Why is Zone System EI often about half rated ISO/ASA?

markbarendt said:

For a guy that's not into geekery you sure spit out a lot of jargon and numbers.

That is meant in fun Rob.

There seems to be a "not seeing the forest for the trees" thing going on though. The big ideas seem to be getting lost in the focus on the minutia of the formulas.

There are two real world things a meter doesn't have when you first turn it on, what are they?

Click to expand...

Mark, I think what's missing is linking the various elements together. The photographic process doesn't work in isolation. We can focus on the details, but they eventually should relate to the bigger picture.

Bill Burk said:

If you are rounding, then it's pretty significant because you are taking us from 100 speed film to 80 speed film.

Click to expand...

Well I can only tell you what Minolta is calibrated to. You will have to ask Stephen about his 0.08 value.

All I know is that my calibration works and produces the patches by metering and placing on zones 0 thru 10 and it works exactly. Its a bit fiddly the first time but once you've done it once and graphed the resulting negs, then you know exactly what density you need for zone 10 and approx density you need for zone 1 and that means you can pretty much nail it without doing any printing the next time. The zone 1 density will vary according to film dev combo so its not fixed but is between about 0.08 and 0.15 (Ralph has quoted as high as 0.17 I think).

Putting it at a fixed 0.1 density is extremely difficult and virtually impossible because you would need to tweak development time and dilution to do so and that will also affect highlights so the target can really only be "as close as possible" erring on the high side if you haven't nailed it. And of course you can only adjust EI in third stops so the equipment doesn't provide for a greater level of accuracy than that anyway.

markbarendt said:

For a guy that's not into geekery you sure spit out a lot of jargon and numbers.

That is meant in fun Rob.

There seems to be a "not seeing the forest for the trees" thing going on though. The big ideas seem to be getting lost in the focus on the minutia of the formulas.

There are two real world things a meter doesn't have when you first turn it on, what are they?

Click to expand...

Well I've been suckered into using your language because you don't understand mine

Stephen Benskin said:

You didn't happen to mention he was talking about 100 speed. My impression was you were claiming 0.1 lxs was for all film speed. Now 0.1 and 0.08 are a third of a stop apart. But meter manufactures set Sx at 100, plus I can also see the tech thinking about the reversal speed equation, which is 10/Hr reducing the difference. We are actually in agreement on much of this. What I don't understand is why you were such an ass when I presented the same information earlier. Instead of quibbling over minor details, you dismissed most of it out of hand.

The standards tend to exclude most theory because it is superfluous to the purpose of the standard. The 1974 standard was special in that it had a very nice appendix that did explain a lot of theory but the standard also makes it clear the appendix falls outside of the standard. The ISO b&w speed standard doesn't mention the method for determining speed is actually the Delta-X Criterion which has it's own set of equations. It also doesn't say that if the development is above or below the parameters of the standard that the resulting speeds using the fixed density method would be progressively off. However it does use the Delta-X Criterion and is explained elsewhere. So just because it's not in the standard doesn't mean something isn't relevant.

Using the equation A2 / T = B*S / K, it's possible to find the calibration level of B. K has been found to be 1.16 cd/ft2 or 12.5 cd/m2. I've found that different units of measure tend to work better with certain equations. I'm going to use candela per square foot for what I hope will be obvious. Solving for B, the equation becomes (A2 / (T*S)) * K = B. Using the Sunny 16 rule with f/16 at 1/ISO, T and S = 1. The equation becomes 256 * 1.16 = B. B = 297 cd/ft2 or 3197 cd/m2.

To verify this, it can be plugged into the exposure equation )q*Lg)/A2 = Eg. Eg = 8. The constant q has been determined to equal 0.65.

(0.65 * 3197) / 162 = 8.11.

The value for I can also be determined in a similar manor.

Click to expand...

Well perhaps you can tell me what, if I decided to apply my brain to it, learning and understanding all these formulas and calibration methods would do to improve my photography? My instincts, having been there before, tell me absolutely nothing. I can already meter something and place it on my chosen zone and have proved it works via my zone patch test. I have already explained that for the vast majority of my shots my shadows will be shifted up my soft curve which is extremely easy to print. So what am I lacking and why would I want to waste my time on something which gives me nothing worth having?
As I said its a purely academic exercise and not a useful methodology for me. The far simpler methodology I already have works very well and especially for roll film. I just don't need all these formulas and charts to do photography. You seem to have a very big problem with that and between you and mark telling me I don't understand how a meter works I decided to make a point about it which I think I've convinced you of. i.e. they don't offset from 0.1 speed point.
I think we'll leave it at that...

Here's a diagram of the speed determination taken from Current Exposure Meter Technology paper that Stephen gave me...

...showing the relations of the exposure meter reading to the ASA speed point.



Now reversal film is calibrated to highlights and brought down. And maybe that's what you keep thinking about.

But black and white negative film is calibrated to the 0.10 density speed point and brought up.

H sub-m is the speed point. For 100 speed film the exposure required to get the film up to 0.10 density is, in meter candle seconds, 0.0080 mcs (-2.1 log mcs)

The meter is set to desire 10 times that 0.08 mcs (-1.1 log mcs) at the film plane.

Now why might your meter be calibrated to put 0.010 mcs (-1.0 log mcs) there for 100 speed film when that's the amount of light an 80 speed film needs... Perhaps the K factor is being used to put more light at the front of the lens, because some of that light will suffer transmission loss...

RobC said:

I think I've convinced you of. i.e. they don't offset from 0.1 speed point.

Click to expand...

This point of yours remains in dispute.

Here are some bullet points showing how the speed determination methods stack up:

From Todd/Zakia Photographic Sensitometry, under Film Speed and Spectral Sensitivity.

Threshold Speeds

1. H&D Speeds by 1890 Hurter and Driffield...
2. ASA, BSI, DIN and GOST speeds. All these methods are based on a small net density, i.e., a specified density above base plus fog.
{for ASA, E sub m is the 0.10 speed point, and the film speed is 0.8/E sub m}

3. Gradient methods... the 0.3 gradient

ASA Speeds for Color Negatives

The speed of a color negative film... is similar to that of a black and white negative film...

Midtone Speeds

The ASA Speed for a color reversal material is found by ... {basically the square root of the product of the highlight and shadow is E sub m and then the ASA film speed is 8/E sub m}

High Density Speeds
{in this category are microfilm, x-rays, positive films for pictorial use and litho films (though litho film has no standard) }

RobC said:

Well I've been suckered into using your language because you don't understand mine

Click to expand...

Well you are actually you are talking Stephen's and Bill's.

RobC said:

Well perhaps you can tell me what, if I decided to apply my brain to it, learning and understanding all these formulas and calibration methods would do to improve my photography?

Click to expand...

Now this could be a good discussion.

Suppose you load your camera with a 400 speed film and find yourself wanting to take some low-light hand-held shots.

Even though you've calibrated and found (for example, 200) your personal EI and a highlight reading, placed high, will give you a recommended exposure of 1/15 second at f/1.4 and that's the lens you have... You can use your knowledge and understanding to realize that you can get away with 1/60th second and probably come out with a superior picture because it will be less blurry.

I do that all the time... I set the meter at 250 which is my personal EI and work by Zone System, placing shadows, or placing palm of my hand on Zone VI or switch to incident mode.
Whenever the metering tells me I need a faster shutter speed... I immediately switch gears and recognize that I am really always getting a real 400 so I have that shutter speed in reserve with no loss in quality. If I need two shutter speeds, now I am pushing it and I usually stop there because I don't like pushing.

But I have no problem exposing 400 speed film at 400 when I need the speed. The rest of the time I am on Zone System-compatible speed of 250.

The next steps are where it starts to get fun. Suppose you become intrigued by the work of someone who uses film in an interesting way like William Mortensen...

Now he staged photographs and lit them with no more than two stops of subject luminance range. He developed his film to gamma infinity and the tones that he achieved in his portraits is amazing. You could immerse yourself in his anti-Adams rhetoric, or you could use some of this sensitometry to accomplish the same thing. That's one of the directions I plan to go. Gotta get me a couple of hot lights first...

and is that all on roll film or sheet film?

There's nothing stopping me using ISO speed for a roll of film if I want but I can't do it for 1 frame of 36 frame roll. But I can place any subject on any zone I like. I'm not forced to expose for the highlights for every shot.

Now what do you do when you are photographing at night and you have some bright lights and very deep shadows. What do you expose for and how do you work out your exposure.

The knowledge can be applied to both roll and sheet films...

I tend to shoot a whole roll or two at once. And I consider a lot of my film to be suitable to develop normally.

If the roll is full of Normal shots, I'll develop it to my normal contrast... 0.62 CI. If its from a foggy day, the whole roll would be that way anyway... I can develop longer.

I don't go around with 3 cameras, one with N and one with N-1 and one with N+1... I think that would be hard to manage. But on extended trips with 4x5 I will sort the sheets into separate boxes.

To your second question: I keep a copy of this chart handy. It's a chart that Michael Kenna gave to Vaughn, and he shared it here.

Look at Post 23

(there was a url link here which no longer exists)

Stephen Benskin said:

Now 0.1 and 0.08 are a third of a stop apart.

Click to expand...

How do you work that out?

RobC said:

How do you work that out?

Click to expand...

8/0.1 = 80
8/0.08 = 100

Reversal speed equation 10/HR
10/0.1 = 100 - probably what the Minolta tech was referring to.

RobC said:

and is that all on roll film or sheet film?

There's nothing stopping me using ISO speed for a roll of film if I want but I can't do it for 1 frame of 36 frame roll. But I can place any subject on any zone I like. I'm not forced to expose for the highlights for every shot.

Now what do you do when you are photographing at night and you have some bright lights and very deep shadows. What do you expose for and how do you work out your exposure.

Click to expand...

I need to pack my tripod, flashlight, lunasix and spot attachment and set it for a zone1 ISO reading... I point the spot at a zone 1 using the flashlight as needed push the reading rocker and read off the exposure using the flash light.

I need to remember to remove the uv filter and write POTA on the cassette and on an avery label on the camera.

All you would need too do is set a similar ISO on your spot to my lunasix.

Stephen Benskin said:

You're back to playing games so maybe we should.

Click to expand...

Cruella de Vielle see previous post...
Rob's scheme is ok for medium and low contrast he is keying to a high light...
Ansells scheme was uniform for any scene?

Xmas said:

I need to pack my tripod, flashlight, lunasix and spot attachment and set it for a zone1 ISO reading... I point the spot at a zone 1 using the flashlight as needed push the reading rocker and read off the exposure using the flash light.

I need to remember to remove the uv filter and write POTA on the cassette and on an avery label on the camera.

All you would need too do is set a similar ISO on your spot to my lunasix.

Click to expand...

I meter for a highlight and sod the shadows. Actually they usually turn out OK.

Fixating on a shadow is not a good thing to do at night.

In many subjects at night (not all ) you will have very deep dark shadows. It is night after all. But you will also have highlights to mid tones and often those mid tones and highlights DON'T fall into reciprocity. So if you meter a shadow and correct for reciprocity you push the mid tones and highlights way up the curve with a large amount of over exposure which then means you are into a rescue job to get them back.

