Here's a little (or long...winded if you prefer) writeup I did yesterday for a thread about Manual exposure and metering that suggests an experiment that could be fun and maybe useful:
Here's a little experiment to both understanding how your meter works and how you need to work with it whether shooting in Manual, Av, Tv or P, and also give you some insight into how you can proceed in learning Manual mode. By the way, I'm a proponent of learning to use all these modes because there are times when each can be advantageous, even the P mode!
For this experiment you want good, bright light, like a bright day light (if you are indoors you would do well to set a White Balance as well as have a good bright light source.
Also worth noting: first, to get the most latitude in post processing you should shoot in Raw when facing challenging lighting conditions. Also, for what I'm about to lay out, you should be aware that even if shooting in Raw, the camera histogram, preview/review, and the highlight alerts will reflect the in-camera histogram and jpeg, which will not give you the most accurate results if shooting Raw, especially regarding highlights and, for bright colors, saturation. To really push things, you should have your Picture Style set to Neutral and then turn your Contrast in that picture style all the way back. Otherwise the histogram and the highlight blinkies will try to tell you that you are screwing up when in fact you are not if you are shooting Raw.
Next, you want two items -- one bright white and the other black. They should both have some detail on them, so something like a towel or an article of clothing as long as they are properly white and black could do.
Think of this as practicing for a wedding shoot -- the white article is a bridal gown and the black is the groom's tux. Lay them in a way where they have the good light and where you can get a decent framing on them (they don't need to fill the frame, but should at least fill the center metering circle).
For reference, take that gray card and set it between them or next to them in the same lighting.
Set your camera to one of your "program" modes (Av, Tv or P) and to Spot metering.
To start, you can just observe the meter as you focus on/meter each of these objects. You will note that the "needle" hits just about center for each. What this means is that the camera meter has been programmed to interpret a subject you are using to meter as "medium" in tone and, if you take that shot your exposure settings will really try to render that subject as "medium". In other words, snap a photo at that black object and it will come out way brighter than the black it really is, although everything else will be seriously overexposed. And that bridal gown will be totally blown. Or, if you focus/meter on the white object, again the camera will bring down the exposure to make it "medium", meaning of course that that groom's tux will lose all kinds of meaningful detail.
So, what do we do about this? Well, where Manual shines is when, if you are say shooting that wedding in consistent light, you can actually use that bridal gown or the tux to set a good exposure for the whole scene.
So, try this: set the camera in M now and go back to focus/meter on that white dress. What you will need to do is tell the camera "this is bright white" -- adjust your shutter speed, aperture, and if needed your ISO until the meter needle is a full two stops above the center and, guess what? You are close, although for real white in bright light, that is underdoing it -- try 2 2/3 stops if you really want to nail it. For this, for cameras other than a 1 series this may go off the scale (many if not all of the xxxD and xxD bodies limit the scale to 2 stops) but, like I said, real white in good light actually can be shot at three stops +EV. So above 2 stops in those conditions or in a bit more subdued lighting +2EV can pull you through.
Now, if you take that shot and include all three items in your frame, you will see that, hey, black is actually black and gray is actually gray and white is actually white! Shazaam!
Now do this: keeping the same camera settings, focus/meter on that gray card. You should find that the needle lands nicely at just about the center! If you had actually maxed out the white at +3 EV the card actually would have landed a bit brighter than the center, because of how the meters are calibrated. If you got spooked and set the whites to just +2EV, well it would be darker.
And then, do the same with the black article. Again, depending on how you exposed the white, and also how "black" the black really is, your meter needle should be at least -2EV or a bit darker -- to find out how dark on a prosumer body you would have to tweak your settings to brighten things until the meter registers -- now you know. But, when you get the shot into post processing you still will have usable detail in both the whites and the blacks, as long as you don't try to boost the blacks to gray!
Then, try this: hold your hand out with the palm up to where it gets the same crisp light and note the meter value -- for example a caucasion skin tone will typically be about +1EV...can you see how this can be a substitute gray card? Or, if you have a nice blue sky overhead, you can focus/meter it and it will be typically a bit brighter, somewhere between +1 and +2 EV. In other words, you don't need a gray card, bridal gown or tux to nail down exposures -- you learn how to work with things at hand.
Now, with this experience in hand, it can be a real pleasure when you are shooting a scene with consistent lighting within your field of view, even given some shadows and such -- you can fire away one shot after another and get the same consistent results. But the drawback with assuming all is fine is if, say, you see something in your peripheral vision and turn so that your orientation to the light source has changed, so the scene lighting is now changed. That calls for some quick action to re-figure the lighting and your exposure values. It happens all the time when out in the field. You will typically learn from your mistakes.
All this is just getting started. Like I said earlier, I'm a proponent of learning about all your modes, and also there are the various metering modes that each behave differently. Sometimes Evaluative metering will be the best approach, maybe with one of the Program modes. It's all a matter of experiencing and learning as you go.
_______________
