or not ....
But it is quite fun to do B&W as compared with E-6 ....

I might probably print it out by paper next week and see ....

Good day!
Jack

Ken Lee

22-Sep-2008, 19:56

"One of the variables that you haven't mentioned is shutter speeds. Have you tested yours? Don't be surprised if you find them plus or minus 25% from the marked speed".

You're right - excellent point - particularly if when testing, exposure is controlled by changing the shutter speed, rather than the aperture. If the speeds are off, but one doesn't know it, then the whole test is "of questionable value", especially if we plan to use a variety of lenses and shutters.

I have tested my shutters, using a Calumet Shutter Tester (http://www.kenleegallery.com/html/shuttertester/manual.html).

weasel

23-Sep-2008, 03:17

Your results look reasonable to me.
When I do testing of a new combo of film and developer, I figure that what I am testing more than anything else is my entire process. If my results are wildly different than the manufactures' recommendations, chances are I have screwed something up.

Stephen Benskin

29-Sep-2008, 17:34

Jack,

Since you appear to be more advanced, you might find this interesting. Any speed determination using a fixed density point where the development parameters don't adhere to the ISO standard will not yield accurate results. This is a somewhat academic argument especially when you consider all the various influences to film speed in actual shooting situations. Still, there exists a fundamental misunderstanding on what film speed is and how it works.

If you're interested, check out these papers:

Jones, L.A., and Russell, M.E., Minimum Useful Gradient as a Criterion of Photographic Speed, JOSA, Vol. 25, Dec 1935.

Jones, L.A., The Evaluation of Negative Film Speeds in Terms of Print Quality, Journal of the Franklin Institute, Vol. 227, No3, March 1939.

Jones, L.A. and Nelson, C.N., A Study of Various Sensitometric Criteria of Negative Film Speeds, JOSA, Vol. 30, Number 3, March 1940.

Nelson, C.N. and Simonds, J.L., Simple Methods for Approximating the Fractional Gradient Speeds of Photographic Materials, JOSA, Vol. 46, Number 5, May 1956.

Nelson, C.N., Safety Factors in Camera Exposures, Photographic Science and Engineering, Vol. 4, Number 1, Jan-Feb 1960.

All the above should be in your local university library. Because of the advent of digital, some may no longer be found in the stacks. You'll have to ask about deep storage.

And

My paper Delta X Criterion (unpublished) which can be found over at Apug.

jack_hui

29-Sep-2008, 19:14

Stephen,

I am not as advance as you throught! hahah!!

But anyway, thanks for you info, I might try to look for the paper online.

Thanks
Jack

ic-racer

29-Sep-2008, 20:42

Still, there exists a fundamental misunderstanding on what film speed is and how it works.

My theory is that these 'zone systems' that exhibit a varying film speed with changes in development cause OVEREXPOSURE. And (as clearly seen on the posted graphs) the OVEREXPOSURE LATITUDE is great (the shoulders are not seen in any of the curves). Therefore, the user sees NO ILL EFFECT from following these systems and there is no Darwinian selection advantage of the more complicated speed systems (like 0.3 G etc.) :)

Again, this is just my crazy THEORY as to why that mindset (ie Film Speed changing with development) persists.

C. D. Keth

29-Sep-2008, 22:02

Stephen Benskin

30-Sep-2008, 00:00

Stephen,

I am not as advance as you throught! hahah!!

But anyway, thanks for you info, I might try to look for the paper online.

Thanks
Jack

Jack,

No problem there. On another note. I was looking over the graph. Don't forget to incorporate flare into the LER (log exposure range) which corresponds to the luminance range of the subject. In other words, the 7 stop range on the graph (LER) really represents more of an 8 stop range when shooting (LSLR - Log Subject Luminance Range).

The basic equation is Aim Contrast Index = Desired negative density range (or simply DR) / LSLR - flare. It's the minus flare part causes all the fun. For more on that, there's a paper hidden somewhere at APUG entitled "What is Normal". Do a search and it might pop up. I think it's in the B&W: Film, Paper, Chemistry section.

Steve

Stephen Benskin

30-Sep-2008, 01:47

My theory is that these 'zone systems' that exhibit a varying film speed with changes in development cause OVEREXPOSURE. And (as clearly seen on the posted graphs) the OVEREXPOSURE LATITUDE is great (the shoulders are not seen in any of the curves). Therefore, the user sees NO ILL EFFECT from following these systems and there is no Darwinian selection advantage of the more complicated speed systems (like 0.3 G etc.) :)

Again, this is just my crazy THEORY as to why that mindset (ie Film Speed changing with development) persists.

I did say it was mostly academic. A disposable camera makes acceptable negatives the majority of the time. From that stand point why discuss anything having to do with photography, except maybe some might want to know why a disposable camera can make okay negatives most of the time. Truly, any reasonably logical system works fine under most conditions. It's the extreme conditions that make the difference. It's just that for some reason I find photographic theory interesting. Sometimes I'll post something just in case someone else might find theory interesting too.

BTW, the Delta-X Criterion was designed to make the fractional gradient method more user friendly. In addition, the fixed density method in the ISO B&W negative film speed standard uses the Delta-X Criterion in it's contrast parameters. We are basically using the criterion without having to use the formula when determined under the stated conditions. And since Delta-X is in essence the fractional gradient method, the fractional gradient method is still alive and well.

Again, what does it matter when I could just go get a disposable camera? Maybe knowing something about what exposure is can save time and some of the frustration laboriously following and blindly believing in some test that on paper looks more like a cooking recipe and falsely claims to find "real" or "personal" film speeds. I know I would have preferred someone explaining variability in a system, normal distribution, and standard deviation along with the ZS stuff when I first started. It would have saved me time stressing over a 0.09 or 0.11 result. I also would have appreciated knowing there was more than the ZS method out there. I can't remember how many times I've heard some ZS student claim that AA invented sensitometry, or the gray card, or forced Kodak or meter manufacturers to change something. Let's not forget about the diabolical K-factor conspiracy AA talks about in this book.

Phil Davis used to tell a story about a photographer who had recently moved and was setting up a new darkroom. He was frustrated because he couldn't seem to duplicate the test results he had obtained in his old darkroom. He carefully re-examined every aspect of the process, taking extensive notes and dutifully logging every result. After weeks of exhaustive and tedious work, he concluded that the table that he uses to processes on was five inches too low. Could understanding a little more theory have helped this guy?

It's been said that Bruce Barnbaum likes his negatives so dense that if dropped, they would dent the floor, and I don't think anyone can question the quality of his prints. That's the beauty of large format. Because enlargements aren't as great as the smaller formats, may of the undesirable effects of denser negatives are inconsequential.

As for the various Zone System methods of film speed determination, most do result in a denser negative. This is caused by two factors. First, the difference between the speed point and the metered point on a curve is 1.0 logs or 3 1/3 stops. ZS uses a four stop range. Second, ZS testing doesn't account for flare. Interesting enough, the results produced will be close to the Pre 1960 ASA standard which was before there was an adjustment in the safety factor (post 1960 every film doubled its film speed).

