Zone System practice and expsure metering

[Doug Kerr]

Let me say at the outset that I have absolutely no experience in the use of the Zone System for exposure planning. What follows is an analysis on wholly "theoretical" terms. Its conclusions and implications might be misleading in the real world.

************

Many discussions of the use of the Zone System for exposure planning with digital cameras assume that, "for openers", a shot metered (let's say in spot mode) on a certain object in the scene will (if we don't in any way intervene) get a photographic exposure such that this object will have an exposure result we would consider "appropriate" for Zone V.

What is that likely to mean in technical terms?

Let me start with some stuff about exposure metering. Assume that we have a camera whose internal automatic exposure control metering system:

• works on the basis of measuring the average luminance of the scene.

Now, few cameras actually have that mode, but sit tight.​

• is "calibrated" consistent with the ISO standard for internal exposure metering systems.

That is defined in a way that one of the inputs to its "equation" is the ISO speed of the film or digital sensor.​

• feeds the "equation" in the automatic metering system not the ISO speed of the sensor but rather the ISO SOS value (which is about 1/2 stop less).

We might expect this for modern digital cameras, where typically the "sensitivity" is stated in terms of ISO SOS.​

Then, if we shot a test scene of uniform luminance, we might expect that the photometric exposure for that entire frame would be about 18% of the saturation photometric exposure.

Not 12.8%? No, that would be the result if we actually fed the "exposure equation" the ISO speed of the sensor, as intended.​

But we are interested in the metering system being used in, for example, its "spot" mode (not consistent with the first assumption above). What then?

Well, this might be different for other camera makes (in which all bets are off), but in Canon cameras, generally, if we aim the camera at a certain uniform-luminance test scene, the exposure that is set will be the same for any metering pattern, for example:

• Center-weighted average
• Partial
• Evaluative
• Spot

and maybe even point (I don't have any machines that have that).

So if that is the case, if, with the metering in spot mode, we meter on an object, the photographic exposure set by the camera will give that object a photometric exposure of 18% of saturation.

Depending on just what kind of tonal scale curves are built into the camera processing, that might be expected to give an RGB color whose luminance was about 18% of maximum (for a neutral scene, perhaps sRGB 118, 118, 118).

Is that "appropriate" for an object we consider deserving of "Zone V" treatment under Zone System practice?

Is that a good color for a "Zone V" gray cat? Who knows.

But the other issue is how will this work for the span of other zones. It is hard to speak of the lower zones, owing to differences in camera dynamic range.

But with regard to the upper end, in theory, if Zone V gets 18% of the saturation photometric exposure, then Zone VII would get about 1/2 stop short of saturation. Thus Zone VII is probably a good estimate of the highest zone we can safely accommodate.

But Zone System practice often assumes that we may have objects with significant texture in Zones II through VIII. And the basic assumption here will not safely accommodate that.

So if we do have objects we consider to be in Zone VII (and they in fact do have luminance two stops higher than objects we consider to in Zone V), we probably need to expose at least 1/2 stop, and more safely 1 stop, less than the metered exposure the camera would use after spot metering on our "Zone V". Of course this can be readily done by setting an exposure compensation of -0.5 EV or -1 Ev before we meter on out "Zone V" object.

On the other hand, some authors speak of Zone VII as the proper "rating" for a bright region for which we might accept rendering that is "a bit washed out". In that case, if the brightest region we have is "Zone VII" (under that outlook), then using the metered exposure as is for a Zone V object would seem to work out well.

"Well, I used to think this was a mid-sized car until I found that lovely parking spot marked "compact cars only".​

************

I would be anxious to learn if this wholly-theoretical outlook comports at all with actual experience on the part of those who use the Zone System in connection with digital cameras.

Best regards,

Doug

 

[Chris Calohan]

I've run my Pentax Spotmeter set to a Zone V exposure value against my Nikon d800 at
-0.3, -0.5, -0.7 and -1 EV using spot mode metering, manual exposure and found that in "normal" sunlight conditions, -0.7 is closest to the Spotmeter. However, I find I must go to -0.3 to get the same reading in less than desirable light (overcast). I am not sure what to attribute this to other than the D800's excellent low light sensor sensitivity.

 

[Doug Kerr]

Hi, Chris,

I've run my Pentax Spotmeter set to a Zone V exposure value against my Nikon d800 at
-0.3, -0.5, -0.7 and -1 EV using spot mode metering, manual exposure and found that in "normal" sunlight conditions, -0.7 is closest to the Spotmeter. However, I find I must go to -0.3 to get the same reading in less than desirable light (overcast). I am not sure what to attribute this to other than the D800's excellent low light sensor sensitivity.

Let me be sure I know what you mean. I take it to mean:

You make an exposure determination with your Pentax Spotmeter in the "default" way (which you describe as a "Zone V" measurement"), and note its photographic exposure recommendation.

But then, spot metering on the same area with your D800, you find that for it to set the same photographic exposure you must set an exposure compensation of:

• -0.7 Ev for "normal" sunlight, or

• -0.3 Ev for "overcast" light.

