Digital Photography and the Zone System

[girod]

Wow, this is truly a very liberating education for me Keith. You have the rare gift to teach, I really appreciate it very much.

So many things make sense to me now like, I see this "banding" so often when I try to recover highlights with the "Highlights Protection" slider; same thing with shadow recovery (though I did not know of combining it with the negative exposure slider) as you have described.

I came home late, the light was not available anymore to shoot for the +2.5EV. I'll do this hopefully tomorrow.

Meantime, can we please continue.

girod

[keithsnell]

Girod,

I'm glad it is making sense.  When you shoot the +2.5 exposures tomorrow, please also shoot one other series of images.  This time please shoot a series of images of a white towel.  Use the same setup as before, and take images of the white towel with 0 EV, and then from +1.7 EV to + 3.0 EV in 1/3 stop increments.  This test will allow us to see what "detail" we can render from images shot at these various exposures.  The purpose is to help us nail down where "Zone VII" is on the exposure scale.  Zone VII is defined as "the highest zone that will still retain good details" (such as textured snow, finely patterned white material, etc.)  When using a modified Zone System to base our exposures on in the field, we need to understand how to optimize our exposures to "retain good detail" in the important highlight areas of the image.

By the way, one of the "shortfalls" of a simple ETTR technique using the histogram, is that photographers are often tempted to "protect" highlights in Zones IX and X.  This would be an incorrect exposure according to the Zone System.  Zone IX is defined as a bright white area with no detail or significant texture, and Zone X is defined as a bright light source or "paper base white, with no image data recorded."   If we wanted to optimize the overall quality of our image (not induce noise in the shadows) we would most likely make a conscious decision to let Zones IX and X be "clipped" in the rendered image.  Keeping in mind of course, the "banding" issue at +3.0 EV.  As long as we are not overexposing large portions of the image, the banding shouldn't be noticeable.

Unfortunately I need to take a break to help get children to bed, but would like to continue this discussion later.

Keith

[girod]

I am assuming that when I take shots of a white towel, it should be focused optimally. I'll take those shots tomorrow.

Warmest goodnight to the kids. I have a 4 and a 2 year old girls, they are my magical inspiration to really get to know my D700 optimally for every scenario.

girod

[keithsnell]

I am assuming that when I take shots of a white towel, it should be focused optimally. I'll take those shots tomorrow.

girod

Yes, good point.  We want to be able to judge the "detail" in the shot, so yes, you should ensure the image is focused optimally.  

One thing I forgot to mention in my initial instructions here:  "Set your camera on manual exposure, spot meter (set for the center of the frame)..." was that you should use the "eyepiece shutter" to eliminate stray light coming in through the viewfinder that may affect your metering.  In the range of exposures you were working with, using the eyepiece shutter could make about 1/3 stop difference in exposure.  (So your +3.0 EV exposures might have actually been closer to +2.7 EV if you weren't using the eyepiece shutter and took the shot without your eye to the viewfinder.)  If, however, you had your eye up to the viewfinder when taking the exposures, then you were also blocking the stray light, and the exposures should be accurate.  

You might also want to use mirror lockup and a cable release for the "white towel" shots.   Given the range of shutter speeds you are likely to encounter, mirror lockup should help prevent camera shake.

While you are taking your "white towel" shots, we can also "skip ahead" to the next test sequence and take the "black point" shots as well.  The shadow range is a bit more forgiving, so we don't need to expose at 1/3 stop increments.  1/2 stop increments would probably be sufficient.  The purpose of this test is to help us understand what exposure will give us Zone III, where we typically place the "darkest area with visible detail," and how much we can "recover" shadows without objectionable noise or posterization in the image.  You should start this sequence (exposure shots of the white towel) at -2.0 EV and continue in 1/2 stop increments all the way down to -5.0 EV.  

Once we understand how a "nominal" exposure will place the tones in our rendered image, and how much "margin for recovery" we have at each end, we can start to make informed decisions about how to optimize our exposures in the field to best capture the dynamic range in the scene.  Understanding "nominal" placement of dark tones and how much we can recover in the shadows if needed is part of this "informed" decision making process.

I should add a couple of caveats:
1)  You will read all sorts of claims on the internet about "shadow recovery" capabilities of various cameras and software.  The bottom line is that the level of noise and posterization that one is willing to accept in an image is a personal choice, and often dependent upon the particular scene.  Each of us needs to perform some level of testing on our cameras/sensors to determine "acceptable" limits for our intended use of the image.  (More noise is acceptable if the image will be displayed less than full size on the web, etc.)  These "acceptable limits" will also change based on ISO, with progressively less "latitude" available at higher ISOs.