Now ask yourself, in a night time shot which is more important, the mid tones to highlights or the deep dark shadows. I have found that becasue I have already calibrated for 10 stop range that I can meter what I want on zonne 7 or 8 and very often that won't be into reciprocity. Had I metered a shadow it would be in reciprocity and with the correction for that I would need to use your pota or some secondary calbration such as N- development to try and get the highlights back. It's a time wasting exercise because its often unnecessary.
And judging a shadow at night and trying to place on a set value is very difficult because at night for deep shadows you're perceptions of luminance change because of the lack of colour in large dark areas. That's just how your eyes work with the rods and cones. It can look darker than it actually is as far as the film is concerned because film doesn't have prceptual vion using rods and cones. Your meter reading may also vary a bit due to lack of colour but not inline with how your eyes work. The result is you can get strange looking results which don't look like night. Shadows way too light and highlights way above where they ought to be.
In other words expose for the highlights and just let the shadows go and that way you'll get a night time image which actually looks like night.
And infact the shadows will often turn out fine (providing you calibrated the way I do for 10 stops) but they won't turn out fine if you have used ISO film speed calibration, they'll be too dark (unless you apply reciprocity corrections which casue highlight problems).
If my "highlights" require reciprocity correction then I'll give them that using Ilfords reciprocity chart ( I use Ilford films) and that will lift the shadows too.

See: (there was a url link here which no longer exists)

A night time shot but it doesn't look natural.

Its no wonder why so many find night time photography so difficult when they are using various peoples reciprocity corrections which seem to be based around exposing for a shadow when exposing for highlight would very often produce a much more natural looking result.
It is night time after all and shadows should be deep and dark otherwise it doesn't look like night.

You forgot about my flash light if I want detail in a dark area that is where zone 1 is.
If it needs correction for reciprocity it gets correction.

I use POTA to limit burn of highlights, yes they are pigs to print.

I also pack an umbrella cause rescent rain leaves lots of reflections.

I can control where zone 1 is and I bracket to allow for reciprocity uncertainty.

The street lights produce gorgeous effects with my single coated lenses enhanced by the overexposure.

It only goes to show there's a 1001 one ways to skin a cat and each of us are always inventing new ways to do it. And our own tried and tested ways work for us.

Rob's scheme is ok for medium and low contrast he is keying to a high light...

Click to expand...

It works for moderate high contrast too since normal is supposed to be 7 1/3 stops and my system is calbrated for 10

i.e. its good for everything except very high contrast and usually I just don't bother with those becasue I know they'll be a complete bugger to print or I just let the shadows go.

I was browsing through the 1971 standard for exposure meters and rediscovered something. Appendix C is taken almost verbatim from Scudder, Nelson, and Stimson's, Re-evaluation of Factors Affecting Manual of Automatic Control of Camera Exposure. The 1971 standard is rather unique in my experience in that it has a number of appendixes explaining the theory supporting the standard. For the most part, standards are very concise and are not the best educational source. Subsection C1 speaks to the importance to know the ratio between the speed point and Hg.

"In order for a meter to be used to set a camera to the proper exposure, the following relationship is assumed to exist:

Eg / Em = constant"

For a 125 speed film:
Eg = 0.064
Em = 0.0064

0.64 / 0.0064 = 10

Another paper used as a reference in the appendix is Nelson's Safety Factors in Camera Exposures. Both Safety Factors and Re-evaluation of Factors Affecting Manual of Automatic Control of Camera Exposure are available in online Here.

In Connelly's paper Calibration Levels of Films and Exposure Devices writes, "it is evident that the relationship between the sensitometric measure of exposure, Hm(Hg) and the photographic exposure requirement Hg must be determined.

Hg / Hm = k1
Hg / HR = k2 "

k1 = 10
k2 = 0.8

Stephen Benskin said:

Hg / Hm = k1
Hg / HR = k2 "

k1 = 10
k2 = 0.8

Click to expand...

I can't get my head around... a 100 speed color reversal film surely is indicated to have the same exposure as a 100 speed black and white negative film.

Yet this implies a difference. How is this difference implemented?

Bill Burk said:

I can't get my head around... a 100 speed color reversal film surely is indicated to have the same exposure as a 100 speed black and white negative film.

Yet this implies a difference. How is this difference implemented?

Click to expand...

