The Nikon metering myth

Roman describes a very thorough, yet complex method -- more complex than most users would want to use, especially when shooting in a hurry. Here are two "quickie" methods I use when the full Zone approach he describes isn't practical and/or absolutely necessary.

1. Eyeball a middle tone (middle gray) and spot meter on it. This takes some experience, but blue skies in sunny days, green grass and asphalt are some of my usual suspects. See http://esfotoclix.com/blog1/?p=23 .

2. If you really want to avoid blowing highlights, you can also spot meter on the brightest area where you want/need to show detail. In order to turn it to near white, yet not lose detail, you will need to place it in Zone 7 (2 stops above middle gray). Simply spot on that area, and open up 2 stops (i.e., if the center reading is 1/100 sec, open up to 1/25 sec). See http://www.outdoorphotographer.com/how-to/shooting/the-digital-zone-system.html

(BTW, with this second method, if your camera has wider DR, as some of these new wonders do, you be able to go to Zone 8)

Though these two methods are simplistic, I find they do the job for most of my work, especially when I'm in a hurry.

Roman's method is more comprehensive, because he's making conscious, deliberate decisions about how to expose his scene based on what's important to him in the final shot . That requires some forethought, like asking, "what do I want in this photo?" -- then basing your exposure on the story you want to tell. Some examples:

• For a given lighting situation, a bride's lovely olive skin tones would look overly dark if you expose to avoid blown highlights on the dress. How would you approach this? By first asking what you want out of the shot. If the shot is about the dress (i.e., a full body shot with wedding train draping the altar), you expose for the dress, then adjust the midtones in PP to lighten her skin tones. If the shot is a closeup of her face, and you want to ensure solid exposure of her facial features, meter for that. So what if the white veil blows out, especially if you use shallow DOF, where part of the veil will get a dreamy glow to it?

• In a night scene, if you expose to avoid the blinkies, chances are that you will get perfectly exposed street lights... and utterly dark and nearly unrecoverable shadows. If the shot is about the street lights, that might be a good approach, but that is seldom the case. A better approach might be to meter for the detail in the shadows (place them in Zone III), then bring down the highlights in PP with corresponding Shadow protection, IOW, reduce the overall contrast.

After sweating profusely about blown highlights with my D80 (and yes, that was a blown highlight machine), I've grown out of that fear. I find that with Spot metering and PP techniques, I can squeeze maximum DR without distracting, blown highlights. As I wrote about http://esfotoclix.com/blog1/?p=1242 , I am now a tad more relaxed about this whole highlight thing and enjoying my photography a lot more these days.

~~~~~~~~~~~~~~

It's easy to argue about equipment and technique, but hard to argue with a good photograph -- and more difficult to capture one .



Gallery and blog: http://esfotoclix.com
Special selections: http://esfotoclix.com/store
Wedding & Portrait: http://esfotoclix.com/wedevent
Flickr stream: http://www.flickr.com/photos/22061657@N03

Uh, that would be my take as well! Perhaps Nikon has technical writers that are also joke writers for late night television, and they've gotten their cards and punchlines mixed up.
Why would 3D matrix work the way it is supposed to? What is funny about that?

Yes, we saw some of that here:

http://esfotoclix.com/tech/MXM/mxm_avg.html
--
~~~~~~~~~~~~~~

It's easy to argue about equipment and technique, but hard to argue with a good photograph -- and more difficult to capture one .



Gallery and blog: http://esfotoclix.com
Special selections: http://esfotoclix.com/store
Wedding & Portrait: http://esfotoclix.com/wedevent
Flickr stream: http://www.flickr.com/photos/22061657@N03

If we combine this behaviour with a AE-L/AF-L set to AE Lock Hold or
AE Lock Only (CSM 19) we can actually focus where we like by half
pressing shutter button and holding it (camera focuses and takes
first reading) than we point camera for example to the sky and press
AE-L/AF-L button (or press and hold what depends on our settings CSM
18). Than we recompose again and take picture. In this way MM can
average readings from the darkest and lightest subject and prevent
blown out skies.