2)  The "white towel" test to determine the level of recoverable shadow detail will give us only a rough approximation (and an optimistic one) of what level of shadow recovery is possible.  The most demanding test will be recovering "skin tones" from the shadows.  You will find that your limits of "acceptable" recovery will be influenced by how well the color balance is retained and recovered in the lower reaches of the shadows.  Some cameras are better than others in this regard.  The D3/D700 is not the best in this regard (D2X probably is) but not the worst either.  As a follow-on to your "white towel" tests, you might want to refine your "acceptable" levels of shadow recovery by performing a subset of this test using skin tones...

We're not to far from our goal now.  Just a little more discussion about "mid-tones," tone curves, and "real world" dynamic range.

Keith

[keithsnell]

Girod,

For the benefit of other website visitors that haven't had the opportunity to perform these tests themselves, I thought I would post the results of my "white towel" exposure testing.  Please let me know if your results agree with these.

Again, the purpose of this phase of my testing was to determine where "Zone VII" is on the exposure scale.  Zone VII is defined as "the highest zone that will still retain good details" (such as textured snow, finely patterned white material, etc.)  When using a modified Zone System to base our exposures on in the field, we need to understand how to optimize our exposures to "retain good detail" in the important highlight areas of the image.

We used a "white towel" as our target because it has plenty of fine detail and texture that will allow us to judge when we have successfully exposed the image at the "highest zone to retain good detail."  Since we had already determined by examining the raw data that  the saturation point for the D3/D700 sensor is +3.0 EV from the "mid-tone" meter reading , I used that as the "upper end" of the test range for this series of images.  

I set up the camera on a tripod, focused on the white towel in even diffuse lighting on the floor below the tripod, set an aperture of f8, closed the eyepiece shutter, spot metered and set the exposure at +3.0 EV (I checked the meter by "defocusing" the image to ensure it wasn't effected by uneven lighting on the towel), and took the exposure using "mirror up" and a cable release.  I repeated this process in -1/3 stop increments down to +1.7 EV above mid-tone.

I then opened each image into Capture NX and used the "highlight protection" slider to pull the histogram back down into the image in an attempt to recover the highlights.  I used the "show lost highlights" view to judge how much I needed to move the "highlight protection" slider.  I set the highlight protection slide at the point where the "show lost highlights" view no longer displayed any clipping.  I then examined each image for the level of detail present.

Here is the +3.0 EV image, with the highlight protection slider set at 80/100.  Both the histogram and "show lost highlights" view imply that all the highlights have been recovered; however, examination of the image at 100% view tells a different story:

+3.0 EV, "highlight protection" applied

+3.0 EV, "highlight protection" applied, zoomed to 100%  (cropped)

Here is the +2.7 EV image, again with the highlight protection slider set so that the histogram and "show lost highlights" view imply that all the highlights have been recovered.  I applied a white balance adjustment to this image, so the highlight protection slider had to be set to 85/100 to remove all the "clipping."  Here's the "detail" available in this image:

+2.7 EV, "highlight protection" applied, zoomed to 100%  (cropped)

Here is the +2.3 EV image.  Same process, highlight protection slider set at 31/100:

+2.3 EV, "highlight protection" applied, zoomed to 100%  (cropped)

Here is the +2.0 EV image.  Same process, highlight protection slider set at 6/100:

+2.0 EV, "highlight protection" applied, zoomed to 100%  (cropped)

Adobe Camera Raw has traditionally done a better job of recovering highlights than CaptureNX, so out of curiosity I opened the +2.3 EV file in Adobe Camera Raw 5.3, set the exposure compensation to -1.0, and performed a click balance white balance adjustment.  Again the histogram and "clipped highlights" display indicated that all highlights had been "recovered."  Here's the detail in the file:

+2.3 EV, processed in ACR 5.3, -1.0 EV on exposure slider, zoomed to 100%  (cropped)

The ACR conversion does appear to recover highlights better than the NX2 conversion of the +2.3 EV file, providing a smoother tonal transition in the areas that are recovered; however, this rendition falls way short of the detail contained in the +2.0 EV image from NX2.  I tried converting the +2.7 EV file in ACR 5.3, and the rendition was unacceptable, showing the same non-recoverable artifacts as the NX2 renditions.  It appears that ACR 5.3 will provide 1/3 stop additional "margin" in recovering highlights, although at the cost of significant image detail.  