I'll proceed as if that is all correct.

The "excellent low-light sensitivity" of the camera should not affect the metering.

One possibility is that there is a difference in the chromatic response between the Spotmeter and the D800 metering system, such that difference in the color temperature between the two lighting situations could cause the discrepancy. (I have never heard that discussed, but . . .)

As to the discrepancy itself, my first thought was that that no doubt the Pentax Spotmeter expects the exposure index to be set as the ISO speed, whereas probably the D800 works on the basis of the ISO SOS (about an 0.5 stop lower number for the same sensitivity).

But in that case, given that we have fed the Spotmeter an "understated" exposure index (from its perspective), we would expect it to give a "too hot" photographic exposure recommendation. To match that result in the D800, we wouild need to apply an exposure compensation of perhaps +0.5.

But your experience is that an EC of -0.3 to -0.7 is needed to match.

So that theory doesn't seem to play out.

Another possibility is that for the Spotmeter, the "Zone V" reading is not what I would call the "default" reading. Or perhaps its "default" calibration, being for a "spot" situation, is not the same as the standard calibration for a meter to be used on a "scene average" basis. I have to look into that matter.

I will contemplate this further over breakfast!

Thanks for bringing this to my attention.

Thanks.

Best regards,

Doug

 

[Doug Kerr]

Hi, Chris,

I don't know what exact model of the Pentax Spotmeter you have, but I got the manual for the Spotmeter V and have reflected some on it.

Calibration

The ISO standard for free-standing exposure meters defines their "exposure equation" in terms of a "calibration constant", which for the reflected light (luminance) mode is symbolized K.

Keep in mind that this is all based on the concept of a meter that measures average scene luminance and derives a recommended photographic exposure from that.​

The larger the value of K, the larger will be the photographic exposure (Ev) for a given scene luminance. The standard allows for the meter manufacturer to choose the value of K to bring about the "exposure strategy" it feels is most useful, over the range 10.6 - 13.4.

A very common value of K is 12.5. (In fact, the American standard for exposure meters, which was superseded by the ISO standard, prescribed a value of 12.50.

My Miranda Cadius exposure meter (a golden oldy) has K = 12.5 (determined by reverse engineering from the table on the back that relates meter readings to luminance).

By the same reverse engineering process, based on a table in the Spotmeter V manual, it seems as if the intended value of K there is about 8.7.

That would mean that the photographic exposure recommended by that meter from spot metering on a scene area with a certain luminance would be about 0.50 stop less than for a meter with K = 12.5 measuring a scene whose average luminance is that certain illuminance.

Were it not for the presumed discrepancy between the exposure index as ISO speed and ISO SOS, this would fit well with the 0.3 to 0.7 stop discrepancy between the Spotmeter and the D800 (see more below).

Zone System operation

The Pentax Spotmeter V (as I see it in the manual) does not have a Zone System scale. It has something intended for similar use: an IRE scale. This is a scale used for quasi-luminance in analog television production. 100 is reference white; 0 is blanking black (7.5 is "black").

Evidently, the design of the Spotmeter is predicated on the assumption that "default" metering on a certain area should lead to a recommended exposure that would place that area at about 35 units on the IRE scale. I deduce that from what I see on this picture from the Spotmeter V manual:

from the Pentax Spotmeter V manual, as captured by Mike Butkus​

In that the "white triangle" fiducial mark corresponds to about 3.4 in the "IRE/10" scale.

Here, it is assumed that the range from 10 IRE units ("1" on the scale) and 100 IRE units ("10" on the scale) is 5 stops.

However, I am not yet able to make a successful relationship as to the luminance implication of 34 IRE units.

Hopefully there is no misunderstanding that the numbers 1-10 on that scale are equivalent to Zones 1-X.

According to a table in the manual, using "8" (80 IRE units) as the fiducial (rather than the white triangle) corresponds to a +1.7 EV offset from "default". If these were actual zone numbers (presented as Arabic numerals), and if the "default" were considered Zone 5, then Zone 8 would have an offset of +3.0 Ev.

Sensitivity of the Nikon D800

The official specifications for the Nikon D800 indicate that the "sensitivity" is on the basis of ISO REI (recommended exposure index).

That is a quantity that has no quantitative definition. It can be freely chosen by the camera manufacturer to produce the "exposure strategy" the manufacturer advocates. It can be described as"

The value that the camera manufacturer recommends be used as the exposure index into a free-standing exposure meter to get a nice exposure much of the time, and which the camera uses as an input to its internal exposure equation.​

Canon, for example, says formally that the sensitivity ratings of their modern cameras are in terms of ISO REI, but then says, in effect, "but that is pretty much the same as the ISO SOS".

Translation: "The intent is that these ratings are essentially ISO SOS, but you can't sue us if it turns out that they aren't."​

But Nikon doesn't mention SOS, as we have no idea whether they normally choose their REI to be essentially equal to the SOS or not.


************

I have (fleetingly) seen things in the forum about an actual Zone System scale for some version of the Pentax Spotmeter unit, or something like that, but was not able to figger out what was being spoken of.

More later maybe.

Best regards,

Doug

 

[Doug Kerr]

Hi, Chris,

Aha! This is an add-on sticker to provide a Pentax Spotmeter (V maybe) with a Zone System scale:

The sticker as a PDF file is available on APUG here:

http://www.apug.org/forums/attachme...one-scale-sticker-zone-scale-pentax-v.pdf.att

I don't know exactly how it is to be aligned (I don't yet have the "instructions") but my guess is that the "V" is to be aligned with the white fiducial arrow (and becomes the new default fiducial).

Best regards,

Doug

 

[Doug Kerr]

Hi, Chris,

Theoretically:

• If we have an exposure meter calibrated per the ISO standard for free-standing exposure meters, with K=12.5 (a typical value).

• Feed that meter, as the exposure index (ISO setting), the actual ISO speed (saturation basis) of the digital camera to be used.

• Take a spot reading with the meter on a certain area (using the "default" setup - no exposure compensation).

• Take the shot with the photographic exposure recommended by the meter.

then, that certain area should receive on the sensor a photometric exposure of about 12.8% of the saturation exposure.

Now, considering the use of a Pentax Spotmeter V, which I conclude is intended to be calibrated per the ISO standard for free-standing exposure meters, with K=8.7, and:

• Feed that meter, as the exposure index (ISO setting), the actual ISO speed (saturation basis) of the digital camera to be used.

That is clearly the intent per the manual.​

• Take a spot reading with the meter on a certain area (using the "default" setup - no exposure compensation - that wouild be using the white triangle fiducial).

• Take the shot with the photographic exposure recommended by the meter.

then, that certain area should receive on the sensor a photometric exposure of about 8.9% of the saturation exposure.

Now is we take that meter and:

• Feed that meter, as the exposure index (ISO setting), the actual ISO SOS of the digital camera to be used.

• Take a spot reading with the meter on a certain area.

• Take the shot with the photographic exposure recommended by the meter (using the "default" setup - no exposure compensation).

then, that certain area should receive on the sensor a photometric exposure of about 12.5% of the saturation exposure.

Interestingly enough, the level of 34 IRE units (the "default" IRE level of the Spotmeter) should, theoretically, correspond to about 9.3% of peak luminance (100 IRE units). Fancy that!

I had hoped to get some insight into the Nikon situation from a blog post by "adangus" of the Nikonians, but so far I only have this transcript of it, which is incomplete:

http://blog.nikonians.org/adangus/2009/03/a-zone-system-study-of-the-nik.html

Well, I have to do some real work now.

Best regards,

Doug

 

[Doug Kerr]

Hi, Chris,

I found another "door" to Allan Angus' paper "A zone system study of the Nikon D80":

http://stonerosephotos.com/blog/2010/02/a-zone-system-study-of-the-nikon-d80/

I haven't yet made peace with all his notation and such.

But his conclusion (and again I haven't reconstructed it) is that it seems as if the ISO REI's assigned by Nikon are essentially closer to ISO speed values than to ISO SOS values.

If that is indeed so, then the picture regarding your Pentax Spotmeter and your Nikon camera fits nicely together.

That is, we would expect the Spotmeter "default" exposure recommendation to be about 0.5 Ev less than that of the camera (with 0 EC), as you have essentially observed.

Still a mystery is why the relationship seems to differ between different illumination situations.

But one gremlin at a time!

Best regards,

Doug

 

[Doug Kerr]

Hi, Chris,

Here is another rather nice do-it-yourself Zone System scale for the Pentax Spotmeter V:

This is offered (physically, stick-on) by Jack Stratton on ebay:

http://i.ebayimg.com/00/$(KGrHqF,!ikE2LhluV)mBNjmn2Zy7g~~_12.JPG

Although he doesn't say, I assume the drill is to mount it with the "V" at the original fiducial location.

Best regards,

Doug

 

[Chris Calohan]

Al Weber and David Vestal made me one during a workshop at the Photo Formulary several years back. It's hand-printed but exceptionally accurate when used with a film camera. I don't know any other two people on this planet save Kim Weston who know the Zone better. Unfortunately, I think Al Weber has passed on. Both David and Al swore there was never any reason to go beyond VIII and thus that's as high as mine goes. The shadow detail is always there as are the delicates in the highlights. I miss shooting film.

 

[Doug Kerr]

Hi, Chris,

Al Weber and David Vestal made me one during a workshop at the Photo Formulary several years back. It's hand-printed but exceptionally accurate when used with a film camera.

Neat-O!

I don't know any other two people on this planet save Kim Weston who know the Zone better. Unfortunately, I think Al Weber has passed on. Both David and Al swore there was never any reason to go beyond VIII and thus that's as high as mine goes. The shadow detail is always there as are the delicates in the highlights. I miss shooting film.

Ah, yes.

Best regards,

Doug