Now that sounds great. I'm going to try it out tomorrow and see what happens.

eNo said:

What's "normal"? The meter's tendency to over-expose in high-DR scenes? Or the need to use exposure compensation? We need to start expressing ideas a tad more clearly to avoid ships passing in the night conversations.

Click to expand...

The need to use exposure compensation is normal...an idea what was fully expressed in my post that you responded to.

I can't include the contents of all my previous posts in my follow up posts. Please refer to the thread to remind yourself of what you were commenting on, and what's being responded to.

.

Expose for middle ground, watch your highlights . That's all you need to do.

Cedarhhill -- Yes, you're right. We are all exposed (pun intended) to P&S technology, that seems to easily outstrip the DSLR manufacturing with regard to flashfill and matrix-type metering. We expect the fancy DSLR we just bought, that says it will do it all for us with matrix, to do just that. When it fails, we whine or we deny it -- that 's what I've learned from the threads here.

I feel matrix should work well in 95% of lighting situations. And the misses shouldn't be a complete catastrophe.

I'm wondering, having read a lot of the posts on Nikon's exposure problems. The problem does seem related to ambient light intensity. When we're outside on a sunny day, the matrix seems to be saying to itself, "white cat on a white rug: more exposure!", and when we have an nighttime indoor event that is too dark for reading, the meter says "Dark on dark: underexpose!" Older metering of the weighted type would just say "make it Grey!" and be right most times, and all we as photographers would have to do is estimate the over tonal strengths of the subject and background.

If this is true, we now face a very unpredictable scene analysis. If we see a balanced scene with consistent bright lighting, instead of metering and snappign a well exposed shot, we now have to think "looks good, but the light is good too -- stop down to avoid overexposure!" See what I mean? In the past, it would have been a perfect shot following the camera's meter.

I don't think it should be that way, if that's the way it is. But that does seem to be the case of "Programmers Gone Wild!" They seem to have programmed the letter of the law, but missed the crucial usage factor.

The review's example shot of the large boat in sunlight, with a bit of foreground shadow on the bow, should have been a slam-dunk exposure. If anything, it might have come out a little dark due to the white superstructure. Instead it came out badly over-exposed -- far beyond anything that front shadow could have produced (it just wasn't big enought in the frame!). The only way I can see that happening is code saying "white cat/white rug=more exposure".

Will Nikon ever come clean on this, and how such programming code gaffs are affecting normal photo meter, AF, and probably flash photography as well?
Stay tuned to find out.

I, for one, think there is a larger issue here, and not a matter of "put up and and shut up" loyalty issue.

Until earlier today I had not done rigorous testing and thought this might just be me remembering wrongly.

However today I have done some more rigourous testing.

And CLEARLY, when shooting LV the exposures are ALWAYS brighter than when shooting w the OVF.

I'll post the shots in a moment but here's the data, for shots taken at 200mm w the 18-200mm 3.5-5.6 (at 200mm this lens thus cannot open more than f5.6):

shot 1: full auto mode w/o flash

• exposure in LV chose: 1/13s, f5.6, ISO 1600
• exposure w OVF chose: 1/20th f5.6, ISO 1600. Result darker of course.

shot 2: full auto mode w/o flash

• LV chose: 1/8th, f5.6, ISO 1600
• OVF chose: same as above but 1/20th only. Result darker of course.

shot 3: Aperture priority mode

• LV chose: 1/15th, f5.6 selected, ISO 3200
• OVF chose: 1/30th, f5.6 selected, ISO 2500. Result darker of course

The way I would evaluate the metering accuracy is this:

Compare with other cameras. If one camera more consistently produces properly exposed images than another then it has better metering. I find the metering on my D7000 to be very similar to my D90 in that both of them tend to over expose in bright, high contrast situations. I was really hoping that the D7000 would do better than the D90, but at this point, my opinion is that it does not. In fact, I find that my Sony A55 nails the exposure with a higher frequency than either of my Nikon cameras. All of this is with matrix metering. With the D7000, like the D90, the solution is to switch to center weighted metering in high contrast situations.