Someone that has been following this discussion might now be asking "if my sensor doesn't saturate or "clip" until +3.0 EV, why in the heck do I need to expose highlights at +2.0 EV in order to retain detail in the highlights?"  The answer is that those "details" span a 1.3 stop range in the image (in pretty much a "bell curve" type of distribution for the image of the towel).  Looking at the +2.0 EV image of the towel again, it certainly doesn't appear to our eye that this is a 1.3 stop range of brightness, but it is.  This is consistent with just about any "real world" scene where you want to "retain detail" in the highlights.  

Lets look at the raw data histogram for the EV +2.3 image (without whitebalance multipliers applied) in Rawnalyze:

White towel spot metered at +2.3 EV, raw data histogram from Rawnalyze

You can see from this screen shot of the raw data histograms that even when metering to expose the image at 0.7 EV below the saturation point of the sensor, there is still a substantial percentage of the green and blue channels that are "clipped" in the raw data.  (The "saturation point" of the sensor is annotated by the dotted yellow line on the right side of the histogram.)  Lowering the exposure to +2.0 EV moves this "clipped" data into the range that can be accurately recorded by the sensor.

Therefore, if my goal is to place "important" highlights into Zone VII, "the highest zone that will still retain good details," then I will expose these highlights at +2.0 EV using the camera's spot meter.  This is consistent with the Modified Zone System that I have been using for three generations of digital cameras, including Nikon, Canon and Kodak digital SLRs.

Keith

[girod]

I am very sorry Keith that I am unable to do my assignments. Yesterday I came home very late from work and very tired that I just dropped dead on the bed. And today, I just came home with the daylight gone.

For the benefit of other viewers, please continue the momentum. I will catch up.

girod

[keithsnell]

Girod,

No problem.  I understand what it is like to be busy (and tired).  You can read along with my posts, and ask questions if anything I say doesn't quite make sense.  I do find that most people learn much better by doing than by reading, so you might want to do the tests for yourself when you get a chance.

This gives me a good opportunity to say that we will be taking my 3-year old son and 1-year old daughter tent camping for the next two days, so I won't be posting on Thursday or Friday.  I'll pick up where we left off once we return from our camping trip.  (Hopefully I will get a chance to photograph a bit on the trip too.)

Thank you again for asking your initial question and giving me the incentive to get some of this information posted on the site.

Keith

[girod]

Hello Keith,

Finally I was able to do the shots following all your instructions (spot meter, tripod, eyepiece shutter closed, mirror-up and cable release) and here are my results or at least, my interpretations:

1) Grey card, +2.5EV: the data are very near the right wall of the histogram; no clipping at all.

2) White point white towel test:
    a. at 0EV: the data are centered at the midtone (Zone V)
    b. +2EV: near the right wall, no clipping
    c. +2.3EV: clipping starts (only Green channel, No WB correction)
    d. +3.7EV: Clipped but 100% recovered with highlight slider (52/100) without banding at 100% 
                      viewing.
    e. On Rawnalyze: without WB correction, the Green channel is clipped at +2.3; with WB correction,
                                aRGB (1998) - NO clipping of all RGB channels upto +3.7EV

2) Black point white towel and skin tests: at (-)5EV, the data are in the left wall of the histogram, photo
    is black or dark with no details visible; after applying shadow recovery slider (100/100), the data and
    photo moved to Zone III to IV, some and very acceptable noise at 100% viewing.
 

What do you think Keith?

girod

[keithsnell]

1) Grey card, +2.5EV: the data are very near the right wall of the histogram; no clipping at all.

This is what I would expect.

2) White point white towel test:
    a. at 0EV: the data are centered at the midtone (Zone V)
    b. +2EV: near the right wall, no clipping
    c. +2.3EV: clipping starts (only Green channel, No WB correction)
    d. +3.7EV: Clipped but 100% recovered with highlight slider (52/100) without banding at 100%  
                      viewing.

Did you mean +3.7 EV or +2.7 EV?  (It seems odd that you skipped from +2.3 EV to +3.7 EV.)

