What Makes Middle Gray, Middle Gray?

OK, Stephen, here is a list of naïve or very confusing questions form this thread, and the Sunny 16 thread, that made me initially question your knowledge of photography.

 

- where do you think the idea of middle gray comes from. In other words, what makes middle gray middle gray?

 

- What about reevaluating the timeless Sunny 16 Rule? Is it just a rule of thumb?

 

- Why use f/16 instead of another f/stop?

 

- Why use 1/ISO for the shutter speed?

 

- What is the relationship between the Sunny 16 Rule and the meter or the camera exposure?

 

- Is there any PROOF to support Sunny 16?

 

I must admit, that in light of the above questions, I probably overlooked some of your other of your comments that would have suggested you are a very knowledgeable photographer, and that you were just sandbagging to bait people into a discussion (I am not allowed to use the word trolling).

Stephen, lets try and do some deductive reasoning. Why is it called the 18% gray card? Nowhere in the photo world do we call middle gray as "18%," usually we refer to it as .75 reflection density. Given that printing was in existence far earlier than photography, I bet you someone, in a effort to quantify or calibrate the printing method, decided to print a series or progression of tones from paper white (it really does not matter what paper) to complete black and it just so happens that the "tone" that fell in the middle corresponded to a tone that reflected 18%. Since in the printing world they use percentages instead of reflection densities to describe the quality of the tone, I bet you the name stuck.

 

I dont think there is a "scientific" or methodological reason as to why the 18% gray card was chosen, most likely it was something carried over from the printing world that was useful in photography and it "stuck".

<i>where do you think the idea of middle gray comes from. In other words, what makes middle gray middle gray?</i>

<p>

Look at my proof. This is just a tiny bit of why middle gray is middle gray. If someone asked me that question. I would give them my proof. This question was preference by saying I was looking for a consensus. I already had a theory.

<p>

<i>What about reevaluating the timeless Sunny 16 Rule? Is it just a rule of thumb?</i>

<p>

This is from the other thread which really didn't go well because there wasn't enough substance in the responses to even begin to tease out a discussion and therefore create a learning experience. However, this is the same idea as the middle gray. What do people think about it? I'm asking people to reevaluate the idea. Notice, REEVALUATE, suggesting I had thought about it before.

<p>

<i/>Why use f/16 instead of another f/stop?</i>

<p>

This was to have people question the reason why f/16. There is a reason other than because it works. It's tied into the exposure and meter calibration formula. Maybe I could have given some information to better direct them.

<p>

<i>Why use 1/ISO for the shutter speed?</i>

<p>

This is also an example from the other thread. But here goes. The key relationships is that film exposure is about the reciprocal of the film speed. In the equation f/16^2 is equal to the theoretical average luminance in footlamberts. If the shutter speed is the same as 1/ISO then they cancel each other out. That means that f/16 becomes a constant that represents the exposure under average conditions when the exposure is for one-second (unity created by the canceling of the film speed and shutter speed). This constant is identical to the constant P in the exposure equation. P in the case of the average illumance is 8. The midtone exposure is defined as 8/Hg. Film speed relates to this because it is 1/10 the midtone exposure or .8/Hm. You can figure Hm by using the equation .8/ISO. It goes to figure then that you can find the midtone exposure with the equation 8/ISO. This can also be expressed as 8 * 1/ISO. Since P is a constant that 16 represents, you can simplify this to say f/16 at 1/ISO. By understanding the relationship, you can better understand the variables that go into the exposure. You can better understand the relationship between all the elements. If you get into the exposure equation further, you will find out more about the variable and assumptions of the conditions of exposure. To me, that means better control.

<p>

The posting of my idea on Sunny 16 that I promised will go into more detail on the above.

<p>

<i> What is the relationship between the Sunny 16 Rule and the meter or the camera exposure?<i/>

<p>

See above.

<p>

<i>Is there any PROOF to support Sunny 16?</i>

<p>

See above. The proof is in the interrelationship between the average illuminance, the exposure formula, and the meter calibration formula.

<p>

Maybe you and the others misinterpreted these questions as pretending ignorance. I didn't because I was hoping to make people think there might be a deeper meaning. It's the lack of awareness of the people that there is a deeper understanding that is truly at fault? It is simple seeing the same thing from different points of understanding.

<p>

Example: The world is flat. <i>Why is it flat?</i> What are you ignorant of something? I just told you it was flat.

<p>

I love that commercial were a couple are fighting over decorating their apartment. The guy says he doesn't want flashy colors. The women says she wants something flashy like beige. Kind of a men are form Mars, women are from Venus thing.

