How "You" Judge Exposure

[smoke665]

I realize this may be somewhat of a rhetorical question, as a lot of it boils down to personal bias, but I'm curious as to how others judge whether they have a "correct exposure" for an image. Histograms give you distribution of values, but can vary wildly from one image to another when the tonal ranges are different. Do you use any type of qualitative or quantitative metrics to give consistency in your images? How do "you" know when "your" exposure is correct?

 

[NS: Nikon Shooter]

Do you use any type of qualitative or quantitative metrics to give consistency in your images? How do "you" know when "your" exposure is correct?

Just as WB is the least important parameter at SR of any take,
consistency is the same when it comes to a series of images…
given they were not taken in the same light conditions that is.

Qualitative or quantitative metrics? Sure: the histogram. Say
you're out in the city for some street photography, chances are
that many subjects will catch your attention and I'll bet none in
similar light conditions. The ONLY thing that is of any importan-
ce is the quality of the recording — the horizontal line at the bot-
tom of the ever important histogram.

The histogram IS your (mine too) best friend because it confirms
or not that the recordings are right. It is only through the quality
of the said recording that consistency will be possible.

Have a good time, smoke!

 

[PJM]

Just as WB is the least important parameter at SR of any take,

What is "SR"?

 

[NS: Nikon Shooter]

-

Sorry Pete… shutter release.

 

[snowbear]

Histograms are fine for digital, but film doesn't really have them.

 

[NS: Nikon Shooter]

Histograms are fine for digital, but film doesn't really have them.

That's where incident light metering comes in!

 

[Strodav]

It starts when taking the photo. Shoot raw, which gives you maximum latitude. I often use a 3 square gray card (white, black, 18% gray) to help me set exposure in the field and take a shot of the gray card to help adjust exposure and white balance in Post, but the easiest thing is to watch the blinkies and adjust exposure to eliminate them. Another easy thing to do is take the shot then look at the histograms, but remember the histograms are of the jpg image you are looking at on your cameras lcd panel. So all your jpg settings, even if you are shooting raw, are being shown in the histograms. That's why you hear ETTR because the camera usually gives you about 1/3 of a stop more room on the high end when shooting raw. If you can't make the blinkies go away or make the histogram fit, then the dynamic range of your subject and background is greater than the dynamic range of your camera. First thing to do is lower ISO as your camera's maximum DR is at base ISO. Next is to chose to either bury shadows or blow out highlights and adjust exposure accordingly. The technique I use more and more is to bracket 1 or 2 stops up and 1 or 2 stops down and fix it in Post.

 

[AlanKlein]

It starts when taking the photo. Shoot raw, which gives you maximum latitude. I often use a 3 square gray card (white, black, 18% gray) to help me set exposure in the field and take a shot of the gray card to help adjust exposure and white balance in Post, but the easiest thing is to watch the blinkies and adjust exposure to eliminate them. Another easy thing to do is take the shot then look at the histograms, but remember the histograms are of the jpg image you are looking at on your cameras lcd panel. So all your jpg settings, even if you are shooting raw, are being shown in the histograms. That's why you hear ETTR because the camera usually gives you about 1/3 of a stop more room on the high end when shooting raw. If you can't make the blinkies go away or make the histogram fit, then the dynamic range of your subject and background is greater than the dynamic range of your camera. First thing to do is lower ISO as your camera's maximum DR is at base ISO. Next is to chose to either bury shadows or blow out highlights and adjust exposure accordingly. The technique I use more and more is to bracket 1 or 2 stops up and 1 or 2 stops down and fix it in Post.

Let's say you use to shoot with a camera that had 10 stops. Now you upgraded to a camera that has 13 stops. Why would you worry about a 1/3 of a stop ETTR?

 

[smoke665]

The histogram IS your (mine too) best friend because it confirms
or not that the recordings are right. It is only through the quality
of the said recording that consistency will be possible.

I use the histogram both in camera and in post to gauge where I am, but a histogram doesn't always portray what you want the final image to be. Some would have you believe that if you just produce a symmetrical Bell Curve, with each edge touching the ends you'd have a perfect exposure, but in reality you rarely see a Bell Curve. The histogram represents 256 shades of gray with 0,0,0 being solid black and 255,255,255 being solid white. In the real world, tonal distribution is all over the place, in a high key for example you'd see data loaded to the right, while low key will show it all on the left. and for some you might only have data in the middle. For example here is one from an accomplished photographer that posts on TPF the image is beautiful.......but it you were to only use the histogram, to gauge the exposure, you wouldn't think so.

There's almost no data in the shadows and almost none in the bright highlights. Going the opposite way, here's one by a certain photographer whose work you're very familiar with. It's a low key image, also very well done.