Considering that the D80, D90, and D7000, all seem to have the same issues with over exposure using matrix metering in high contrast situations I have concluded that Nikon intends it to be this way. In short, Nikon's matrix metering is trying to meter to bring up the shadows, often at the price of losing the sky. I have often wondered if this is a strategy to get better scores for high ISO performance. If a camera's meter leaves the shadows too dark, then the shadows will have more noise, especially when you try to raise up the shadows in post processing.

At any rate, with Nikon I have learned to use center weighted or even spot metering in high contrast situations. Using exposure compensation really doesn't solve the problem for me as well as switching to the center weighted or spot metering mode. With a bit of practice one can get a feel for how the camera is going to meter in certain situations. So if you combine center weighted metering with the exposure lock button you can get good results in bright conditions. Not a perfect solution, I know.

Yep, and 3rd blown highlight question for me would be...does a blown highlight occur naturally to our vision when we physically look at the scene. My goal is matching human vision so if the highlight is beyond our own physical DR when we focus on the subject...as far as I am concerned...its game for letting the highlights fall where they may.

And not sweating them is good advice....nice article.

Roman
--

“There is only you and your camera. The limitations in your photography are in yourself, for what we see is what we are.”
~ Ernst Haas

We are officially live!!!!
http://www.commercialfineart.com/
Old Web Site
http://www.pbase.com/romansphotos/

these are just two of a large number - this does happen systematically.

in this very case the OVF shot works better because the lettering is perfectly legible. Whereas the LV shot has OVER-exposed the yellow area and this makes the lettering much less cleanly rendered:

darker shot, taken w OVF
http://www.flickr.com/photos/antoinebach/5230173644/

brighter shot, taken w LV
http://www.flickr.com/photos/antoinebach/5229578921/

Interesting Martin! Thanks. That does seem to explain some of the problems.

It seems like something was tweaked in the programming to make it over-expose a bit too much. Hopefully a firmware fix will be available.

The other two evaluative systems I've successfully used are Canon and Pentax. I will say both of these are slightly more conventional -- they tend to underexpose brightly lit subjects and the reverse for dark ones. But that's more along the lines of what we expect. The Nikon seems excellent in making decisions about exposure most times, resulting in better metering than these other two brands. But mine D7000 seem chronically to need about -0.5 EV, which doesn't please me very much.
Still hoping for a simple fix.

Danel said:

solution is to switch to center weighted metering in high contrast situations.

Considering that the D80, D90, and D7000, all seem to have the same issues with over exposure using matrix metering in high contrast situations I have concluded that Nikon intends it to be this way. In short, Nikon's matrix metering is trying to meter to bring up the shadows, often at the price of losing the sky. I have often wondered if this is a strategy to get better scores for high ISO performance. If a camera's meter leaves the shadows too dark, then the shadows will have more noise, especially when you try to raise up the shadows in post processing.

Click to expand...

You are probably right Nikon might be metering hot for low light. But..

Then why do it for low ISO? If anything I'd say both require a different strategy. I still think the D200 was way better for metering than the D80, least when I tried it. But they got slammed for noise at high ISO. I guess that's why they meter a bit hot.

I've used a few models from various makers. Sony tend to be a bit more cautious (but the underexposure hurts them at high ISO on some models) Canon a bit hot, but not as much as Nikon. Pentax a bit under too though probably a bit better with recent models. I don't think you can rate metering on no. of zones alone.

Still think under is better than over. You've got a lot more latitude in the shadow areas not so much highlight end just like slide film over exposure isn't great. I know more than a few D80 users who gave up on matrix metering and use it in CW only.

We could argue that good photographers won't mind and can take control, but I think Nikon should tweak thing a bit even if they were more cautious at low ISO. I'd say they could dial back the high ISO a bit too.

Gunzorro said:

Interesting Martin! Thanks. That does seem to explain some of the problems.

Click to expand...