If you meant +2.7 EV, then I think the difference in "recoverable" highlights can be attributed to the blue channel.  I noticed after taking my shots that the window light I shot under was very cool.  It was a "blue sky day" here in Colorado, and since we are at higher altitude (7,000ft) the color temperatures are even cooler.  With that cool of light (equivalent to cool "shade," the blue component of the raw data (without white balance multipliers applied) was actually slightly greater than the green channel.  This caused BOTH the green and blue channels to clip.  Highlight recovery can interpolate data when only one channel is clipped, but when two channels are clipped it doesn't have enough information to successfully interpolate the missing data.  In "direct sunlight" or cloudy conditions, the light would not be as cool, and therefore the clipping would be primarily in the green channel (and therefore more "recoverable" with interpolation).  My results would be the "conservative" results.

I would be very interested in comparing the "detail" in your +2.7 EV shot against the "detail" in the +2 EV shot.  Remember that our goal was to "retain detail in the highlights" by placing them in Zone VII.

Given your results, I will try to find time to reshoot my sequence in "warmer" light (by shooting in full sunlight or on a cloudy day, instead of the cooler "shade" lighting I shot the initial sequence in).  I'd like to compare the detail available from interpolation (i.e., with only one clipped channel) against the detail available from "unclipped" data.

    e. On Rawnalyze: without WB correction, the Green channel is clipped at +2.3; with WB correction,
                                aRGB (1998) - NO clipping of all RGB channels upto +3.7EV

Did you really mean +3.7 EV?  That seems odd.

It also seems odd that Rawnalyze was showing clipping at +2.3 EV without WB correction, but no clipping with WB applied.  Note that the "clipping" indications are different if WB is applied.  You won't see the yellow dotted line, but instead will see a "spike" on the right side of the histogram.  The "height" of this spike indicates the amount of clipping.

2) Black point white towel and skin tests: at (-)5EV, the data are in the left wall of the histogram, photo
    is black or dark with no details visible; after applying shadow recovery slider (100/100), the data and
    photo moved to Zone III to IV, some and very acceptable noise at 100% viewing.

This is fairly close to what I observed.  My level of "acceptable" noise was at -4.7EV, whereas the noise at -5.0EV became too noticeable for me.  I did however, shoot a few other test scenes with a range of tones from white in sun to black in shadow, and found that I could apply "shadow protection" to recover the "black in shadow" (which was -5.0 EV from midtone) without any objectionable noise.  The bottom line is that in a "real world scene" with a black textured object in shadow I can successfully recover the shadows (If I choose to do so) without introducing objectionable noise in the shadows.  Successful shadow recovery is dependent on accurately placing Zone VII in the image.  If I underexpose too much (in an attempt to "protect" highlights in Zones IX and X) then I won't be able to recover shadows without introducing noise.  This reinforces the need to make a conscious decision of where to place the tones, and meter accordingly, during the initial image capture.

Make sense?

Keith

[girod]

Thanks Keith.

The recoverable highlights was from +2.3 to +3.7. The light was a direct sunlight (~4PM, New Orleans) coming thru my window. You're right the clipping was solely in the Green channel.

There was more detail in +2EV compared to the +2.7. But at 100% viewing, the fibers of the white towel are still very detailed at +2.7

I was looking for the yellow dotted line in Rawnalyze even with WB applied. You're right, with the WB applied the "spike" showed up in the Green channel starting at +2.7

I will reshoot my entire White point white towel test series. Meanwhile, can we please move on.

girod

[keithsnell]

Girod,

A couple of thoughts:

1)  There is a bit of inconsistency between our results.  One reason is because we were testing under light with different color temperature (shade vs direct sunlight).  However, this doesn't explain all of the differences we are seeing.  I believe there might also be a bit of inconsistency both due to our interpretation of the results and perhaps a misunderstanding of how to "apply" white balance in Rawnalyze.

Here is a screen shot of the histograms displayed in Rawnalyze for the +2.0 EV image with white balance applied:

"white towel" exposure test at +2.0 EV, histogram in Rawnalyze with "selection" WB applied

Using the "composite" view, I made a large selection (right click and drag) from the center of the image, then clicked Ctrl + W to apply a white balance correction to the image using the "selection" to calculate a neutral white balance rendition.  The "hash marks" on the horizontal axis of the Rawnalyze histogram display indicate 1/3 stop increments.  With white balance applied the red and green channels are essentially just below clipping and the blue channel is clipping slightly.  

Here is a screen shot of the Rawnalyze histogram of the +2.3 EV image with white balance applied:

"white towel" exposure test at +2.3 EV, histogram in Rawnalyze with "selection" WB applied

Note that at +2.3 EV, with the "proper" white balance applied, all three color channels indicate clipping.  In my case (because I was shooting in "cool" light) I was unable to recover the detail in the towel because both the green and blue channels were clipping in the raw data.