As for calculating film speed from a fixed density point on a curve, development over normal will result in underexposure (I don't really like using the terms under and over when referring to exposure) while development under normal will result in overexposure. I know it sounds counter intuitive, but when considering the method that best defines exposure is based on local contrast and not a point of density, it makes more sense.

Personally, I use a calibrated EG&G sensitometer with a calibrated step tablet and plot it all with a program I wrote. Even with that, believe it or not, I had this very heated discussion with a non PhD in Kodak's research department about the inaccuracies of an intermittent type sensitometer compared to consistent light source and rotating exposure wedge. I felt that my intermittent EG&G gave me consistent enough results for my needs, he didn't. Of course, Kodak's densitometers read to 4 or 5 decimal places.

So, whether it's a disposable camera or a densitometer that can read to 0.0000 accuracy, the question comes down to what you want out of it, what you're happy with, and or course, the image.

ic-racer

30-Sep-2008, 07:20

BTW, the Delta-X Criterion was designed to make the fractional gradient method more user friendly.

Sounds interesting, do you have a link to the text?

Personally, I use a calibrated EG&G sensitometer with a calibrated step tablet and plot it all with a program I wrote.

I have an EG&G and a white-light Wejex. One of these days I'm going to post a shoot-out comparison between the two. Do you describe you program in you paper?

So, whether it's a disposable camera or a densitometer that can read to 0.0000 accuracy, the question comes down to what you want out of it, what you're happy with, and or course, the image.
Agree

Ken Lee

30-Sep-2008, 09:19

Stephen's article can be found here:

http://www.apug.org/forums/forum37/44881-what-normal-article.html

Stephen Benskin

30-Sep-2008, 12:12

I have an EG&G and a white-light Wejex. One of these days I'm going to post a shoot-out comparison between the two. Do you describe you program in you paper?

There are a few programs actually. I have one for data entry and curve analysis for film and one for paper. I wasn't happy with the Phil Davis plotter. There is too many BTZS hardwired functions in it. I wanted something that was more strictly sensitometry based. There's a four quadrant reproduction curve program, which I found key to research. Just getting it to work right with all the minutia forced me to have to nail down certain concepts in my head. It's like you think you know how something works until you have to teach it. There's also a process control program.

Thing is, I'm not great at math and probably the world's worst programmer. I learned to program in order to make the plotting programs. So, they're definitely not robust. In fact, the four quadrant has never left the Visual Basic environment. It's now in kind of a mess since I change computers and forgot about 80% of what little I knew about programming.

I've attached two examples from the Reproduction Curve program. One is the basic four quadrant curve for normal conditions. The other is a screen shot of the program which illustrates how it's more for research than show.

I attempted to attach the What is Normal? article here but it exceeded the max upload size. Sorry.

Here is the link:

http://www.apug.org/forums/forum37/44881-what-normal-article.html

Stephen Benskin

30-Sep-2008, 12:40

This is a continuation of the Reproduction Curve Program post. I had to break it up for the uploads.

The attached images are examples of some of the analysis I've done using the program. Two use only the reproduction quadrant which compares the values from the original subject to the print values. One deals with the choice of exposure placement in printing for a negative that's been over processed. The other one illustrates the resulting tonal shifts that occur when developing the same scene to different contrast Indexes and then matching the negative to the appropriate paper grade. Each print would have almost identical highlight and shadow placement.

The other example uses the camera image curve and the film curve. All the arrows and explanations were added later in Photoshop, but this is the example on why ZS testing produces EIs lower than the ISO. How many times have we heard someone claiming that the true film speed is 1/2 stop or so slower than the manufacturer's ISO? How many times are the manufacturers blamed for it? I was only able to discover the answer by seeing how the different elements interact using the reproduction curve program.

Most exposure/development/printing systems/methods tend to compartmentalize each step making it harder to understand of what is implicitly interconnected. Look again at the reproduction curve example of matching different CIs to different paper grades. There's differences in the reproduction of local contrast. Different parts of the same scene will be reproduced differently depending on some basic choices. For example, a normal CI on a grade two produces the greatest separation in the lighter tones while a lower CI printed on a higher contrast paper has greater separation of the shadows. Knowing this might allow greater creative choices in the field. In addition, the lower CI/higher grade examples lighten the reproduction of the mid-tone as well as the relative lightness of all the upper tones. The trade off is that it then darkness the relative darker tones as compared to the relationship the tones have in the original subject.

Kirk Keyes

30-Sep-2008, 13:06

Stephen - your 4-quadrant plot is beautiful. Your prograpm really has come a long way in the last couple years. I still think there's a market for something like this.

Kirk

Stephen Benskin

30-Sep-2008, 16:17

Stephen - your 4-quadrant plot is beautiful. Your prograpm really has come a long way in the last couple years. I still think there's a market for something like this.

Kirk

Hey Kirk,

How have you been?

As you've no doubt know, my motivation is sporadic and my math skills pathetic. Besides, after the total lack of interest from magazines in anything of technical substance that isn't about Photoshop, I've kind of stopped putting any more effort into that stuff and am just focusing on my own photography. But with the economy and my marketing skills the way they are it's hard to distinguish the two from each other. The last couple of days I've have an urge to get chatty for some reason.

Did you and Sandy ever finish that project that we started to work on?

Steve

Jim Noel

30-Sep-2008, 17:49

There is only one flaw in your process - Your exposure reading should be based on the box speed of the film -400, not 200.

I use the same process every time I get a box of film with a different emulsion number except that I use a blank wall on the north side of the house.

This system works well and is probably the most economical method.

Stephen Benskin

3-Oct-2008, 16:12

There is a tenant in testing that the results must adhere to real world results. This can frequently be difficult. The new film standard (1993) replaced the older standard for just this reason. The old standard had a developer that wasn't available on the markets. It was found to approximate many of the commonly used commercial developers. Then came T-grain films. Most people wouldn't have noticed or don't remember, but when the T-Max films first came out, they didn't have ISOs. They had EIs. The ISO designation indicates the manufacturer adhered to the ISO standard specifications. Kodak found the ISO developer didn't accurately produce the speed values obtained with many commercially available developers. The standard was changed allowing any developer to be used as long as the developer type is indicated (which few manufacturers do).

The specification for ISO exposure says to use a non-intermittent sensitometer. How can a sensitometer obtain results similar to real world usage? Truth is, it can't. Certain general assumptions must be applied which are derived from statistically average conditions. Together, the testing results closely reflect real world shooting results. Why not just test under real world conditions? In a word, variations.

Controlling the variables is key to accurate test results. Let's look at the typical ZS testing and some of the variables. These are just off the top of my head. The list isn't meant to be at all inclusive or complete.

The test target: Through some testing under identical conditions, I found that a white and gray card produces results consistent to predictions and reflective of sensitometric tests. A black card would produce a density placement lower than predicted. The materials from the test target play a factor. What wavelengths does the target reflect? Infrared is a concern. Many meter's photocells are sensitive to infrared. This can influence the meter reading. Ever wonder why infrared film doesn't have an ISO? While infrared film is sensitive to the visible spectrum, it's intended use is to record beyond the visible spectrum. The ISO film speed standard tests only for the sensitivity of the film under visible light.

