D40x matrix metering focus dependence

In matrix metering mode, the exposure setting of the D40x appears to depend on

the focus point even with the AE lock option set to OFF. With AE lock ON, the

effect is the same, but this is expected. Below are examples showing what

happens when (1) the focus point is placed in a dark region and (2) when it is

moved to a bright region. This is a 'feature' of the D40x that I'd like to be

able to defeat.<p>

<img src="http://img403.imageshack.us/img403/8360/matrixfocusdarkts8.jpg">

<img src="http://img403.imageshack.us/img403/2605/matrixfocuslightgk0.jpg">

Berg, It is an excellent "feature". I make use of it!

Use Center-weighted metering instead.

The logic behind the feature is that the computer thinks your focus point is on your subject,

which you will want "properly" exposed. "Properly" meaning something of a midtone.

 

There's no way to turn it off, but there are many work-arounds, such as using AE lock

(metered on the subject of your choice), exposure compensation or good old manual

exposure.

Thanks for your replies.

 

Brad - I agree that most of the time you'd want the region in focus to be properly exposed, but there are situations where you just want the overall scene to be properly exposed regardless of the focus point. Center-weighted metering is not a good solution.

 

According to the manual, if the AE lock option is set to OFF, "pressing the shutter release button halfway does not lock exposure, if ON is selected, exposure will lock when the shutter release button is pressed halfway." This provision apparently does not work as stated.

The D80 works similarly. I now use centre weighted metering with generally more reliable results and need to reach for the exposure compensation button less often and more reliably.

"Center-weighted metering is not a good solution."

 

> Perhaps the problem is the D40x then, since I can adjust my D200 and D300 to average the scene similar to Matrix metering without giving added weight to the luminance value under the selected AF point or if the scene is brighter or darker on the top (just try turning your camera upside down and see what happens to the exposure, with a bright sky and dark foreground the differences in exposure will be dramatic). For me Center-weighted metering set to average is a great solution and gives me the EV I would have selected over 95% of the time.

I think that center weighted metering is better suited for a scene that's evenly illuminated, but when there are both bright and dark regions in the scene, matrix metering tends to work better.

 

I guess what really bugs me is the fact that the description for the AE lock function does not reflect the actual operation of the camera. As you and other posters have mentioned, there are different ways to work around the problem, but they all involve extra steps.

Berg, you're confusing two separate issues here. The AEL option does work as described,

but the camera is programmed so that exposure depends on the active focus point.

Therefore, the camera must choose a focus point before it evaluates exposure.

 

Hopefully this will help clarify AEL: Activate AEL on the shutter button. Press the shutter

release allowing the camera to choose the AF point and evaluate exposure. Now, with the

button still held down, change composition, and change it a lot, so that the new exposure

is very different from the initial scene. (For example, choose a scene in shadow, then

change composition to include mostly sky.) With AEL active on the shutter release, and

with the shutter release held in the half-way position, the camera will not allow the

exposure to change. Try the exercise again, without AEL active on the release and the

camera will allow the exposure to change and match the second (sky) scene.

 

Hopefully this will help clarify the AF point exposure issue: With the camera on a tripod

compose a very high contrast scene, and one in which a very dark area covers one focus

point and a very bright area another. Using the mulitselector on the back of the camera,

toggle back and forth between those 2 focus points. Since you are using a tripod, the

composition isn't changing so any changes in exposure are based on Matrix metered

recommendations (which in turn are based on the active focus point.)

Berg, I think your problem is more about the D40x and how Center-weighted metering is implemented than it is about the Matrix Metering. On the more advanced cameras you can change the size of the center circle used, going from 6mm to 13mm, along with my favorite which is "average"; these adjustments are not available on the D40x. Using Center-weighted set to average I rarely see a difference compared to Matrix except that the exposure is not as overly influenced by the presence or location of highlights using Center-weighted. The D40x is also a problem for Matrix metering because there are a limited number of AF points, and Matrix metering is implemented differently in different models based in part (but not completely) on the number of AF points.

 

With my D200 and D300 cameras, I rarely have any exposure issues under even the most demanding circumstances. It is worth noting though that I shoot in RAW which provides more DR, and shooting at or near base ISO utilizing ETTR I have a lot of latitude to raise shadows while avoiding clipped highlights. A little more time spent with my files in PP translates for me in much less time fiddling with compensating and adjusting settings while I'm shooting. For me this makes the most sense since I don't want to miss fast changing light or a brief moment messing with the camera.

This was a big disappointment I had with the D80. It has a Matrix metering system that

works very differently from that of the D70, the D200, and D300. If I want a spot meter, I'll

choose that. Matrix metering is supposed to take the ENTIRE scene into account, and not

bias toward any part of it, and average and compute the best exposure for the entire scene.

The D80 clearly did not do that, and I lost good photos because of it (and was constantly

spinning the +/- dial to correct the meters weaknesses. Fortunately the D300 has a meter

that is a lot more consistant in this regard.

Brad - I appreciate your explanation, and I think that you are correct, but I don't quite agree with your conclusion that the camera is working as described since it's never stated that the focus point will have any influence on the exposure value of a scene in the matrix metering mode.

 

As described in Nikon literature, the matrix meter essentially measures brightness in multiple areas of the scene (multi segment) then compares the different brightness values and evaluates the contrast range of the scene. The information is then compared against a reference database to decide on an appropriate exposure value. It appears that the algorithm used in the camera also puts added emphasis on the brightness of the segment when the focus is locked on.

 

Anthony - I agree that the 1005-pixel metering sensor in the D200/300 is a lot more sophisticated than the one used in the D40x.

 

Thank you all for your replies.

Dave - Thanks for your comments about the D80 having similar characteristics to the D40x. It would be very helpful if Nikon were to better document these differences in their products. This is definitely not a desirable 'feature' in my view.