TeleFragger wrote in post #13620431
so in class we learned about the histogram.. far left = underexposed, far right = over exposed and in the middle is pretty much 18% grey (i know 1/2 black and 1/2 white is 50% grey..so not sure why they say 18%...)... anyway..
what the instructor told us to do (this was in the landscape portion of the class) was to look for the bright spot, dark spot and neutral spot.. meter off of the neutral spot and move around and make sure you are not more than 2 stops towards over exposed as you will loose quality.. you can go more than 2 stops in underexposed and get it back....
now he said take the pic and make sure the histogram shows up the middle.. so instead of me using the white shirts couldnt i use the ground? you see the asphalt looks to be a mid tone? or am i looking at that wrong? i think out of the class the white balance (not gone over in detail) is where im struggling..
I can't see your photos at work, but I don't really think the above is a good explanation of the histogram.
Don't think of the histogram as showing you whether the photo is overexposed or underexposed. Instead think like this: the histogram shows you how much bright stuff is in the photo, how much dark stuff is in the photo, and how much medium-toned stuff is in the photo.
A handful of examples:
If you take a picture of a polar bear in the snow, the histogram should be a huge hill on the right. That's not overexposed - it's correctly exposed, because you're taking a picture of a bunch of white stuff. If you were to take a photo that put the "hill" right smack in the middle of the histogram, it would be underexposed - the snow would look dirty and grey, and details in the shadows (bear's nose and eyes) would be completely lost. But the camera thinks it's right, because the camera doesn't know any better, and just tries to make everything "average" with the hill in the middle.
If you take a picture of a bride in a big white dress, standing in front of a black wall, you're going to have a big spike near the right edge of the histogram (the white dress), and a big spike near the left edge (the black wall), and a very pronounced valley in the middle. The middle won't be completely flat, because you've got other mid-tone elements such as the bride's skin and hair, and shadowed areas of the dress. Correctly exposed, and no hill in the middle. If your spikes are running up against the edges of the histogram, you're either underexposed (left spike falling off the edge, you're losing detail in the black wall) or overexposed (right spike falling off the edge, you're losing detail in the white dress). HOWEVER! sometimes the camera just can't get the detail in the shadows AND the highlights, and you have to make that choice - do I lose detail in the dress, or the wall? I would usually choose to lose detail in the wall, and retain all the detail in the dress.
At a hockey game, the ice in the rink is mostly white. The hill should fall closer to the right than the middle. You'll have some elements in the middle and low end of the histogram: the markings on the ice, the players' uniforms, etc. But if you're shooting from above and most of your shot is the ice in the rink, you're going to need a little more exposure than simply putting the histogram and the meter needle in the middle. If you're shooting from ice-level, you'll probably have about half a frame of ice and half a frame of bleachers/spectators/other - so your histogram will probably have a spike near the right (the ice), then probably taper off toward the left.
It's up to you, the smart photographer, to look at the subject matter, and look at the histogram, and determine whether you're getting the right exposure. You can't just try to aim for the middle all the time, or your results will never be any better than the "average" that the camera wants to make. The camera isn't smart, it doesn't know what you're taking a picture of, it just wants to put everything nice and safe and in the middle, and it doesn't know if that's the right thing to do or not.
Also, note that you use the histogram primarily for exposure, not necessarily for white balance.