The surround: The area surround the test target will influence the results. The ZS indicates that the frame is filled by the test, but most lenses "see" beyond just what is cropped in the image frame. Flare plays a factor. Eighty percent of camera flare is produced by the subject. It's not just about the sun hitting the lens. What are the values of the wall you stick the target on? The target itself can also produce flare even if it's solid. This is one reason why I found a white target preferable to a gray. The lighter tones are less affected by flare.

Color temperature: Under what lighting conditions are you testing under. ISO says daylight color temperature. Problem with shooting in the sun is specular reflections from the surface with can influence the density readings. The problem with shooting in the shade is the higher color temp. The film's spectral sensitivity and the light meter's spectral sensitivity are two variables to consider. Let's say the meter's photo cell is very sensitive to blue light, but the film has a lower than average sensitivity to blue. How will that affect the results?

Filters: While not a factor in testing, many of the photographers who do ZS testing shoot landscapes or exteriors of some kind. In many cases, this means the use of filters. The very nature of a filter influencing the placement of various tones along the characteristic curve brings into question the need to test the film beyond the manufacturer's indicated ISO speeds.

Influence of variables from limited testing: Most ZS practitioners will do one set of tests for a given film. The ISO film speed is determined from multiple tests with different batches over a period of time. This limits the influence of variables.

Hold Time: This is closely related to reciprocity failure. In the period of time after an exposure is made and the film is processed, there is a tendency for the silver to release electrons which occasionally causes a developable silver grain to revert back to an undevelopable silver halide grain. This is most common with the grains that receive the least amount of exposure energy, the shadows. The result is speed loss. Most of the loss occurs within seconds after exposure. The rate of speed loss eventually levels off and stabilized.

The old standard had a hold time of two hours. While this was a significantly long enough for most film to stabilize, the new standard made changes. It even makes a distinguish between professional films and amateur films. Why? Pros tend to develop the film within a shorter period after shooting it. The hold times are now something like one to two days for professional films and seven days for amateur films.

How many people who do their own testing maintain a consistent hold period. I've actually never seen a hold time even mentioned on any testing procedure. Hold time consistency is particularly important for those who test each new batch.

Accuracy and repeatability of equipment: This has been cover thoroughly else where. I only want to add that accuracy and repeatability are key reasons why testing equipment is so expensive. Since camera equipment's primary function isn't for testing, it's only as a good as it has to be for the production of images.

Development: As with everything, there is variability to every action. Nothing can be perfectly repeated. The degree of development will influence the density of any given point in relationship to a set amount of exposure. Sensitometric exposure makes it possible to determine contrast and speed from a single exposure and with a limited amount of film. A single exposure eliminates many of the variables, but the length of the film isn't something most people think about.

No matter what system of development is used, different points along the length of the film will receive different degrees of development. For example, a dip-and-dunk processor has the film hanging in a large tank. Film agitation comes from a burst of nitrogen through holes in tubes at the bottom of the tank. The film closest to the burst tubes receives greater agitation than that at the top. Also, the film toward the bottom is in physical contact with the developer longer as it is the first part that's first in and last out. For pictorial purposes, this generally doesn't matter, but it's another variable that can have an accumulative influence in testing.

Most people who go through the ZS testing don't ever test again for a given film, but those who retest new batches, generally only do the speed part. Sensitometric testing always confirms the degree of development. The ISO speed standard indicates that the contrast conditions must first be determined before speed can be determined.

Method of evaluation: While this isn't about variables in a system, it deserves a mention here. Determining speed using a fixed density method is inaccurate. So is determining speed using just black print method print, as is almost every other method to one degree or another under certain circumstances and conditions. This brings up the question just careful do you need to be in testing when the method of evaluating the test results is questionable?

It's up to the individual to determine what is acceptable for themselves. Just keep in mind the limitations and degree of accuracy and limit the variables as much as possible. Also, remember the potential for obtaining inaccurate film speeds caused by testing error and variability in testing conditions can be greater than any variation in film speed resulting from a specific film/developer combination or different batch.

Ken Lee

3-Oct-2008, 17:29

Stephen -

"...this is the example on why ZS testing produces EIs lower than the ISO. How many times have we heard someone claiming that the true film speed is 1/2 stop or so slower than the manufacturer's ISO? How many times are the manufacturers blamed for it? I was only able to discover the answer by seeing how the different elements interact using the reproduction curve program."

I confess: I'm one of those people :)

I'd love to get to the bottom of the question, and sincerely appreciate all the postings which have shed light on this question. Yours are articulate, and you cover all the bases.

You mention that you have discovered the answer. What is the answer ?

Stephen Benskin

3-Oct-2008, 23:48

Ken,

I wrote an article that appeared in the Jan/Feb 2005 issue of PHOTO Techniques titled "Flare and Accurate Film Speeds." The original manuscript was 7000 words. I had to cut it down to 2000 for the article. I had to cut out a lot of details including most of the testing data. I think it really hurt the article. Plus, I don't think the people at the magazine really understood what the article was about anyway. Mike Johnston would have.

I don't have a pdf version on hand at the moment to upload, but the two graphs below basically describe the concept. The captions are included on this post. I'll be happy to answer any questions you might have or help explain what's going on in the graphs.

BTW, notice the exposure for the camera image? 1/125 at f/16 - Sunny 16. If you're interested we can cover the exposure formula and compare the actual exposure numbers flare and non-flare.

Steve

Graph 2.
The right quadrant shows how the 7 1/3-stop scene luminance range is distributed along the flare curve. 18% reflectance falls a little over ½-stop higher on the curve than the 12% equivalent reflectance of the exposure meter calibration point.

Under normal flare conditions, the resulting mid-tone and shadow densities on Plus-X developed to a CI 0.62 should fall at 0.63 and 0.12 over Fb+f respectively. Without flare, they will fall at 0.61 and 0.05 over Fb+f.

The shadow exposure in Zone System method of speed determination falls four stops down from the exposure meter reading. If no flare exists, the shadow exposure will fall 0.23 log-H units below the speed point. This would indicate a film speed 2/3-stop slower than in actuality leading to a 2/3-stop over-exposure.

Graph 3.

In this example, the results from a standard Zone System speed test with the film set at the recommended ISO will produce a negative density of 0.05 over Fb+f. In order to produce the desired 0.10 over Fb+f, the film’s EI rating needs to be adjusted by 2/3 stop. The roll of Plus-X has an ISO of 125. According to the Zone System test, the film’s EI falls at 80.

In actual shooting conditions, flare has a greater influence in the shadow exposure than the middle or highlight exposure. So, while the Plus-X will produce a density of 0.10 over Fb+f in the test, it will produce a density of approximately 0.19 over Fb+f under real world average flare conditions.

Ken Lee

4-Oct-2008, 04:48

Steve -

Thanks for your patience.

Is the following statement correct ?

Films are capable of their official speeds in the laboratory - but in real-world practice, flare "fogs" the film, by around 2/3 stop. To get sufficient shadow details, we need to increase exposure by around 2/3 of a stop.