It explains nothing. The explanation on that page is nonsense. Nikon does not explain how matrix metering works, so no one knows what it's doing. Pages like that are just a lot of presumptions backed up with nothing concrete.

.

RomanJohnston said:

Yep, and 3rd blown highlight question for me would be...does a blown highlight occur naturally to our vision when we physically look at the scene. My goal is matching human vision so if the highlight is beyond our own physical DR when we focus on the subject...as far as I am concerned...its game for letting the highlights fall where they may.

Click to expand...

Good point, Roman. While cameras and eyes don't see the same (our eyes constantly adapt, giving us the impression they instantaneously can handle more DR than they actually do while they do the equivalent of changing ISO), it's always a challenging goal to give our images the impression that they match human vision.

RomanJohnston said:

And not sweating them is good advice....nice article.

Click to expand...

Thanks. One of the most liberating things I've done in the last year.

~~~~~~~~~~~~~~

It's easy to argue about equipment and technique, but hard to argue with a good photograph -- and more difficult to capture one .



Gallery and blog: http://esfotoclix.com
Special selections: http://esfotoclix.com/store
Wedding & Portrait: http://esfotoclix.com/wedevent
Flickr stream: http://www.flickr.com/photos/22061657@N03

antoineb said:

these are just two of a large number - this does happen systematically.

in this very case the OVF shot works better because the lettering is perfectly legible. Whereas the LV shot has OVER-exposed the yellow area and this makes the lettering much less cleanly rendered:

darker shot, taken w OVF
http://www.flickr.com/photos/antoinebach/5230173644/

brighter shot, taken w LV
http://www.flickr.com/photos/antoinebach/5229578921/

Click to expand...

Did you block light from coming into the OVF from behind in either of these shots?
--
http://fruminousbandersnatch.blogspot.com/

I figured that as in LV no light reaches the OVF, there was no reason to block it.

Should I have covered the OVF while shooting LV, and if so, why?

thanks

antoineb said:

I figured that as in LV no light reaches the OVF, there was no reason to block it.

Should I have covered the OVF while shooting LV, and if so, why?

Click to expand...

I was wondering if the OVF was blocked in LV (or the camera did it automatically) and not blocked when shooting in OVF mode, therefore resulting in more light reaching the meter in OVF mode and resulting in a darker exposure for OVF than LV. The issue is making sure the OVF is block in OVF mode.

I don't have the camera and therefore don't know how it works, but this might explain the differences in exposure.

--
http://fruminousbandersnatch.blogspot.com/

I purchased a D7000 as a second body to the D3 and for the DX x1.5 on tele.

My initial observations, which I posted on DP, were focus and exposure issues, mainly with my 17-35mm.

Specific concerns were "blown out highlights", especially compared to the D3.

I thought that this was a dynamic range issue, or "pink on green" sensitivity, but it appears to me that Matrix does meter/weight for the background to a greater extent than the D3. On the D3 the 17-35 and Matrix "recognizes" weighs the subject in the foreground more than does the D7000. Backgrounds of Trees, even snow, the D3 handles it all with less input from me.

Out of frustration, and to better judge Nikon's intent/programming, on high DR scenesI have been trying the "Portrait Scene" mode with improved results. Next step will be to emulate/reverse engineer back to Aperture priority.

Hi guys,

very interesting disucssions, and tips, on how to get the most accurate reading from the d7000 in various situations and using the alternative metering modes.

Coming from a d80 to d7000 I don't need any extensive research done to see that the d7000 does over exposure more than the d80 using matrix metering.

It's very clear that's the case, although I thought d80 was as bad as it can get.
With the d80 I only used CW metering which has worked fine for 5 years.
Expected having to do the same with the d7000, a bit disappointed I admit.
That's until I tried the "Fine tuning exposure" setting opportunity.

Using an exposure compensation, only for MM, on -2/6 and the issue is gone, completely.

The exact compensation may vary from user to user or from camera to camera but it now allows me to use MM for all (most) average shooting. CW and SM still has there place depending on what you are up to.
No issue whatsoever with blown highlights any longer.