<I>I dont think there is a "scientific" or methodological reason as to why the 18% gray card was chosen, most likely it was something carried over from the printing world that was useful in photography and it "stuck".</i>

<p>

Jorge, there is, but it falls into the catagory of psychophysical / perceptual evaluation on middle gray. One of the problems I'm face with here is that I don't want to give the away everything from the entire series of articles. Be it said that it sort of begins with Munsell. The first values were perceptual. His son later related them pscyhophysical while keeping the Munsell nomenclature. Around the same time, CIE was established. Also around the same time the Journal of the Optical Society of America conviened the second Committee on Colorimetry chaired by....? (wait for it) Loyd Jones. I personally think the committee was conviened in response to CIE 1931(?).

<p>

Mechanical printing could use the original Munsell patches. After all they are designed to communicate color.

</i><p>

Stephen, considering all that's been said about feigning ignorance, I'm going to put all pretense aside. I are ignorant. All of what you posted about sunny-16 is good, and I'm sure it will make sense when I go back and sound out the words. Right now, I'm stuck on just one small part of it. And once I digest that, I'm sure it will all make sense.

<p>

"What? You mean the ISO speeds have some basis on physical reality?"

<p>

Do you feel like weeping at the futility? How do you discuss astrophysics with apes when they haven't considered what the visible horizon might mean? That the earth is or isn't flat is a step beyond even that.

Stephen, I glad you appreciated my input. I'm still following all this, but it seems like its gone in circles so many times I'm still processing it all before I even consider to add anything else. I don't really have the time to keep up with this, the come-and-go nature of my schedule makes it hard to follow a thread this active and intellectual.

 

Do you mean Fechners law? I think that can help in explaining the psychovisual aspects of what makes gray, gray. Our eyes do compress the darker tones of the scale, making us more sensitive to subtle changes in the highlights, this also has an effect of "pushing down" what we see as middle gray on the "absolute scale" if you could say there was one. :D

 

As far as Ansel Adams camping out at Kodak, I would have loved to see that... or at least have read Bob Shells article.

 

As far as flare factoring in to film speeds, Eek that reminds me of 4 quadrant graphs. I wouldn't say it causes you to determine film speed, but more defines how you use it. Take that as it is, I just jumped back in to this thread today more or less. I'll try to come up to speed, but in an hour I'm off to teach the current freshman some more about editing images on computers.

 

If I'm wrong about something, let me know. I'm always learning.

<i>How do you discuss astrophysics with apes when they

haven't considered what the visible horizon might mean? That

the earth is or isn't flat is a step beyond even that.</i><P>

I think Brad's simple question cuts to the heart of much of this

debating (well, the debating about the debate).<p>

An inorganic chemist probably understands the reactions that

make film work on a much more fundamental level than even an

experienced photographer and printer who consistently gets

excellent exposures and can easily make great prints. In the

context of photography, though, whose knowledge is more

substantial? At some point between "FP4+ exposed at EI 80 in

average conditions should be developed for . . ." and "a certain

quantity of energy from light cause the silver electrons to . . .", we

move from knowledge that better enables the photographer to

do what he needs to do to knowledge that's only usefully

important to someone designing new film emulsions.<P>

If your primary interest is in making better photos, you'll have a

different idea of what is "substantial" information and discussion

than someone who takes a serious interest in technical detail

and theory. The friction here seems to stem from differing views

over which sort of understanding is "superior."<p>

For me, which committee named which term 70 years ago is a

mildly-interesting bit of trivia. And to the ape, astrophysics is not

substantial--he wants to make the best of his life in the jungle.

Now where did I put that banana . . .

Exactly, Fechner and Weber. If you didn't have that cold, the discussion would have worked. Pressure.

 

I said flare is an important factor in the determination of film speed, not "causes you to determine film speed." Changes in flare do effect how film speed is used, but there is a 0.30 to 0.34 flare value that is incorporated into the calculation of film speed.

 

As for tone reproduction curves, they are really key to understanding the relationship of all the elements. Otherwise, it's the four blind men describing an elephant.<div></div>

<i>Jorge, there is, but it falls into the catagory of psychophysical / perceptual evaluation on middle gray. One of the problems I'm face with here is that I don't want to give the away everything from the entire series of articles. Be it said that it sort of begins with Munsell. The first values were perceptual. <p>

 

Jorge, I forgot something - 1/antilog(.74) = 18%.</i><p>

 

LOL...Stephen, your antilog message is funny, is like those people measuring the Egyptian pyramids and comming back with all kinds of relations. Perhaps the reason we are looking for a density of .75 is because someone took the log of 18..:-)<p>

 