Stephen Benskin

4-Oct-2008, 07:07

Ken,

Kind of the complete opposite. The ISO film speed standard has flare incorporated into it. Zone System testing does not. If you test under a low to no flare condition and don't compensate for it in the speed calculation, then the indicated film speed will be lower. Flare actually adds a approximately 0.34 to 0.40 average flare to the exposure.

Flare also makes it impossible to be absolutely certain the shadow exposure in any given situation. High flare situations will produce greater shadow density and low flare situations will result in lower than expected shadow density.

The 2/3 stop discrepancy comes from the difference in stops ZS uses between the meter reading and the difference in the ISO standard. The difference between the average shadow and the point of the meter reading is 4 1/3 stops. The statistically average LSLR is 2.2 or 7 1/3 stop range. In many cases, the 1/3 stop is dropped, and most of the time it doesn't matter. In testing it does. Notice that 4 1/3 difference minus a 1 1/4 stop comes out to around 3 1/3 stops between the meter reading point and the speed point. This is difference is indicated on graph 2 as Delta 1.0. This is the how flare is automatically incorporated into the ISO standard.

So, exposure theory has a difference of 4 1/3 stops and ZS has 4 stops. That's a third of a stop difference in what they consider the shadow placement. In the graph, for the sake of simplicity, I used a stop flare factor. If the flare factor is 1 stop and the Zone System shorts the delta shadow by 1/3 stop, that leaves a discrepancy of 2/3 stop between the two systems.

For the article, I exposed both sensitometric and ZS tests and processed them together. The data from the tests supports the theory illustrated in the curves.

Another point to think about when thinking about film speed is that the speed point isn't necessarily the aim point for the shadow placement. It only means that it is the point where the film speed is calculated.

Steve

Stephen Benskin

6-Oct-2008, 01:12

I was just over at Apug.com looking through the article section and discovered they are currently updating that section and many articles aren't available. I've tried to upload the Delta-X article here, but it's too large even as a pdf. Anyone have any ideas?

Delta-X is a way of using the fixed density method of determining film speed and making it as accurate as the Fractional Gradient Method (0.3G).

venchka

6-Oct-2008, 09:20

Thanks one and all and especially Stephen Benskin for this information.

After wading through most of it, I have to ask myself, "Self, given all the variables, why on earth would you pretend to know how to conduct a test to find your 'personal film speed'?"

I was doing ok with Ilford Hp5+ at box speed in Xtol 1:3. Then I thought I ran a test that would tell me that my personal film speed for this film+developer combination was 250. Maybe it is. Maybe it isn't.

Dazed and confused in East Texas.

Ken Lee

6-Oct-2008, 11:05

Part of the problem, is that people tend to describe things in terms of f/stops, zones, and log values, interchangeably.

This adds to the mental gymnastics required.

venchka

6-Oct-2008, 11:16

It just gets worser and worser. What about different shutters, apertures, etc.? No two shutters are ever going to be EXACTLY alike at every speed.

Given that I have yet to overexpose a shadow and given that I can muddle through to a development time that doesn't totally blow out all the highlights, I think I'll trust my "test" and be blissfully ignorant exposing HP5+ at an ASA (ok, ISO) of 250.

That's my story and I'm sticking to it.

Ken Lee

6-Oct-2008, 11:37

From what I have been able to grasp from this thread, that number sounds right.

Back around 1970, when I first started studying from Fred Picker (when he lived in White Plains NY), I tested Tri-X in HC-110 dilution B. I got the same result he did, which was 250. Fred told me that Ansel used the same number too. After all, it was Ansel from whom Fred got the idea of using that film/developer combination, and from that identical result, Fred took some comfort that his methodology was sound.

According to Stephen's analysis, the numbers are basically off by 2/3 of a stop. That means we should lower our number by 2/3 of 1/2, or 1/3. Or, we should adopt a speed which is 2/3 of the stated speed.

Two-thirds of 400 = 266. which is so close to 250, that I doubt anyone could tell the difference except in a testing laboratory. Even the stated speed of 400, sounds like it may have been rounded off, for marketing purposes. After all, would you buy a film if the speed were stated as 392 ? Four-hundred is a nice round number. :rolleyes:

venchka

6-Oct-2008, 11:39

Maybe there is hope for me yet! Thanks Ken!

Stephen Benskin

6-Oct-2008, 14:01

Ken,

Sorry, you still have it backwards. Simply put, the manufacturers are correct. There is this universal myth derived from a conceptual misunderstanding that most film speeds are 1/2 stop slower. This is incorrect. Much of the myth can be tracked back to Adams. I know he has a reputation as a technical guy, and he was compared to most photographers (not scientists) at the time, but on most of the critical points, he was simply wrong. Anyone with a basic understanding of sensitometry can go through The Negative and find factual or conceptual errors on practically every page. Picker was worse. He was little more than a charlatan.

Thing is, it's no big deal. And that's my point. With all the variables in testing and in shooting, how accurate does the film speed setting have to be? People who are shooting at 1/2 the ISO still produce excellent images. The best example is that prior to 1960, the ASAs were around a stop slower for all films and quality was just fine. "Close enough" should be the catch phrase. I'm just saying be aware of this. Don't waste too much time with unreliable testing. For me, the EI isn't as important as the contrast. (venchka, just for the record, the correct term for rating the film at a setting other than the ISO is Exposure Index (EI). If you had testing equipment accurate enough to determine the speed under your specific conditions, it would be Effective Film Speed (EFS).) You also have to consider that there is a difference between film speed and exposure. Knowing what the film's speed is and deciding where you want to place the exposure aren't the same.

In the attachments below there are two three quadrant examples. One shows accurate exposure and the resulting placement of print tones (graph 5). One shows the technically over-exposed Zone System placement on the film's curve (graph 6). But unless you are strictly adhere to the just black method of printing, you will simple print for the shadow or "print down" the extra density, as was done here. Nothing wrong with that. In fact, it's a smart practical approach. Notice how there is almost no difference in tonal placement between the two examples? I believe it's just better to understand what you are actually doing and how something actually works. After all, what difference did it make to people's lives when they thought the Sun revolved around the Earth? They still had to pay rent.

The graph also just happens to be a good example of while the light meter "reads" and the mid-tone exposure is 12% reflectance, the print value from that point can still be ~18% gray.

Here's something interesting to think about. A film might have an ISO of 400. 400 what?

Steve

venchka

6-Oct-2008, 14:06

...

"Close enough" should be the catch phrase.

Steve

Now there's a scientific principle I can understand and adhere to. I can do that. :cool:

Wayne

Ken Lee

6-Oct-2008, 16:32

Stephen -

Thanks again for your patience, for taking the time to explain.

I am perfectly willing to cast aside what Ansel and Fred recommended about this subject. We all have, as they say, feet of clay, and those guys were not saints. They were just guys. And as you point out, one f/stop is not that much, when tossed in among the countless other variables, and a forgiving film.

In an effort to understand the terminology of your charts, I made a quick review of the Phil Davis 2007 BTZS article which appeared in View Camera, and which is still available for reference here (http://www.viewcamera.com/pdf/2007/btzs.pdf). Even so, much of the terminology evades me, especially when it is presented as acronyms. But that's my shortcoming. I am happy to learn, and appreciate the privilege.

