Histograms and Exposure (light) Meters

[forestkeeper]

Hello everyone. Hope you are enjoying your weekend. I just have a small question concerning our camera's histogram vs. its light meter. Gregory Cazillo, one YouTube photography instructor, teaches that when we are using the histogram to properly expose our compositions, we should shoot to the right side of the histogram, where brighter lighter color data is kept. Mike Brown, an English YouTube photography instructor says for the most part, we should keep the data centered in the histogram, but there will be times when a composition will have more darks than lights and vice versa.

I was shooting at Wahcleena Falls this morning and although I worked to keep my

histo-yellow.jpg (5.3 KiB) Viewed 4679 times

histogram centered or to the right........ my

camera-light-meter.jpg.jpg (4.3 KiB) Viewed 4679 times

light meter was showing that I was way overexposed (where the red dot is). So.... do we follow what the histogram reads or the light meter for a proper exposure?

[BrianEdwards]

I generally stay a little bit to the + side like you were. Shots at '0' turn out kinda dark and under exposed

[TJ_T]

It also depends how you're metering. If you're metering on a highlight or the sky (which i doubt at Wahclella) then your meter may say that you're overexposed since it's only metering the spot you have set. But your whole sensor is capturing data from a lot more than that.. which is what your histogram is going to show you.

[jdemott]

There are two parts to your question: first, where does a proper exposure fall on a histogram; and second, is the exposure meter or the histogram a better tool for setting exposure.

First, if you are shooting jpeg files, you have limited ability to adjust exposure in post-processing, so you should aim for an exposure that gives you the image the way you want it to look. E.g., if you want a very dark, moody image, then should "underexpose." If you are shooting RAW files, then you should generally try to get the histogram centered without any clipping of highlights or shadows...then you can just adjust the exposure slider in post-processing to get the image you want. Technically, there are reasons to prefer a histogram that is shifted a bit to the right, BUT ONLY IF you can do so without clipping the highlights. Clipping the highlights results in a blown out area of the image with no information and no way to recover in post-processing. But under-exposing a little bit results in shadow areas that tend to be a bit noisy or muddy, something that can often be corrected in post-processing. So be very careful if you try exposing "to the right."

Second, in general, I think that the histogram gives you better results overall, particularly if your camera is one that lets you see a preview histogram as well as a histogram in review mode. The blinking highlights feature is also very useful for avoiding blown out highlights. The exposure meter is most useful for understanding what the brightness levels are in different parts of the scene so you can evaluate where important parts of the image will fall in the histogram (and possibly make adjustments like using a split neutral density filter).

The more shots you take, comparing the readings you get in both the histogram and the light meter, the more you will understand about exposure. In the case of your Wahclella example, the white water of the falls should be exposed so that it is a bright white, but not so bright that the highlights are blown out. The camera's light meter tends to set the exposure at a medium gray, so if you metered on the waterfall and centered the needle of your meter, the water would look gray rather than white. If you metered on the dark rocks, the rocks would come out gray and the water would be blown out. A proper exposure will have a histogram with the white water to the right side of the histogram but not all the way to the right, and the dark rocks to the left.

[markesc]

Expose for highlights, develop for shadows...

Raw file types have an advantage for bringing out shadows and rescuing highlights better than having the camera make the decisions for you...so always shoot Raw if you actually want to take full advantage of the camera you bought. Sort of like driving an automatic vs. a manual. One is convenient, but one yields better end results in nearly all cases.

[forestkeeper]

Thanks everyone! I do shoot raw and manual all the time. But there is so much to learn as I am learning portraits, landscapes, sporting events and street photography. I just did a runway shoot today which was pretty fun. If I knew what I do now, thirty years ago, I would have majored in photo journalism. Oh well, by the time I make pro, I'll be in my late 60's. I guess it's never too late.

[Etexan]

A little late to post on this thread but someone may see it. I will be shooting in the gorge next week. Here is my take on exposure:
Your camera's meter tries to average the scene to 18% grey.
If you are shooting where most of the frame is dark greens and a small percentage of the frame is a bright waterfall, the camera will over expose trying to get the scene to 18% grey. You would probably see a large hump in the middle of the histogram (the trees and ferns) with a spike on the right edge (waterfall) probably blown out (overexposed).
I don't rely on the camera's meter in this situation. Using a hand held spot meter, I know from calibration that my camera's dynamic range can safely expose 2.5 stops above 18% grey.
With this knowledge, I spot meter on the brightest spot in the scene (probably the water fall). I can take this reading and add 2.5 stops (moving the bright waterfall from 18% grey to as bright as I can get it without blowing out the highlights. This forces the histogram as far to the right as possible without clipping the whites.

Remember that even shooting RAW, your camera's histogram and display are getting information from a Jpeg.

That's why I use the hand held meter. I know it will give me the correct exposure.
Hope this was not too technical for you but it's the best way to assure a correct exposure.