I dont know about the psychophysical studies you mention, but it seems this falls more into anatomical studies done for other purposes than print reproduction. When you hear hooves you gotta think horses not zebras. I kind of got the feeling the reason we got stuck with the 18% value is very simple and is lost in the history or in the meshing of printint and photography. I have no doubt someone afterwards decided to make a study of this, but I just cant see someone printing in the 1700 or 1800's thinking about the psychophysical responses...or later on in photography in the early 1900's.<p>

 

Anyhow if you are not going to tell us why you think we got stuck with the 18% value then pooo on you...:-)

Mike D. I'm not into lenses. Some people are. To me, lenses are just a tool. If you want to know if they are working correctly, then do a bench test. I can't understand what people have to talk about. For me, it is not worth my time. That is why I don't participate in lens discussions. Why do some people feel they have the right to denigrate the interests of others especially if they don't have to be there? This is totally inappropriate behavior. If this thread doesn't interest you, don't be here. But to tell people they are wrong for being interested in something is rude and boorish.

 

You kind of make your own point, Mike. What is substantial to one isn't to another. If it isn't to you, you don't have to be here. You could be off eating bananas.

 

I still think it is a defense mechanism masking feels of insecurity.

<i>To me, lenses are just a tool. If you want to know if they are

working correctly, then do a bench test.</i><P>

LOL! Actually, I just go out and shoot a test roll of typical

subjects!<P>

But okay, okay, you win. It's your party--you get to say what kinds

of questioning are good and what kinds are bad. I'll just sit here

quietly pondering my insecurites while I scratch myself. (At least

I didn't throw any feces . . .)

Jorge, I'm sorry I thought I just missed answering something on your post. I know I didn't go into a great deal of information about Munsell and all, but it was an attempt to answer you. Munsell was a painter who taught. He wanted a way to communicate color to his students. So he worked out the Munsell system. CIE is somehow based on this system. Exactly how I'm not sure.

 

Study into how we see began in the 1700s, and appears to flourish in the later half of the 19th century. Sometime in the 1700s someone asked three artists to paint him a mid gray patch. He didn't specify the lighting conditions or anything else. They came back with samples that very closely matched. I think the mind might have a greater sensitivity to middle gray. It could be ingrained.

 

I know it is less likely to conflict with how tones near it are perceived. A light tone next to a dark tone will appear darker and visa-versa. There were tests to determine the background value to be used in the perception tests. You guessed it, middle gray worked best. Where middle gray exactly falls is determined later, but I think that we always had a general concept of what it should be. Along these lines, it wasn't until the Committee of Colorimetry released their report in 1940 or 41, that there was a consensus to what color actual is. The members fought over the issue as to whether color was a purely psychological phenomenon or is it purely physical. My hero, Lyod Jones, suggested a psychophysical approach.

 

The Munsell value scale was designed to range from black to white in perceptually equal spacing. Value 5 is the middle step. I believe it was originally equal to 19.4% reflectance, but Munsell didn't place reflection values on the patches. His son repeated the tests under stricter scientific conditions. His Value 5 was equal to 18% reflectance. BTW, there are 10 Values from black to white. Remind you of any other system, eh?

 

That's about what I've discovered is the basic visual aspect of middle gray.

This thread is so long that I would have thought the answer to be plain by now. What is the consensus? From my viewpoint, it is difficult to find in nature a substance that has a diffuse reflection density greater than 2 and less than .05. The eye is pretty well logarithmic in response, as is witnessed by the fact that a just noticeable difference is a constant ratio, not a constant difference, over much of the range of visual brightness. I would think it would be well to call a diffuse reflection density of 1 middle gray. If you ask someone to identify a gray patch on a white or black background as middle gray with no other reference, lots of luck.

Jorge, your e-mail address doesn't seem to be working. Do you have a new one?

You are absolutely right about the just noticeable difference. I believe it is about 2%, but the ratio is only valid when comparing one tone to another and not when the tones are in a series. Carl�s excellent post explains this well.

<p>

Your other assumption of middle gray appears to be based on the conditions found in the physical world. As I have proposed, there are a number of separate areas to consider and each have their own middle gray. Your value of RD 1 equals 10% reflectance. This correlates to Jack Dunn�s 9% value in his book <i><u>Exposure Manual</i></u>. Now that the physical world's middle gray is defined, there is the psychophysical middle gray, the camera image middle gray, and the exposure meter middle gray to consider.