So this leaves me with 2 questions:

1) When conscientious people test popular films like TMY using the BTZS methods, do their results generally confirm the manufacturer's claimed effective film speeds - at least when dealing with a "standard" brightness range or SBR ? (I understand that EFS changes with development time, so it's a "moving target" as it were. That often struck me as a shortcoming of the Zone System).

2) Have people adapted BTZS for scanning of negatives ?

Stephen Benskin

6-Oct-2008, 20:11

So this leaves me with 2 questions:

1) When conscientious people test popular films like TMY using the BTZS methods, do their results generally confirm the manufacturer's claimed effective film speeds - at least when dealing with a "standard" brightness range or SBR ? (I understand that EFS changes with development time, so it's a "moving target" as it were. That often struck me as a shortcoming of the Zone System).

2) Have people adapted BTZS for scanning of negatives ?

Ken,

I hate to throw more terms at you, but Davis' use of SBRs is outdated. When he first started writing the BTZS book, SBR and the like terms were being phased out and replaced by international terms having to do with light and exposure. The new terms also had to be psychophysical and not psychological. Davis was aware of this and has a footnote in a later addition. He chose not to change it because SBR had become embedded in BTZS terminology and it was in the programs he wrote which he didn't want to change. The current term is Subject Luminance Range but since the acronym is SLR, many write it either as LR or LSLR (log subject luminance range).

1. In a word, no. In order to determine anything other than a films EI, you need a known light source. That means it has to be calibrated and repeatable. BTZS uses an enlarger in most cases. It also uses a fixed density method and the inaccuracies that accompany that; however, the use of curves is imminently more helpful in understanding the process. I'm not a fan of any "system" or "method" because they generally have to simplify which can lead to falsehoods or confusion. BTZS is fundamentally sensitometry and that makes it fundamentally good. Apart from the incident exposure part, I feel it's the best of the popular methods.

Because BTZS doesn't use a known light source, you assign a workable film speed to the curve that comes closest to the ISO conditions. The other curves will extrapolate the speed from there. This is a perfectly fine approach and I hope I have illustrated how determining "the" film speed isn't that critical as determining how you like to exposure.

Without calibration, a sensitometer is only good for repeatable results. Calibration involves doing a preventative maintenance check of the machine, but mostly it's about taking a very accurate light meter and reading the output of the bulb (which changes over its lifetime). When I first had EG&G calibrate my sensitometer, it cost around $300 and all I got back was basically a piece of paper. Later, they really didn't want to deal with calibrating a product they no longer sold, so the price rose to around $1200. I checked how much a meter would cost that could handle that degree of accuracy and it was about $4500.

2. I have no idea about BTZS and scanning. I've begun having my negs drum scanned and I haven't really changed anything.

Steve

Stephen Benskin

8-Oct-2008, 19:59

Ken,

Let me know if this doesn't make sense. The majority of the people who do speed testing using the Zone System method obtain results indicating the EI of the film to be 1/2 to 2/3 stop slower than the film's stated ISO speed. The consistency of the results indicates the test are being performing in a uniform and consistent manor, that film to film ISO stated speeds are consistent from manufacturer to manufacturer, that most general purpose chemistry, exposure meters, f/stops, and shutter speeds are fairly accurate and don't vary much between manufacturers and brands.

Now, assume that the basic assumption of the Zone System testing procedures is flawed and if followed tend to produce film speeds of 1/2 to 2/3 stops slower than the stated ISO speeds. The logical conclusion would then be that with general purpose developers, the indicated ISO speeds are accurate and correct.

Steve

ElrodCod

11-Oct-2008, 06:18

Ken Lee

11-Oct-2008, 07:36

I had a look at the 1974 edition of Fred's Zone VI Workshop (http://www.amazon.com/exec/obidos/tg/detail/-/0817405747/qid=1075561755/sr=1-2/ref=sr_1_2/002-7002570-7186453?v=glance&s=books). The methodology seems reasonable to me, as it did around 35 years ago. In a nutshell, you make contact prints from a variety of film speed/development times, giving just enough exposure to render the film edge at maximum black.

You don't really need a gray card. As I recall, Fred used a horse with shaggy hair, and a caucasian model, in his book. The contact prints tell the story rather convincingly, and cold-light enlargements match the prints. Strictly speaking, no sensitometer is required either. I've never owned one, and never saw the need, since the empirical approach is rather direct and compelling.

While I respect the sensitometric approach, and the ernest scientific method behind it, I find it noteworthy, that even with the latest calibrated equipment and software, one is still obliged to determine one's own personal working parameters. If my personal speed were 1600 instead of 400, I would be worried - but I feel safe in leaving the sensitometry to the erudite professionals who work for the manufacturers.

Which means that we all have to follow this where it leads us. I'm grateful that the forum provides a place where we can discuss all this.

And thanks again Stephen for taking the time to explain things so patiently and thoroughly !

Stephen Benskin

11-Oct-2008, 19:04

Factual and conceptual errors aside; Ansel's and Fred's prints look pretty good. Your statements might be more meaningful if you posted some of your pictures instead of graphs. As the "charlatan" often said..."show me a print."

Elrodcod,

I could upload a print, but that won't conclude anything. If you go back and read the "close enough" post of mine, you will better understand what I am talking about. I'm not saying that you can't achieve a quality image with many different techniques. In fact, it's been more than proven that you can. This is not what I was trying to discuss.

My point is that while you can obtain quality images under most average condition using almost any system or no system at all (disposable cameras), it is in the extreme conditions where the differences or problems will show. As I said, for many people this is will be more of an academic discussion.

My other goal was to make people aware that Zone System testing doesn't give you the true film speed, that you really don't have to do extensive speed testing, and it's nice to be aware of the many variables.

Steve

ic-racer

12-Oct-2008, 08:06

Controlling the variables is key to accurate test results. Let's look at the typical ZS testing and some of the variables. These are just off the top of my head. The list isn't meant to be at all inclusive or complete.

I think about those errors all the time, but I have no way to determine the standard deviation produced by them. In theory, I like to think that I have my EI at a spot about two standard deviations from the minimum exposure to produce an excellent print. This ensures me acceptable exposure in 98% of cases. (Probably the same as the 'safety factor' films used to have). I have never had a problem with over exposure. All the contemporary films I have tested have had the shoulder way, way out there.

Stephen Benskin

12-Oct-2008, 08:43

I think about those errors all the time, but I have no way to determine the standard deviation produced by them. In theory, I like to think that I have my EI at a spot about two standard deviations from the minimum exposure to produce an excellent print. This ensures me acceptable exposure in 98% of cases. (Probably the same as the 'safety factor' films used to have). I have never had a problem with over exposure. All the contemporary films I have tested have had the shoulder way, way out there.

Exactly, know the variables, understand the strengths and weakness, and make choices based upon knowledge.

Kirk Keyes

13-Oct-2008, 17:42

Factual and conceptual errors aside; Ansel's and Fred's prints look pretty good. Your statements might be more meaningful if you posted some of your pictures instead of graphs. As the "charlatan" often said..."show me a print."

Elrod, I own a couple of Stephen's prints, and they are top notch. I wish I was as good a photographer as he is. And you probably would too, if you'd seen his prints...

