Experiments with Nikon D600 exposure - seems too aggressive

Started Sep 21, 2012 | Discussions thread
Mako2011 Forum Pro • Posts: 24,580
Re: Experiments with Nikon D600 exposure - seems too aggressive

creaDVty wrote:

Focus is one thing but even matrix meter sees your placement of the focus point (focus bias) as an instruction to meter that point as the mid-tone....to some degree.

I agree that placing the focus on a particular point in the scene biases the camera to meter for that point. However, I don't think it necessarily meters that point as a midtone. For example I believe it can guess when you're shooting a beach scene or snow scene, and if you had the AF point on a snow man, there's a reasonable chance it will be metered correctly as white or light gray, not middle gray. Is that what you meant by 'to some degree'?

Remember there is no guessing going on. Matrix metering looks at all the zones of the RGB meter and uses algorithms to match an exposure solution preprogrammed in. It then also looks at the focus point and adjusts the exposure solution slightly based on that target/subject. So if matrix metering sees a snow scene it will select an exposure solution appropriate for that.....if you also put a black car under the focus point then it will adjust the exposure in a direction similar to had you spot metered on a black car and the entire result will be off. By "to some degree" is what I mean by how far off from a snow scene.

As for the car, we know it's black but the camera doesn't so yes it has to guess whether it's black, white, or gray. If the entire scene had the same tone (all black or all white or all gray) then yes it's impossible for the camera to guess the actual tone. But I would have thought that given that there was a large white object nearby, that the meter would not be fooled. I'm supposing the camera had two options:

1) The car is gray. The nearby pickup truck is a white object that is intensely lit by the sun.
2) The car is black. The nearby pickup truck is white.

Or 3) The matrix algorithm dictates this solution for the scene but the focus point is on a zone 3 object so it will shift the exposure to honor "focus Bias". And that is what it did.

I would have thought the camera could correctly guess it was the 2nd scenario. Besides, it couldn't hurt in terms of preserving detail because option 1 would mean that the white object would be blown whereas option 2 would retain most of the information. What if this were a wedding? What's it gonna do with a bride and groom? Blowout the expensive wedding gown and have a dark gray tux? I hope it's smarter than that. Maybe not?

Try this.....Matrix meter. A or S priority. White wall background. Bride in white dress fills 2/3 of scene on left. Groom at same distance in black tux fills 2/3 of scene on right. Camera on tripod so scene in viewfinder does not change. Two shots. One with focus point on groom chest and one with focus point on Brides chest (focus distance same in both). You will get two different exposures...both wrong. That is why the black car scene was "off to a degree" due to the focus bias.

Well, I hope no one shoots me for saying this, but my approach would have been to underexpose the scene sufficiently to not blow the sky, then bring up the exposure for the midtones in post. It appeared from the histogram that there was still a lot of room left in the shadow side.

Not a bad strategy....meter for the mid-tones then check the highlights

I don't interpret my strategy the same way you do. To me, I would summarize it more like: protect the relevant highlights, and let the rest of the tones fall where they may. I don't look for a midtone value then meter it. Maybe I ought to, but I'm just saying that's not something I do right now.

But you did exactly that. Just in a different order...you checked the highlights first and put them in the correct place on the meter...zone7. That then puts all the mid-tones in the scene at Zone 5 unless the highlights in the scene are actually not up at zone 7. Meter for the mid-tones and check the highlights gives the same result really as when you check you bring the highlights back to zone 7 if needed. Your method is also a good one and very appropriate for many situations.

Very helpful and relevant, just like the rest of your advice. Thanks again!

I learn each time I read what you ask and see if I think I know the answer. I really appreciate you asking. I'm learning too.

Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow