Comparison of Reflectance from 18% and 12% Zone Models

[Chan Tran]

Ralph! What the Monitor column represents?

 

[Bill Burk]

I looked at the butkus site and read a few instruction manuals for early Weston meters. The manuals tell you to exclude the sky.

Now I have to concede a point.

I no longer think that averaging light meters were designed to read a statistically average scene brightness distribution.

If the arrow pointed to average scene brightness distribution like I have been saying all along -- you wouldn't have to exclude the sky. So I must be wrong.

It sounds more like the instruction manuals were written to explain how to use something in a way it wasn't originally designed.

Bill

I work for Kodak but the opinions and positions I take are not necessarily those of EKC. Especially on this thread.
---
p.s. It's amusing that Weston company proudly explains ASA standards for film speeds were based in part on their research, but the standards changed just enough that ASA speeds and Weston speeds don't match.
---
p.p.s. I would be very amused if it turned out ASA standard meter calibration point was based in part on Weston meter calibration research but changed 18% to 12%.

 

[dpgoldenberg]

I looked at the butkus site and read a few instruction manuals for early Weston meters. The manuals tell you to exclude the sky.

p.s. It's amusing that Weston company proudly explains ASA standards for film speeds were based in part on their research, but the standards changed just enough that ASA speeds and Weston speeds don't match.
---
p.p.s. I would be very amused if it turned out ASA standard meter calibration point was based in part on Weston meter calibration research but changed 18% to 12%.

On a historical note, there is a direct connection between the work at Weston and the Zone System. A couple of months ago, I posted a link to a set of articles by John Davenport, published in 1940, which Ansel Adams acknowledged (somewhat belatedly) as the basis for the ZS. Davenport, in turn, acknowledges "Mr. W.N Goodwin Jr. who showed that there is a mathematical relationship between scene brightness range, negative density devleopment, and negative density ranges of printing papers". Goodwin was, apparently, in charge of the sensitometry at Weston. For those who missed my original post, the articles can be found at:

http://www.archive.org/details/ConstantQualityPrints

One of the striking things is that Davenport devised clever overlays for the Weston dial, similar to those used by Adams, though not exactly the same.

David

 

[Stephen Benskin]

Bill, I believe this excerpt from Calibration Levels of Films and Exposure Devices by D. Connelly (interestingly from Sangamo Weston Limited) touches on what we both are talking about.

"The essential characteristics of a photograph is that it portrays a differing pattern of luminance comprising the object being photographed, but equation (3) which is intended to use in assessing the exposure required for this range of luminance uses a single value of luminance Lg. The validity of the exposure determination method must, therefore, depend upon the acceptability of the resulting photographs which are produced by substituting a single value of luminance in the determination of exposure to represent the multiplicity of values of luminance of the scene itself.

In practice exposure determined by this substitution is satisfactory only for what may be termed average scenes. Unusual distributions of luminance require special exposure assessment.

From the point of view of the film, satisfactory photography depends upon the proper location on its exposure density characteristic of the densities produced by the image illumination within the camera. The greatest and least significant luminance in the scene are required to cause exposure of the film within the usable part of its exposure density characteristic. This implies that the important characteristics of the luminance are:

1. the ratio of its maximum to its minimum value.
2. its absolute value (of maximum or minimum)

for the former determines whether or not the film can reproduce the contrast range of the scene and the latter determines the exposure time necessary to provide an exposure which will locate the brightest scale of the scene correctly relative to the film characteristic.

The question of what single measure of scene luminance can be made which will allow adjustment of a camera to give satisfactory reproduction of a given scene is settled by the devices available for exposure determination. Until the advent of the cadmium sulphide cell the small electrical output obtained from selenium cells resulted in exposure meters which received light from an extended area in order to provide the required electrical power to operate the device. These exposure meters make a measurement proportional to the total light received, but because the directional light sensitivity of any one exposure device is built into he device and therefore, does not vary, its indication can be taken as proportional to the arithmetic average luminance of the scene. The large majority of exposure determining devices make a measurement proportional to this arithmetical average and in consequence it is this quantity which is used in exposure determination, but because this quantity is no more closely related to the ratio of the extremes of luminance than is the arithmetic average, and moreover is not easily measured, it is not used in exposure determination for ordinary photography."

 

[Bill Burk]

Thanks Stephen,

The beginning of this paragraph makes sense

The large majority of exposure determining devices make a measurement proportional to this arithmetical average and in consequence it is this quantity which is used in exposure determination,

But the last part of the paragraph doesn't appear to say what it was leading up to

but because this quantity is no more closely related to the ratio of the extremes of luminance than is the arithmetic average, and moreover is not easily measured, it is not used in exposure determination for ordinary photography.

The way he was leading up and in context I think he was trying to say:

[Because] it is not easily measured, [Extremes of Luminance] is not used in exposure determination for ordinary photography.

 

[Stephen Benskin]

I've been reading and re-reading that paragraph all day, and I still don't have it nailed down.

 

[Diapositivo]

Somebody said in this forum, in another thread, that the 18% grey card is customarily used because it is used in the printing industry. Sample pictures are taken with an 18% grey card in the scene because this matches, or used to match well with certain procedures in use in the printing industry where 18% is the middle grey. So the 18% number is not originating from the photographic world but from the printing industry.

As it happens, an 18% grey card can be used by photographers as a proxy for an average scene. In theory the reflectance should be around 12% to match results from an incident light meter and in practice using the 18% card would lead to an under-exposure of around 0.5 EV. This is of no material consequence when using negative film. When using slide film, 0.5 of underexposure is what many people typically do (either as a safety margin against highlight blowing, or to increase saturation) and that is why nobody ever really felt the need to buy 12% grey card.

In some Kodak 18% grey card the instructions say that the exposure value obtained should be increased by 0.5 EV. In some other Kodak grey cards no such advice is present.

Less-than-ideal inclination between light source, grey card and light meter would further distort reading.

My definitive answer: do not use grey cards. I cannot figure a situation when a grey card is more practical than an incident light meter. For me it's either reflected light spot metering, or incident light metering. Tertium non datur. Grey cards request people to have 8 arms, to keep the grey card exactly at 45° between light source and lens, and to pay attention not to keep it too high or to low. If you do some tests with grey cards you see how easy it is to get different readings. Guessing would be more reliable when outdoors.

Regarding the 144% stuff, Ralph Lambrecht gave an answer in another thread, around one year ago. IIRC the answer revolved around the notion that both films and human vision are not linear in response, so that zone VIII does not correspond, on film, to 8 times more light than zone V.

Fabrizio

 

[markbarendt]

My definitive answer: do not use grey cards. I cannot figure a situation when a grey card is more practical than an incident light meter. For me it's either reflected light spot metering, or incident light metering. Tertium non datur. Grey cards request people to have 8 arms, to keep the grey card exactly at 45° between light source and lens, and to pay attention not to keep it too high or to low. If you do some tests with grey cards you see how easy it is to get different readings. Guessing would be more reliable when outdoors.

Regarding the 144% stuff, Ralph Lambrecht gave an answer in another thread, around one year ago. IIRC the answer revolved around the notion that both films and human vision are not linear in response, so that zone VIII does not correspond, on film, to 8 times more light than zone V.

Fabrizio

Well how about portraiture and having the sitter hold the card?

Don't get me wrong, I'm with you on the value and use of incident meters but a gray card or similar device can provide a nearly perfect reference point not only for shooting but also for printing when it is included in a reference shot.

This carry through of a reference point allows me to regularly peg my print exposures on the very first try using my PM2L, no guessing, no test strips; just a darn nice proof on the first try.

 

[Stephen Benskin]

more food for thought

I was looking at the numbers and noticed that they are for non-flare conditions, also considered to be testing conditions. I've attached an example of guideline data from my four quadrant program from a non-flare model and a model with flare added. While I don't have a perfect match, the non-flare results come very close to matching the chart from WBM.

 

[Stephen Benskin]

Somebody said in this forum, in another thread, that the 18% grey card is customarily used because it is used in the printing industry. Sample pictures are taken with an 18% grey card in the scene because this matches, or used to match well with certain procedures in use in the printing industry where 18% is the middle grey. So the 18% number is not originating from the photographic world but from the printing industry.

Regarding the 144% stuff, Ralph Lambrecht gave an answer in another thread, around one year ago. IIRC the answer revolved around the notion that both films and human vision are not linear in response, so that zone VIII does not correspond, on film, to 8 times more light than zone V.

The printing industry makes sense. I believe they would have gotten it from Munsell. Using the psychophysical middle gray as a reference guide works well. But did the printing industry influence the photo industry or is it just post hoc ergo propter hoc?

We're talking about the physical reality of the average scene luminance range. It's 2.20. If the meter's calibration point is at 12% of the total illuminance, then three stops up is 100% Reflectance. It's just math. 144% comes from the same conditions assuming the calibration point or metered exposure point is 18% of total illuminance.

Human vision doesn't come into play when we are talking about the negative density range as it isn't evaluated in those respects. It does; however, play a big part with the print. Now, if you are talking about how flare affects the upper range in the negative, it doesn't. See Quadrant 1 of any of my tone reproduction graphs. You can also check out the non-flare / flare data guide example I just upload in response to Ralph's chart. Look at the 4th column, H(mcs), and compare the values.

 

[MattKing]

So Stephen, would you recommend that I make my fortune by coming out with a line of 12% grey cards?