I hope you don't think I'm harping on this too much.  I really want to make sure we are on the same page here because the most "unforgiving" aspect of digital photography is when data is "clipped," i.e., the sensor is saturated.  If the data is clipped, the software can "interpolate" (guess at) the data to try to reconstruct the missing detail, but this interpolation is a poor approximation of the "real" data.  If it isn't important to display this portion of the image with all of the detail that is available in the original scene, then that is OK, but it doesn't belong in Zone VII if it isn't important.  

It is widely accepted by followers of the Zone System that "perceivable detail" belongs in Zones III through VII.  For a high-contrast scene (using a digital camera), you would expose to place important highlight detail in Zone VII and then adjust mid-tone and shadow placement as needed during post-processing.  We need to understand how to expose correctly to make that happen on a consistent basis.  Mid-tone and shadow placement are less important during the initial image capture because we have more latitude in adjusting placement of those tones in post-processing.

2)  I think the results of our tests under different color temperatures of light reinforce that our success at interpolating data from clipped highlights will vary depending on the "warmth" or "coolness" of the light source.  With "direct sunlight," as long as only one channel is clipped, the software can interpolate the data in the other two color channels to "reconstruct" a portion of the missing data in the clipped channel.  The success of this interpolation will depend on how consistent the data was in "real life" in the clipped channel.  With cooler light, we are in danger of clipping both the green and blue channels at the same time.  The challenge is to know when we are truly clipping the blue channel, and not be fooled by the white balance multiplier being applied to the data during the in-camera or CaptureNX2 processing.  

A general rule of thumb for the D3 and D700 is that in "shade" under a blue sky, the blue channel will clip at about the same time as the green channel.  The blue channel coefficient (for the D3/D700) for "shade" is 1.0859, so if you see blue channel clipping when you have "shade" set for white balance (or when you are using auto WB and shooting in shade under a blue sky) then you should expect that the "raw" blue channel data is truly clipping.

There are differing opinions with respect to how "hot" we can expose the highlights and still obtain usable results.  It is true that in some situations, you can expose (Zone VII) highlights at +2.3 EV, or even +2.7 EV, and still obtain usable results by using "highlight protection" or other methods to recover the highlights.  My goal is to be able to consistently obtain optimum exposures in challenging lighting conditions.  I know that I have a lot of latitude in the shadows with the D3 (and D700) and so adopting a "conservative" method of exposing the highlights has served me very well.  Exposing highlights with visible detail at +2.0 EV will almost always enable me to produce a viable image from the recorded data.  Exposing any hotter than this is "treading on thin ice."

I would very much like to continue this discussion, but I need to take a break now to finish getting ready to teach a workshop in Crested Butte, Colorado.  Please check back with me on Tuesday of next week and we can continue our discussion.

Keith

[girod]

Thank you very much for your patience Keith.

I've just redone the "White point white towel test" (New Orleans, 3PM direct sunlight on the floor thru my window) and here are my interpretations:

In CNX2:
1) 0EV - data centered at Midtone (Zone V) and no clipping upto +3.0
2) +3.3 - clipping starts (Green channel)
3) +4.0 - Highlights clipped (still Green) fully recovered with 74/100 highlights slider, no banding that I could discern at 100% viewing

In Rawnalyze:
1) +2.0 - No WB: R&B at middle of left of histogram, Green at midtone to middle of Left of histogram; With WB: R&B at Zone V, G at Zone V to middle of right of histogram.

2) +2.3 - tiny spike of Green clipped both with and without WB.

3) +2.7 - Green clipped without WB; R&G clipped with WB applied.

I think the above is consistent with your tests on your D3 - that for D3/D700, +2.0 EV (??stretchable maybe upto 2.3 - 2.7 with UniWB) is the safe Zone VII.

If you think that the above findings on my D700 have inconsistencies relative to your expectations, I could send to you the NEF files.

Do you think we can now go forward to "Tone curves", maximally capturing the real-world dynamic range, correct and maximal use of the CNX2 and others that you deemed important in this method of exposure.

girod

[keithsnell]

Girod,

Thank you very much for your patience, and for sticking with me in this discussion.  I would like to continue the discussion, unfortunately I am up against deadlines for preparing for our upcoming workshop in Crested Butte.  I will have to defer any further discussion until we return from the workshop next Tuesday.