Stephen Benskin

14-Oct-2008, 01:21

Thanks for the support Kirk, but I maintain showing a print wouldn't prove anything pertinent to the discussion about exposure. All it would prove is if I was a good printer or not. For example, let's take a look at Ansel Adams' Moonrise Over Hernandez. It was by Adams' own account underexposed, but the print was beautiful. I personally like the prints he made from it during late 60s and early 70s. The prints from the 40s and 50s seemed a little weak to me, yet all of them made with the same negative.

As some famous physicist once said, "the universe is only knowable through math." Sensitometry isn't separate from photography, it is the quantification of photography. For those who don't know, Loyd Jones is responsible for much of tone reproduction theory and exposure theory that we use today. That includes speed determination, average luminance range of scenes, average daylight illuminance, the exposure formula, the relationship between the negative and print (grades), and much more.

One of the many disciplines in which Jones was considered an expert was Psychophysics. In regards to photography, it is how the brain perceives colors, tones, quality, etc. Between the invention of photography and Jones' testing, there were many methods used to determine film speed. Hurter and Driffield, the inventors of sensitometry, concluded that since tones reproduced more accurately on the straight line portion of the film's characteristic curve, speed should be determined there. This seems logical when based only from the scientific results. As we know today, that isn't the best way to calculate film speed. What psychophysics does is to connect the science with the art. What H&D didn't do and what Jones did was to explore the science behind what is perceived to be a quality good print?

In the late 1930s, Jones did a series of psychophysical tests. Most notably is what has become known as the first excellent print test. Prints are made from negatives of differing exposures. People where asked to organize the prints in order from worse to best. Jones found that after a point, people couldn't tell the difference. He chose the negative made with the minimum amount of exposure that produced a quality print because of the many advantages that come with it. He then reviewed the technical exposure data of the negative. Finally, different mathematical models were tested to determined which would reach the same conclusion as was made through the exhaustive testing with judges. The best model would be the one that came closest to the judged tests using the greatest variety of film types under the most diverse set of conditions.

Jones found that not only was the shadow of critical importance, but also the shadow contrast. The best prints were made when the contrast of the shadow was .3x the overall contrast of the film. This is basis of the Fractional Gradient Method. If you think about it, the absolutely best, most accurate way to determine film speed is to do all the extensive psychophysical testing. The Fractional Gradient Method comes in second. (An interesting subject of discussion for another time is how the parameters of the instructions to the judges affect the choices and thus the outcome.)

So, sensitometry isn't somehow separate from photography. It is photography. In fact, the Zone System is just a simplified version of sensitometry, while BTZS uses sensitometry more directly. I've taken it a little further for myself and have found it beneficial to be able to utilize a broad range of tools when shooting. When I shoot, I see Zones, curves (camera, film, paper, reproduction), and printing options, all of which help to create the image in my head.

Kirk Keyes

14-Oct-2008, 09:22

Thanks for the support Kirk, but I maintain showing a print wouldn't prove anything pertinent to the discussion about exposure.

You're right, but some people don't like to reason logically...

ic-racer

14-Oct-2008, 09:26

This is basis of the Fractional Gradient Method.

I have always been frustrated by the fact that I am sitting in front of a reasonably powerful computer, yet making it solve a dataset for 0.3 G (fractional gradient method) has proven cumbersome for me. I had software that would give me something like a 3rd or 4th order polynomial for the film curve but it was difficult to solve (with Excel spreadsheet) just for 0.1 above film base, let alone try to get it to find the intersection of 0.3xG and the toe! Did you ever play with this and get an easy way to just plug in the (x,y) dataset and have a program or spread sheet spit out the speed based on the 0.3*G, fractional gradient method?

Kirk Keyes

14-Oct-2008, 09:28

In theory, I like to think that I have my EI at a spot about two standard deviations from the minimum exposure to produce an excellent print. This ensures me acceptable exposure in 98% of cases.

Racer - could you describe how you calculate your safety margin? I'm curious as to how you determined it to be 2 std dev away from mimimum exposure.

Stephen Benskin

14-Oct-2008, 12:13

I have always been frustrated by the fact that I am sitting in front of a reasonably powerful computer, yet making it solve a dataset for 0.3 G (fractional gradient method) has proven cumbersome for me. I had software that would give me something like a 3rd or 4th order polynomial for the film curve but it was difficult to solve (with Excel spreadsheet) just for 0.1 above film base, let alone try to get it to find the intersection of 0.3xG and the toe! Did you ever play with this and get an easy way to just plug in the (x,y) dataset and have a program or spread sheet spit out the speed based on the 0.3*G, fractional gradient method?

I had the same problem with my plotting program. That's why the Delta-X Criterion is so beautiful. It allows for a fixed density method and simply adds onto it a equation. I've written a paper on this but it's too big to upload here. If anyone has any ideas, I would appreciate it, or I could e-mail a pdf copy to those interested.

Fact is, the ISO standard for B&W film speed utilizes the Delta-X Criterion, it's just hidden in the contrast parameters. Many people, including Phil Davis, has complained that the contrast conditions as stated in the standard are too high which may result in film speeds higher than real world working speeds. It's a valid concern if it relates to a fixed density method, but the ISO standard relates to the Fractional Gradient Method, and there film speeds don't really move much with changes to processing.

Below is a table comparing fixed density film speeds to Delta-X film speeds (0.3G).

Kirk Keyes

14-Oct-2008, 13:40

Did you ever play with this and get an easy way to just plug in the (x,y) dataset and have a program or spread sheet spit out the speed based on the 0.3*G, fractional gradient method?

I made a spreadsheet that pretty much replicated the BTZS method as described in the books, and I took the easy way out and simply used rise over run to find all the pionts (speed, dmin, dmax, density ranges). it was quick, and it matched what the book gave.

ic-racer

16-Oct-2008, 08:30

Racer - could you describe how you calculate your safety margin? I'm curious as to how you determined it to be 2 std dev away from mimimum exposure.

Well, all I know is the percentage of negatives exposed with adequate shadow detail (adequate exposure). So its working in reverse from a set of known good negatives. (similar to the Nelson, Jones literature). But instead of finding the exposure for the thinnest negative (ISO, ASA), I would be looking the exposure index with the "2 standard deviation from the mean" safety margin.

This is just a theoretical explaination or model of how I might get 98% success. One certainly could fill in the numbers based on copious exposure information. In my case I don't take many notes so all I know is an estimate of the percentage of adequately exposed negatives. I know the EI I set on the meter, but when I meter I usually add 'a little extra' depending on how I feel that day.

When I first saw Steve's paper, I saw the bell curve and thought he may have been working on the same model, but it was not the case.

I don't know the literature well enough to know if there is a similar model out there, but I suspect it has been done before.

http://img.photobucket.com/albums/v670/ic-racer/bell_curve.jpg

Stephen Benskin

16-Oct-2008, 11:52

Racer,

Do you have numbers for the mean and standard deviation and the information source?

