Middle Grey? Really?

[tirediron]

My guess-timate was that for first generation Nikon bodies (D1/D100 vintage) it seemed to be something almost Zone VI rather than Zone V, so yeah, the 12.5/13% makes sense.

 

[unpopular]

I'd much rather it be slightly lighter than darker...

 

[Derrel]

Another aspect you might be overlooking, as a Sony user, is that Nikon cameras use Red-Green-Blue 3-D Color light metering analysis, as well as reflective value analysis, so that the suggested meter reading is based on the "color" of the actual subject matter that the meter is reading. A Texas-based forum friend of mine (on another forum) who switched to digital with the Nikon D300 was complaining that the dynamic range of "digital" was simply terrible, and he had to "bracket like hell", often over a multi- stop range, to get anything that was even workable. I asked exactly how he was metering and exposing.

He was using what many call the "modified Zone System", and using his old, film-based Zone placement ideas---but instead of a color-blind and "dumb" light meter, his Nikon D300's meter reading ALREADY HAD MADE the allowance (or adjustment,zone placement, whatever word one wishes to use), that would place a white shirt properly or make a dark cat look dark, and not mid-gray...and then he was adding exposure to the already-highlight-biased suggestion the D300 was giving him...and subtracting exposure for reading he made off of dark-toned objects.

After he realized how the 3D, distance-aware, and color-aware metering was actually functioning, he said he felt rather foolish...I felt bad for him...

 

[memento]

this thread makes me feel dumb.

 

[Josh66]

this thread makes me feel dumb.

Just quote someone and then say - "good point".

 

[Helen B]

Another aspect you might be overlooking, as a Sony user, is that Nikon cameras use ... reflective value analysis, so that the suggested meter reading is based on the "color" of the actual subject matter that the meter is reading.

...but instead of a color-blind and "dumb" light meter, his Nikon D300's meter reading ALREADY HAD MADE the allowance (or adjustment,zone placement, whatever word one wishes to use), that would place a white shirt properly or make a dark cat look dark, and not mid-gray...

Two questions.

1) How does a Nikon camera know the reflectivity of an object if it doesn't take an incident reading?
2) Why doesn't my D3 do this? My D3 gives exactly the same mid-grey result whether it is pointed at a black cat, a grey beard or a white shirt. It has no idea what the reflectivity is. Does it have to go to Nikon school?

I suspect your friends problem was more to do with not knowing what the meter was reading off, or not understanding matrix metering.

 

[Derrel]

Good question Helen. Here's how: If a system can recognize highlight tones by color analysis, and can recognize black or other dark-colored objects, by their RGB color profile, there is no need for an incident light measurement: the computer can"work backwards" and get a pretty good idea of the light level incident on the scene, and it can do it in milliseconds, and has a 300,000-photograph image bank from which it draws its analysis. If one measures 420 (or 1,005, or more specific points) points within a scene using a reflectance metering method, in combination with a color analysis, it's pretty easy to compute an exposure. Color temperature of the light present on-scene is also factored into the RGB measurements. "It's done by computers". I can not imagine that one single press of the METER button on a Sekonic or Minolta is so much superior. Point the dome at the light and it's MAGIC!! Right???

Cameras like the D2x and D2h took incident readings of the LIGHT present, off the top of the prism as well as through the lens...apparently now Nikon feels reading only the WB info that comes through the lens is a better method, but white balance is one factor. A second factor is that the camera is aware of the degree of contrast the user has set, or which tone curve will be used to process to image, based on literally thousands of measurements made. Since there is only one type of "film", the metering is most akin to metering for color transparency film. So, in basic principle, the "brightest" and "whitest" part of the scene is where the "film" must NOT be blown out. It's similar to the old 1970's method of using the camera's built-in light meter in center-weighting, and reading the "brightest subject" in the scene, and then making SURE not to over-expose that. The difference between metering for negative films and for positive emulsions is one new factor that d-slr's do not have to worry about.

I liked the Nikon School comment!!! You're quite the comedienne!

Here's what Nikon says:

"[h=1]3D Color Matrix Metering II[/h]
Matrix metering evaluates multiple segments of a scene to determine the best exposure by essentially splitting the scene into sections, evaluating either 420-segments or 1,005 segments, depending on the Nikon D-SLR in use.
The 3D Color Matrix Meter II takes into account the scene's contrast and brightness, the subject's distance (via a D- or G-type NIKKOR lens), the color of the subject within the scene and RGB color values in every section of the scene. 3D Color Matrix Metering II also uses special exposure-evaluation algorithms, optimized for digital imaging, that detect highlight areas. The meter then accesses a database of over 30,000 actual images to determine the best exposure for the scene. Once the camera receives the scene data, its powerful microcomputer and the database work together to provide the finest automatic exposure control available"

from MIR's web site: 3D Colour Matrix Metering Process

I) Brightness data with colour information is read from each of the 1,005 pixels in the RGB sensor.
II) Colour information is extracted from the brightness information.
III) Brightness data is grouped into overlapped areas.
IV) From the colour information (see: II), the colour signal (average colour of the scene) is calculated.
V) From each group determined in (See: III), the brightness signal (average brightness) and contrast signals are calculated
VI) These signals (colour, brightness and contrast) and focus area signal (selected focus area) are compared to the camera's built-in database of more than 30,000 scenes taken from actual shooting experience. The optimum exposure value is then obtained.
VII) Finally, distance information from the D-type Nikkor lens in use is integrated to further optimize the calculated exposure value.