If you have time to do some experimentation on your own, you might want to place a gray card in a "real world" high-contrast scene (one that contains highlights where you would want to retain detail, with bright areas in sun and dark areas in shadow), and use spot meter to determine the "range" of the scene by metering on the highlights, mid-tones and shadows.  You will want to note the exposure values of the different areas within the scene so that you can refer to them later.  Next, use two different methods to expose the scene.  First, meter on the mid-tones with spot meter (meter on both the gray card, and other tonal areas within the scene that you judge to be "mid-tone) and take the exposure.  Try an additional "mid-tone" exposure with -0.5 EV exposure compensation.  Next, meter on Zone VII and take the exposure (at +2.0 EV, then if you wish, at a higher EV).  Examine where the tonal values fall for the highlight, mid-tone and shadow areas when the images are rendered in Capture NX2.  

After you have completed your experiments with a "high-contrast" scene, repeat the same experiments with a "low contrast" scene, where all the tonal values fall within Zones III through VIII.  

There are two components to this exercise.  1)  Observing how different metering techniques affect the results  2)  Observing the "non-linear" placement of tones in the rendered images using the various tone curves within Capture NX2.

Keith

[girod]

Welcome home Keith.

I'm in the process of implementing (or experimenting) the 2 methods of exposure (Zone VII exposure and the centered-exposure) that you have so patiently illustrated, for both the high contrast and low contrast real world scenarios. I will try the -0.5EV in centered-exposure and the +2.3, +2.7, +3.0 EVs in Zone VII exposure. I will also try these using the UniWB (for D700) + custom linear tonal curve.

Meanwhile, I have some questions:
1) "Examine where the tonal values fall for the highlight, mid-tone and shadow areas when the images are rendered in Capture NX2."  - by looking at the CNX2 histogram? or looking for the actual number of tonal values? where? in Rawnalyze? how?

2) "Observing the "non-linear" placement of tones in the rendered images using the various tone curves within Capture NX2." - the CNX2 picture control (Neutral, Standard, Vivid etc)?

3) "..modify the behaviour of your raw processor in order to force it to interpret how you have chosen to record the data (to maximize dynamic range) and output the data in the way that you want it presented." - how can I do this in CNX2?

4) Could you please explain more on midtones and tone curves?

5) How do you create a custom linear curve for the D700. I did it this way: In CNX2, opened "Picture Control Utility", selected "Neutral" then "Use Custom Curve" (this, I've read in dpreview forum, apparently defaults to a linear "flat" curve); exported then loaded to my D700. Is this right?

Thanks,
girod

[keithsnell]

Hi Girod,

In answer to your questions:
1)  The easiest way is to use the "watch point" function in NX2 to see where the tonal values fall within the RGB scale.  

2)  First, 0 EV exposure does not place the resulting rendered value at the midpoint, but above the midpoint.  Typically a -0.5 EV exposure compensation is required to place a tone at mid-tone in the rendered image using a linear tone curve in NX.  If you are using a non-linear tone curve (i.e., one of the supplied tone curves with NX), then even more negative exposure compensation will be required to place a tone at mid-tone in the rendered image.  Second, what tone curve you select will determine how much the other tonal values are shifted in relation to mid-tone.  Using a "standard" or "vivid" tone curve will shift the shadow and highlight values in relation to mid-tone, applying an increasingly aggressive "S" curve to the image.  

3)  If you record a mid-tone at 0 EV and capture NX places that mid-tone value higher than mid-tone in the rendered image, then you will need a curves adjustment or use the midpoint slider (under quick fix, levels and curves, or LCH editor to place the mid-tone in its "proper" position.  (If indeed, that is where you want the midtone to be.)  This is most noticeable when skin tones in portraits look too "hot."  On the other hand, if you expose Zone VII highlights at +2 EV, you are in effect shifting the mid-tones lower on the scale.  This actually places them about where they should be (around 118 in RGB), in effect compensating for the "hidden" exposure compensation NX applies to the midtones in a normally exposed image.

4)  "Mid-tone" is generally considered to be 118 on the RGB scale.  Therefore you would expect that if you used "highlight protection" NX would only shift the highlights and leave the midtones and shadows alone, and if you use "shadow protection" NX would only shift the shadows and leave the midtones and highlights alone.  This doesn't seem to be the case.  What appears to be happening when you use "shadow protection" is that NX keeps the white point set, but compresses both the shadows and midtones, and even some of the highlights up toward the white point of the image.  In other words, the "rendering" of mid-tone in NX is variable, depending on what tone curve you select and how you apply shadow and highlight protection or adjust the black and white points of the image.

5)  You created your custom linear curve correctly.