Kirk Keyes

16-Oct-2008, 13:09

IC, Steve -

A Gaussian or Normal distribution (where the data conforms to a Bell Curve like IC has in his graph) will have the following numbers:

A range of +/- 1 std dev (or 1 sigma) means about 68% of all the results fall within that range.
A range of +/- 2 std dev (or 2 sigmas) means about 95% of all the results fall within that range.
A rannge of +/- 3 std dev (or 3 sigmas) means about 99% of all the results fall in that range.

If you are counting all the results on on side of the mean, like in IC's graph, you get:
1 std dev error => 84% above
2 std dev error => 98% above
3 std dev error => 99.9% above

If you give yourself enough over exposure that you don't have more than 2 sdt dev of your exposures outside of that range, then 98% of your exposures will not be underexposed.

The trick is being able to predict how much overexposure you need to have to make sure you meet that goal.

So this really can't be something that's tested or determined using film tests. It can only be determined by examining how many of your exposures are failures or successes based on your own exposures.

Kirk Keyes

16-Oct-2008, 13:11

I guess if one could measure how many stops off each negative that one makes is from the minimum exposure needed for a good negative, then it would be easy to do this with a little bit of simple statistics.

ic-racer

16-Oct-2008, 14:16

Again, the last model is just a model, I don't keep records to analyze my exposure info, but it would make an interesting paper if it had not already been done.

That statistical model was a refinement of my "Bullseye" model. The "Bullseye" model simply places the exposure latitude equally on either side of scene brightness range. Exposure is 'aimed' for the center of the H&D curve. In that more refined statistical model, exposure is 'aimed' for the peak of the bell curve. With the statistical model, depending on how 'good' you are with your technique, the bell curve can be narrow and the resulting exposure will be less than the "Bullseye" approach.

Of course, if you live in a laboratory, your bell curve will be so narrow as to make your EI equal ISO with that statistical model :)

Anyway here is the introduction to the "Bullseye"

This diagram is just to get ones bearings, and shows a "Conventiional" exposure based on ISO or any permutation of the Zone system. As you can see there is almost no underexposure latitude, but there is plenty of overexposure latitude.

http://img.photobucket.com/albums/v670/ic-racer/Min.jpg

ic-racer

16-Oct-2008, 14:29

So, here is the "Bullseye" method. Unlike the statistical model, it is easy to calculate. One can set the exposure index to put Zone V smack in the middle between the minimum point (ASA/ISO or 0.1 above film base) and the maximum point (D-max or 0.2 below D-max).

So, with this system, an ISO 400 film can be put in a non-metered classic camera and a exposure data sheet for ISO 100 film is used to guess at exposure (ie "sunny 16" f8 at 500th etc).

Again this would be a cool experiment to give the ISO 100 datasheet and camera loaded with modern ISO 400 film to a novice and compare "best prints" shown to a panel. The comparison would be to an accurate exposure with a spot meter by a "zonie" based on 0.1 above film base or based on the film's ISO.

Of course the drawback with this system is that some of the negatives will be dense, which could lead to difficulty printing with some enlargers or some papers.

http://img.photobucket.com/albums/v670/ic-racer/Bullseye.jpg

Stephen Benskin

16-Oct-2008, 18:44

Thanks for the information Kirk, you shouldn't have gone to the bother. I think I wasted your time because I probably didn't state my question properly. A normal distribution curve has to relate to something. For instance, the mean Subject Luminance Range is 2.2 with a standard deviation of .38. This determines the shape of the curve as well as the range of the samples. I was just wondering the parameters Racer was using is all.

I'm in agreement with Racer's slightly higher placement on the curve. The first excellent print test determined that exposure greater than that required to reach the first excellent print produced prints that were indistinguishable from the first excellent print. It is a sound principle of exposure. I do believe it falls under a different topic of discussion than speed determination which was the topic I was discussing.

I have a question in regards to the two graphs illustrating the principle. They demonstrate the principle well. I'm just wondering if the subject luminance range is just a relative example or if it is supposed to represent realistic values? I'm asking because it appears that the range is around 0.8 or so, and it definitely falls under 1.0. It's just that the average subject luminance range is 2.2 and even if you incorporate flare, it will still be around 1.8 log-H. Zone I - VIII should be 2.1 log-H and with flare 1.7 log-H.

Kirk Keyes

16-Oct-2008, 20:58

Steve - so if the mean LR is 2.2 and standard dev is 0.38. So I can do the math in my head, let's say that the std dev = 0.4, OK?

Like IC Racer pionted out, we only need to worry about scenes that have Lumanence ranges wider than normal, as subjects that have ranges narrower than normal will not be under or over exposed.

So with a mean of 2.2, then 1 std. dev covers up to a LR of 2.6. That means that 84% of all scenes have a LR of less than 2.6.
With 2 std dev, it covers up to a LR of 3.0, and 98% of all exposures will have a LR less than 3.0.
With 3 std dev, that covers up to a LR of 3.4, and 99.8% of all exposures will hve a LR less than 3.4.

So if you wanted to cover 98% of all exposures, then you need to have a film than has a relatively straight line section that will take an exposure range (LR) of 3.0. And that means you need to put the "bullseye" exposure point for the middle of the exposure at 1.5 above the minimum exposure value.

Not all films could handle this situation, and it gets complicated if one starts to do N-1 or N+1 developments, but then why would one, if they are trying to have a simple, non-complicated exposure system...

Kirk Keyes

16-Oct-2008, 21:01

I'm just wondering if the subject luminance range is just a relative example or if it is supposed to represent realistic values? I'm asking because it appears that the range is around 0.8 or so, and it definitely falls under 1.0.

I suspect he's simply reproducing some graphs with example photos that really were not intended to match, so the numbers look a little wacky when put side-by-side.

ic-racer

17-Oct-2008, 15:29

I'm just wondering if the subject luminance range is just a relative example or if it is supposed to represent realistic values?

I just copied and pasted an H&D chart from the internet, so the values don't match the little shaded graphic and should have probably been erased. (or, thanks to photoshop "What numbers?... ;) )

ic-racer

17-Oct-2008, 16:00

I'll give some background on what prompted me to come up with these models back in the 80s.

At the time, I was writing an image analysis program at the NIH for CAT scans. At the time the GE cat scan 'exposure' of the patient gathered 1024 gray levels from each area of the patient. The computer screen could only show 256 gray levels, so my gray level software allowed many ways to view the 1024 set. Here is a perfect example someone else wrote as a java applet.

http://www.emory.edu/CRL/abb/WindowLevel.3/chest2.html

So, I had a mindset that each exposure I made was to include the ENTIRE subject brightness range (as projected onto the film plane) and later, in the darkroom I would decide which 'gray level window' would show in the print, by using multigrade paper. Totally analogous to the way the CT scanner gathered all the gray level data and the CRT monitor showed it.

Although these technical details show the scientific basis of what is going on, the main thing FOR ME was the MINDSET that I WOULDN'T decide which values are, or are not, going to be in the print, at the time of exposure (opposite of zone system visualization). I wanted all tonal info from the scene at the time of exposure, (like a scientific instrument would record it), and later, I would decide which values will appear in the print.