Moose Peterson's simplified version: "How does the 3D Color Matrix Meter work? It has 1005-pixel CCD which are what actually reads the light. The amazing thing to me is not the 1005-pixels, but that each pixel has incorporated in it one R (red), G (green) or B (blue) filter. Each sensor then not only evaluates the scene's brightness and contrast, but also the scene's color.
The information received from the 1005-pixels including scene brightness, contrast, selected focus, distance information and the scene's color characteristics is analyzed. This information is then processed by a powerful microcomputer and its database said to hold at least 30,000 exposure possibilities. This is where the new 3DII is different. It has a database of 300,000 compared to the original's 30,000."

Happy now?

 

[Derrel]

I just picked up my D2x and metered a black cast iron fry pan from 8 feet away, using the 80-400 AF-D zoom lens. Meter reading was f/5 at 1/3 second to 1/4 second, depending: the pan has some oil in it, and there is a faint glare depending on exactly where the meter is directed.


Moving left 10 inches to the white enamel top, exposure is f/5 at 1/50 second, occasionally shifting to 1/60 second, depending on slight glare variations that I can barely even see through the finder. Light is overhead fluorescent.


Huh. Doesn't seem like the readings are the same to me. Huh....my camera is actually doing what Nikon SAYS it will do!!


Tone curve was set to Normal for both meter readings in Aperture Priority mode.


Since the incident light is perfectly the same level, it seems like Nikon might actually be adjusting the exposure settings, based upon how the light metering system evaluates subject differences. Imagine that! Engineering claims actually backed up by a real-world test...


I just did a Nikon School assignment in my own home! Perhaps you misunderstood my original post. I was talking about pointing a Nikon at REAL-WORLD subjects, not metering a gray card. But then, you say your camera gives the same readings on white, black, or gray?

Best,
Derrel

 

[Helen B]

Happy now?

I think that you have misinterpreted the information. Moose Peterson's summary is good (of course) but it doesn't support your hypothesis, nor does any of the other information you quote. Nikon's remarkable matrix metering is all about comparisons, not about "If a system can recognize highlight tones by color analysis, and can recognize black or other dark-colored objects, by their RGB color profile, there is no need for an incident light measurement: the computer can"work backwards" and get a pretty good idea of the light level incident on the scene". The RGB colour profiles of black, grey and white are all the same. If the meter sees only dark tones or only light tones it can't tell what they are are and produces grey. This is very easy to see in practice. The matrix system has to see a wide range of tones to be able to come up with some notion of what is light and what is dark - but it still doesn't know what the reflectivity is. Nikon's matrix metering is extremely good, but it isn't magic. You have to help it sometimes.

 

[Helen B]

I just picked up my D2x and metered a black cast iron fry pan from 8 feet away, using the 80-400 AF-D zoom lens. Meter reading was f/5 at 1/3 second to 1/4 second, depending: the pan has some oil in it, and there is a faint glare depending on exactly where the meter is directed.


Moving left 10 inches to the white enamel top, exposure is f/5 at 1/50 second, occasionally shifting to 1/60 second, depending on slight glare variations that I can barely even see through the finder. Light is overhead fluorescent.


Huh. Doesn't seem like the readings are the same to me. Huh....my camera is actually doing what Nikon SAYS it will do!!


Tone curve was set to Normal for both meter readings in Aperture Priority mode.


Since the incident light is perfectly the same level, it seems like Nikon might actually be adjusting the exposure settings, based upon how the light metering system evaluates subject differences. Imagine that! Engineering claims actually backed up by a real-world test...


I just did a Nikon School assignment in my own home!

Best,
Derrel

Er, Derrel, that's exactly what I'm saying the meter will do as well. It isn't taking the darkness of the pan or the brightness of the enamel into account, because it doesn't have any way of telling how reflective they are. It gives different exposures in the same amount of incident light. It has no idea what the reflectivity is. You've disproved your own hypothesis.

 

[Kerbouchard]

I agree. If Derrel had actually pushed the shutter, he would have seen that both images would have turned out grayish. As a matter of fact, anyone who wants to set up their own Nikon School, and take three pieces of paper. A black piece, a grey piece, and a white piece. Using a tripod, filling the frame with the paper, and taking the shot with the meter zeroed will result in three close to identical pictures.

I just picked up my D2x and metered a black cast iron fry pan from 8 feet away, using the 80-400 AF-D zoom lens. Meter reading was f/5 at 1/3 second to 1/4 second, depending: the pan has some oil in it, and there is a faint glare depending on exactly where the meter is directed.


Moving left 10 inches to the white enamel top, exposure is f/5 at 1/50 second, occasionally shifting to 1/60 second, depending on slight glare variations that I can barely even see through the finder. Light is overhead fluorescent.


Huh. Doesn't seem like the readings are the same to me. Huh....my camera is actually doing what Nikon SAYS it will do!!


Tone curve was set to Normal for both meter readings in Aperture Priority mode.


Since the incident light is perfectly the same level, it seems like Nikon might actually be adjusting the exposure settings, based upon how the light metering system evaluates subject differences. Imagine that! Engineering claims actually backed up by a real-world test...


I just did a Nikon School assignment in my own home!

Best,
Derrel

Er, Derrel, that's exactly what I'm saying the meter will do as well. It isn't taking the darkness of the pan or the brightness of the enamel into account, because it doesn't have any way of telling how reflective they are. It gives different exposures in the same amount of incident light. It has no idea what the reflectivity is. You've disproved your own hypothesis.

 

[unpopular]

NERD FIGHT!!

 

[ann]

:lmao: However, very civil. What a radical thought for the internet.!

 

[GeorgieGirl]

I'm trying to get my brain wrapped around this concept...I will work on the three paper deal tonight. I'd like to be able to have that sink in a good bit more.

 

[memento]

NERD FIGHT!!

Good point! :lol