But again, if you were to only use the histogram to gauge the exposure. this would have ended up in the trash. That's when the Qualitative Metrics come in.

but remember the histograms are of the jpg image you are looking at on your cameras lcd panel. So all your jpg settings, even if you are shooting raw, are being shown in the histograms. That's why you hear ETTR because the camera usually gives you about 1/3 of a stop more room on the high end when shooting raw. If you can't make the blinkies go away or make the histogram fit, then the dynamic range of your subject and background is greater than the dynamic range of your camera.

Good point, I've learned that I can go all the way to the blinkies without blowing the whites. I don't chimp every shot, but in the past I've always leaned toward creating a full data file (at least some data stretching from the left side to the right. Now I'm beginning to wonder if the quest to fill the file, only to clip it in processing isn't an exercise in futility. Your camera sensor, or you handheld meter is going to expose for 18% gray, point it at snow, sand, or other reflective object and it will overexpose. It requires reducing what it tells you by 1.5-2 stops. So, here again it's experience behind the camera that gets the correct exposure.

 

[NS: Nikon Shooter]

In the real world, tonal distribution is all over the place…

In the real world? This sounds like an absolute and I know that the only
absolute is relativity. In the real world, in my book, everything is possible
from the the narrow to the wide bell and from the U shape to the spike.

When a bell is seen — like in the first pict— I see that the extent of avail-
able DR is greater than the recording needs. that would render a dull ima-
ge that could use DRL = Dynamic Range Levels = black and white points
setting… technically. Artistic intent has its own set of prerogatives.

 

[Ysarex]

I realize this may be somewhat of a rhetorical question, as a lot of it boils down to personal bias, but I'm curious as to how others judge whether they have a "correct exposure" for an image. Histograms give you distribution of values, but can vary wildly from one image to another when the tonal ranges are different. Do you use any type of qualitative or quantitative metrics to give consistency in your images? How do "you" know when "your" exposure is correct?

I expose all my photos the same way. I know my exposure is correct when the sensor in my camera is fully utilized. My goal behind the camera then for all photos is the same; place the diffuse highlight at the sensor saturation threshold and click. I love digital -- it's so damn easy!

 

[smoke665]

n the real world? This sounds like an absolute and I know that the only
absolute is relativity. In the real world, in my book, everything is possible
from the the narrow to the wide bell and from the U shape to the spike.you'

When a bell is seen — like in the first pict— I see that the extent of avail-
able DR is greater than the recording needs. that would render a dull ima-
ge that could use DRL = Dynamic Range Levels = black and white points
setting… technically. Artistic intent has its own set of prerogatives.

Oh I agree, there are all sorts of possibilities, that's what I was pointing out, the histogram in and of itself is a guide, but not the authority. In the first histogram, you would think the image would be dull, but it wasn't. I see this type of curve in the soft and dreamy type images. Yes it was artistic intent by the photographer, in fact it's pretty much his style.

I expose all my photos the same way. I know my exposure is correct when the sensor in my camera is fully utilized. My goal behind the camera then for all photos is the same; place the diffuse highlight at the sensor saturation threshold and click. I love digital -- it's so damn easy!

Yes, but my question was judging the "correct exposure for the image". You can have an image technically correct for exposure, but sadly lacking from an artistic standpoint.

 

[zulu42]

I start by thinking about the final image and where are my exposure related priorities, then spot meter in camera for highlight control, shadow detail, or a middle gray depending on the subject and image intent. examples: a white car or bird gets metered for highlights, a black dog I make sure to meter for some detail in the black fur. Shooting people, family on the run I might occasionally meter my hand in the viewfinder and center the needle.. I personally don't use a histogram until I see it in post. With digital I more often meter to preserve highlight detail. Let's admit that today's sensors have amazing latitude if you don't blow out either end of the range. With film I'm generally more slow and careful with exposure if the camera has a trusted meter. I love the predictable results of careful exposure. For cameras with no good onboard metering I use a phone app or use sunny 16 estimates which also work well after some practice.

 

[Ysarex]

Yes, but my question was judging the "correct exposure for the image". You can have an image technically correct for exposure, but sadly lacking from an artistic standpoint.

That's a bit trickier. There is no "correct" lightness for the finished image -- that's an artistic judgement call and so there's no measurement either. But for starters if there are any rules at all in photography this is rule number one: A good photograph has a black point. If the image doesn't contain black somebody screwed up. There will always be exceptions but that's the rule with the fewest exceptions. After that blown diffuse highlights are a mistake. The photographer who claims they blew the highlights because that's the effect they want is usually trying to cover up failure. Again there are of course exceptions.

So after the black point is set the diffuse highlights should be placed as bright as possible without clipping. If the image then appears too light or too dark or too flat or too contrasty then adjust the midtones holding the black and white end points in place.

 

[Rickbb]

I judge mine based on how close it looks to what I had in my mind when I took it. And how much I like the way it turned out.