<p>

BTW, there�s an interesting paper <i>XU, H., <u>Lightness and Reflectance of Munsell Gray Samples, Color Research and Application</u>, Vol. 18, No. 6, Dec. 1993, p. 422-425</i>. He reevaluates the just noticeable differences between the Munsell Values. He found a discrepancy between the numbers. By creating equal JND units between each step, he came up with 13.3% for Value 5. I don't think the Munsell Color Science Laboratories have adopted the recommendation.</i>

I shudder at the thought of making this thread even 1mm longer, but the topic of flare in the eye caught my interest. As we age, stuff (for lack of a medical description) accumulates in our eyes. Floaters become more common. I know that my ability to see into dark shadows next to bright lights isn't as good as when I was a teenager. Though I have no way to see or quantify it, I have to assume the same stuff that scatters light in my eye under those conditions, affects shadow contrast under more normal conditions. It also must affect my evaluation of prints, so maybe it all cancels out :-) I have to assume from this that maybe there's a perception difference between young and old, with regards to contrast and shadow detail. I've also read that people who have their lenses removed due to cataracts find a much bluer image with the replacement lenses.

I agree, its a psychovisual phenomenon. Our eyes are set up to separate the spectrum in to different "channels" which our brain interprets as color... driving a lot of the perception issues as well.

 

Without our perception and visual mechanism, color doesn't really exist, its just another wavelength.

 

As far as the chemical properties of film and how and why it does what it does, I actually understand a fair amount about that, I just really don't want to go there. :D I've had plenty of chemistry as well as just straight photographic chemistry, its incredible how film works, I think.

I don't know if JNDs would be equal between each step. I think I missed something in the discussion, where within the scale are these steps falling? JNDs at one level of intensity could vary signifigantly from what they would be at another, I thought.<br><br>

 

<i>Changes in flare do effect how film speed is used, but there is a 0.30 to 0.34 flare value that is incorporated into the calculation of film speed. </i><br>Very true, I forgot about that. This is built in to allow us to not worry about flare when using film, because its already assumed to be there at some degree or another. I wasn't so much thinking calculation as I was the absolute concept of whats being measured (films sensitivity) but now that I look at it, that really doesn't matter if we don't have any practical way to measure/label and use it (IE film speeds). Silly me.

That should be "significantly"

In astronomy and space imaging reflectance is known as albedo, and has been widely measured. Here is a NASA map of the earth's albedo as measured from space: <a href="http://www.gsfc.nasa.gov/gsfc/earth/pictures/20020624earthshine/modis_albedo_1440.gif">modis_albedo_1440.gif</a> (37KB).

<p>

Typical values range from 0.1 to 0.2 for vegetation, and 0.4+ for snowy arctic forests.

This thread is increasingly interesting.

 

Thanks to Mark and Michael for enlightening me on 'trolling'.

 

Stephen, sorry there's no ref. on eye flare -- I suspect it may have been in a recent BJP so I'll try to get hold of the author and ask if he has any refs. Like you, it's a subject I hadn't considered.

 

Conrad, a fascinating note on eye contrast decreasing with age -- I wonder if this relates to the fact that Brandt printed his pictures more and more contrasty as he got older. I've started another thread about this.

 

Lex, amen to your comment that AA and (worse still) his supporters were not averse to bullying and denigration: look at what they did to Mortensen. I didn't hear the 18 per cent gray story from Shell, but from another source whom I would regard as extremely reliable but I won't name him until I've checked HIS sources -- he's the kind of guy who will put up or shut up, and he told me the story as gossip, not gospel.

 

Thanks everyone

 

Cheers

 

Roger

Dear Mike (Dixon),

 

Sorry -- I meant to include this in my previous post and forgot. I suggest it matters a great deal which committee decided what, even 70 years ago, if you are trying (as Stephen is) to re-evaluate something. On what basis did they decide? Have circumstances (flare, ease of determining fractional gradients) changed since? In short, is it still good science, or folklore?

 

I think it was JFK who said "Those who do not study history are condemned to repeat it" (no doubt someone will correct me if I have misattributed this), but I'd suggest that those who do not study sensitometry are condemned to the Zone System. I'd be much obliged if the more enthusiastic devotees of the ZS take that remark as light-hearted, and not a killing insult.

 

Cheers,

 

Roger

Below is the proof for a theoretical no flare condition. It can also be applied as a proof for the physical world conditions. This graphic example represents approximately the same conditions Patrick described in his post. It also proves Jack Dunn's 9% reflectance figure.<div></div>

Although not a perfect example, I use 7 stops instead of 7 1/3, below is a proof of a possible 18% reflectance. Remember that the older uncoated lenses increased the flare factor by approximately one stop. So it is possible that at one time, the average reflectance could have been 18%. So, the argument as to 12% or 18% may be solved. Both are right depending on the time period to which you are referring.<div></div>

The way I see it is that the camera image-flare hypothesis appears correct because the flare model fits three different established conditions. There is the current 12% (ISO standard), the past 18% (if there was one), and the physical world 9% (Jack Dunn). I think this justification is correct and logical. Anyone else agree?