Stephen Benskin

19-Oct-2008, 13:52

3312easy

22-Oct-2008, 01:27

It's too abstruseness

Ken Lee

22-Oct-2008, 03:14

"It's too abstruseness"

When experts chat directly with other experts, they often use a kind of informal language, which non-experts find hard to follow.

As a fellow non-expert, I find it hard to follow, but I'm sure that if they discover anything which will be helpful to the rest of us, they are good guys, and will share it and explain it patiently.

ic-racer

22-Oct-2008, 06:55

Racer,

I've attached a couple of examples which you might like better than the internet example. They are from Jack Dunn's The Exposure Manual, fourth edition. Dunn also wrote a paper for the Photogaphic Journal in 1963 titled Expose for the Middle Tones which succinctly states his view on this subject.

BTW, one of the graphs is for reversal film, but the over/under idea is basically the same. Also, I've attached the first excellent print test graphed results from the Loyd Jones series of tests.

I'd like to hear more about your Bullseye Method. What do you say to starting a new thread to discuss it further?

Exactly, it is information similar to that contained in those graphs that gave me the idea for the bullseye method. Especially the last one from Jones showing the plateau. Also, the "Maximum Exposure For Acceptable Print" value would be based on something like the example for reversal film. Like 0.2 below film base.

So the 'test' of the bullseye method would be exactly like Jone's paper, but continuing the exposure series to fill in the right-hand side of the plateau curve. Then one would place one's estimated exposures at the middle of the plateau.

I wrote up a little story showing how two students of a "Fred Picker" style zone system workshop based their zone system tests on both the toe and (serendipidously) the shoulder of the H&D curve. One student has no overexposure latitude, and the other has no underesxposure latitude. A third student could put his exposure index between the two and have the best of both worlds.

http://www.apug.org/forums/forum37/47878-mythbuster-you-have-expose-shadows-develop-highlights.html

Here is the explanation of what each student in the story did: http://www.apug.org/forums/forum37/47878-mythbuster-you-have-expose-shadows-develop-highlights-3.html#post596960

I thought it was pretty funny but I don't think anyone got it.

Stephen Benskin

27-Oct-2008, 11:36

Racer,

I hope you don't mind my asking so many questions. Your system sounds like it fits into my concept of close enough. I have a few more questions that should help to lock it down for me.

Do you use a reflected or incident light meter? Spot?
Do you take into consideration the luminance range when metering?
If so, do you average the luminance range and place the average at the Bullseye location or do you weight the exposure based on distribution of values?
Is the goal of the method to keep the upper and lower exposure points equal distance from the toe and the shoulder?
Is it accurate to characterize the Bullseye method as average metering plus two stops? If not, how does it differ?
You mention an advantage of using the straight-line portion of the curve in shadow reproduction, is this on any psychophysical evidence or conjecture?
Does the method include changes to development based on a given luminance range?
If so, are the exposures based on the characteristics of the target curve for that luminance range?

Ken Lee

27-Oct-2008, 12:29

I'm not a historian of film stocks, but in the early 20th century, film was slower. With large cameras and their larger lenses, depth of field was correspondingly narrower, and "f/64" exposures were longer.

Applying the principle of "Sunny 16" on a film whose speed is, say, 10, suggests roughly 1/8 at f/16, or 1/4 at f/22, 1/2 at 32, or... 1 second at f/64. And that's in full sun.

Perhaps this explains why Zone System shooters took for granted, as it were, the unspoken imperative to expose for the low values (IE just enough exposure to guarantee the low values): It tends towards the minimum exposure, giving you the flexibility to choose a faster shutter speed and/or smaller aperture. In other words, in a world where film speed was at a premium, it gave you the most film speed.

Stephen Benskin

27-Oct-2008, 15:28

Plus it minimizes grain and maximizes sharpness. Not that big of a deal with large format, but it became a growing concern with the increase use of the smaller formats. One of the reasons why the change in 1960.

ic-racer

27-Oct-2008, 15:52

[QUOTE]Racer,

I hope you don't mind my asking so many questions. Your system sounds like it fits into my concept of close enough. I have a few more questions that should help to lock it down for me.

Do you use a reflected or incident light meter? Spot?

The concept centers around guesstimation with hand-held cameras without meters, but would also benefit averaging reflected meters and built-in meters and 'automatic' settings on vintage cameras. As these are probably the most error prone methods of determining exposure.

Do you take into consideration the luminance range when metering?

No need to because I suspect (though only through other's tests) that the luminance range on the film plane in just about any pictorial scene will safely fit on T-max.

If so, do you average the luminance range and place the average at the Bullseye location or do you weight the exposure based on distribution of values?

Conceptually it works best for me to think of a middle value on which to base exposure letting the high-end and low-end fall where they may. Again, since its more of a 'guessing' model, if you were capable of measuring scene range etc, a conventional exposure model would give a thinner (maybe better??) negative.

Is the goal of the method to keep the upper and lower exposure points equal distance from the toe and the shoulder?
Exactly. But it is more to allow for error (sloppy photographer) because if you had the means to exactly center the exposure you may be better with a conventional method. So, the "bullseye" refers more to a 'large target' than an exacting technique.

Is it accurate to characterize the Bullseye method as average metering plus two stops? If not, how does it differ?

Yes. I came to that with some simple (perhaps too simple) calculations. Last time I made a full curve for T-max was when it first came out (I have not repeated it since, because it is a little tedious) and I got about 12-14 stops between the toe and shoulder. So if there are 5 stops from white paint to black paint and I photograph a painted barn with one side in the sun and the other side in the shade, the black-to-white ranges will overlap by about 3 stops (based on the Kodak exposure datasheet for open shade). This gives an average scene luminance range for this 'off the cuff' example as 8 stops. So 8 stops of scene (ignoring flare) minus 12 stops of film range gives 4 stops extra, which can be equally distributed on high and low end. Therefore, 'add two stops' works in this example.

You mention an advantage of using the straight-line portion of the curve in shadow reproduction, is this on any psychophysical evidence or conjecture?

Mostly conjecture. Perhaps some of the 'look' of classic analog photography is due to the shadow values falling on the toe, but if the toe is to pronounced I get a scooped mid thing in a print where blacks are black, whites are white and middle gray looks like zone IV (too dark). T-max film does not have much of a toe, so at the popular IE of 200, the majority of the scene is on the straight portion anyway (ignoring a hump that can show up sometimes near the high values but before the real shoulder). Again all conjecture and conceptual meandering. For example I think "Why punish the low values by relegating them to the nasty toe?" Also, conceptually, if you think about only using the straight portion of the HD curve, development and exposure are more clearly independent of each other. Whereas when the low values are compressed in the toe, overexposure increases the density range of the scene on the film slightly as the low zones spread out into the straight portion. The required decrease in development needed in that case tends to promote concepts of 'development/exposure' interactions.

Does the method include changes to development based on a given luminance range?

No, multcontrast printing is required.

My thought was the many great photographers from the 35mm/Leica era (Frank, Winogrand, Cartier Bresson, etc) did NOT use meters, but knew the film's exposure latitude well enough to guess and get it right the majority of the time. I figured (when T-max films came out) that it would be even easer to guess at exposures because of an extra stop or two of latitude those films could give (in theory, based on the fantastic long H&D curve).