According to Connelly, "Film speed is a number which indicates the characteristic of light sensitive material by comparing the exposure required to produce certain effects on the film, with an arbitrarily chosen constant exposure." Knowing the ratio between the speed points and Hg is "the basis upon which calibration techniques of exposure devices are decided." "Since however an exposure device should give satisfactory results for both monochrome and reversal color films without change of calibration level, n1 * k1 and n2 * k2 must be equal. n1 is the b&w speed constant and n2 is the reversal speed constant. This way the same speed indication will be effective for films that have different film speed systems.

The ratio between the film speed point and Hg is defined by the speed constant. Hm is 10 times or Δ1.0 lo-H from Hg. Hg constant is 8 so 8/10 = 0.80. Black and white film speed equation is S = 0.80 / Hm. Reversal film is 8/0.8 = 10 and the reversal speed equation is S = 10 / HR.

I guess this brings us back to where this thread began with the difference between the metered exposure point (Hg) and b&w speed point (Hm) being 3 1/3 stops.

Is there any problem using an exposure meter to attempt to infer the illumination incident on the scene?

I would think that since the exposure meter is designed to plan exposure at the film plane, it has already been adjusted for the light loss through transmission through the lens. (K again).

Doesn't that make the conversion from meter reading in EV to light falling upon the scene (using an EV to candlepower chart) a little bit wrong?

Bill Burk said:

Is there any problem using an exposure meter to attempt to infer the illumination incident on the scene?

I would think that since the exposure meter is designed to plan exposure at the film plane, it has already been adjusted for the light loss through transmission through the lens. (K again).

Click to expand...

You are correct.

Doesn't that make the conversion from meter reading in EV to light falling upon the scene (using an EV to candlepower chart) a little bit wrong?

Click to expand...

Nothing in the exposure meter changes. Todd and Zakia in A Review of Speed Methods write "The speed of a photographic material is not a fundamental concept, but provides an index number useful for calculating camera settings. "Reversal film speed point has been measured at the same point on the curve but the speed constant changed changing the film speed. For instance, exposure at the same point HR = 0.064 will produce a film speed of 8/0.064 = 125 for the old equation and 10/0.064 = 156 for the new. This will place the exposure on a different spot on the film curve without changing anything with the meter.

I was thinking the problem of using an exposure meter to infer incident light... Because the light is falling on the subject in front of the lens and has some absolute value, while the meter is trying to compute what's behind the lens after some losses.

Stephen Benskin said:

Reversal film speed point has been measured at the same point on the curve but the speed constant changed changing the film speed. For instance, exposure at the same point HR = 0.064 will produce a film speed of 8/0.064 = 125 for the old equation and 10/0.064 = 156 for the new. This will place the exposure on a different spot on the film curve without changing anything with the meter.

Click to expand...

I expect this part happens behind the lens at the film plane. I'll try to figure out the difference 8 versus 10 as we go along... But I think it will help to know if we are talking about the film plane exposure.

Bill Burk said:

I was thinking the problem of using an exposure meter to infer incident light... Because the light is falling on the subject in front of the lens and has some absolute value, while the meter is trying to compute what's behind the lens after some losses.



I expect this part happens behind the lens at the film plane. I'll try to figure out the difference 8 versus 10 as we go along... But I think it will help to know if we are talking about the film plane exposure.

Click to expand...

You can determine the incident light value using a reflected light meter but the reflectance of the target value needs to be specific. I was talking about film plane exposure.

Stephen Benskin said:

You can determine the incident light value using a reflected light meter but the reflectance of the target value needs to be known.

Click to expand...

Probably would be difficult - I won't propose to do that since I have an incident meter.

But then incident meter applies C to give the film plane exposure recommendation.

When using an incident exposure meter to try to evaluate incident light... Do published charts that convert EV[100] to incident light values... back out C? If they back out C I think they would have a better chance of correctly telling how much light is falling on the scene.

Stephen Benskin said:

I was talking about film plane exposure.

Click to expand...

Or is EV[100] a film plane exposure that incorporates C... making those published charts slightly wrong?