On a more serious note, thank-you for another fascinating and informative thread.

 

[Stephen Benskin]

So Stephen, would you recommend that I make my fortune by coming out with a line of 12% grey cards?

There’s no need for a 12% card (another lost fortune). And I’m not talking about how the card is held either.

No one ever said that the tones of the original subject have to be reproduced exactly 1 to 1 in the print. In fact, based on what is perceived as a quality print, you wouldn’t want them too. And who said the meter is designed to reproduce the values exactly anyway? What the meter does is to place the range of exposure on a desirable section of the film curve. We’ve already seen that the exposure from the subject can be placed higher up on the negative film curve than is indicated by the ISO speed without loss of quality, but the reason why these points weren’t chosen for the film speed standard had little to do with the reproduction of tones.

Then there is the film and paper characteristic curves. They can combine to shift the tones all over the place. Take a look at the second set of data in post #16. Zone V is based on 12% (RD 0.92), but because of the combination of the film and paper curve, the print RD is ~18%. Small differences in such things as paper LERs, film processing, paper exposure, and yes flare can easily produce a Zone V print reflection density of 0.74 from a Zone V 0.92 subject reflection density.

In addition, when we talk about the 7 1/3 stop average scene, it doesn’t include the “accent” luminances – speculars on one end and deep cracks and small crevices on the other. There’s always room around the edges for interpretation of tonal placement in the print. And added to that is that there isn’t a perfect relationship between the best print possible from a negative and the matching of the negative density range to the print LER. I’ve recently uploaded this example in another thread. It shows the exposure placement from negative with different negative density ranges against a grade 2 paper curve.



All of the prints from these negatives were considered to be of excellent quality. Subject choice, details, texture, local contrast, and tonal distribution all can have an effect on what is perceived as excellent. As you can see in the example, there can be a large margin of acceptable exposure placement in a print. Perhaps this latitude is the reason why there is still a debate over the issue.

 

[Bill Burk]

It all seemed so simple back in the day when dad and I made a copystand out of a sheet of plywood, clamped on a couple photofloods, screwed a close-up filter on the Spotmatic II and put a graycard on top of the book to set the exposure before shooting.

 

[Stephen Benskin]

It all seemed so simple back in the day when dad and I made a copystand out of a sheet of plywood, clamped on a couple photofloods, screwed a close-up filter on the Spotmatic II and put a graycard on top of the book to set the exposure before shooting.

And a lot of very smart people worked really hard to make it that way. "You push the button..."

BTW, there's more of that excerpt from Connelly's paper.

"Equally it should be noted that the geographic mean is a convenient measure for the sensitometric determination of film speed of reversal color films as will be shown later, and as such is of considerable importance. The use of the cadmium sulphide photo-resistive cell has made available much more electrical power than formerly and this, of course, appears as an immediate advantage; but some caution is necessary. If the greater power availability is used to reduce to a small value the area of the scene from which light is received, the measurement made is no longer of the average scene luminance which it has been stated above is that on which the photographic constant P, as defined later, is based. In this case, it would be necessary to know which part of the scene to select to obtain a measurement equal to the average, or alternatively, a measurement of either the maximum or the minimum luminance could be made, but then a different value of the constant P would have to be determined and used."

That last part is a little unsettling.

Bill, instead of eventually retyping the whole article, I've decided to upload a pdf version of it. Now you don't have to read it in installments.

 

[Bill Burk]

Found the omission that made this paragraph hard to grasp...

The large majority of exposure determining devices make a measurement proportional to this arithmetical average and in consequence it is this quantity which is used in exposure determination.

It may be noted here that it has been stated that the geometric average of the scene luminance is the better measure for the purpose of exposure determination

, but because this quantity is no more closely related to the ratio of the extremes of luminance than is the arithmetic average, and moreover is not easily measured, it is not used in exposure determination for ordinary photography.

 

[Bill Burk]

... the measurement made is no longer of the average scene luminance which it has been stated above is that on which the photographic constant P, as defined later, is based. In this case, it would be necessary to know which part of the scene to select to obtain a measurement equal to the average, or alternatively, a measurement of either the maximum or the minimum luminance could be made, but then a different value of the constant P would have to be determined and used.

I think this can be loosely translated... "If we make a spotmeter, then photographers will have to know where to point it."... "and if they point it at a shadow or a highlight, they'll have to know where to set the dial"

 

[Stephen Benskin]

I think this can be loosely translated... "If we make a spotmeter, then photographers will have to know where to point it."... "and if they point it at a shadow or a highlight, they'll have to know where to set the dial"

It's the different value of the constant P that makes me wonder, but the paper doesn't discuss spot meters any further. BTW, P is a big part of the answer in the thread about the relationship between film speed and the metered exposure point.