exposure metering - why average?

[romanr]

Hi everybody,

Since I recently bought DS and it seems to have an underexposure problem I have read a lot about metering, tests and how all that works. I was very surprised with the the ideology of the exposure settings - average = 128.

I am not a professional photographer I am a physisist. So know a lot about measurement noises and calculations. And as far as I know when the measurement is made the main goal is to have best signal to noise ratio! In case of digital photography that mens that the average must be as high as possible as long as no clipping occures.

The best example is "gray wall" test. Most tests give 128 average with maximum value of 10-15% above ( http://forums.dpreview.com/forums/readflat.asp?forum=1036&thread=15761967 ). THIS IS NONSENSE. So result is almost twice as bad signal to noise ratio and twice as bad color resolution.

Why do the do it like this? I believe that 16 segment metering alowes to do much better.

I appreciate all your comments.

Roman

 

[jeffkrol]

I'm not sure I get your point. But if you have an outdoor scene, mostly of grass, trees, shade and your matrix meter samples all these areas and exposes to an overall 18% grey are you not in effect increase the exposure of the image above what it's true value should be and does this not increase your signal to noise ratio?

Any dark scene that is averaging below 127 will have an exposure increase to 127. You now have the choice to increase it even more thus again increasing s/n ratio Those averaging above will have its exposure decreased. The decrease will save the high end from blowout but then it is for you to decide to manually increase the exposure to boost your s/n ratio back up. Balancing act.

Not sure if this makes sense but that's how I understand it. As with everything in life you have to start somewhere. I don't really believe matrix metering is very good for any camera for the most part (Did I waffle enough?)and would prefer to spot meter when I can.

Nikons what... 20,00 point system and built in logic may work ok. Don't think Pentax's is that advanced though.
--
360 minutes from the prime meridian.

 

[rke21]

You get more noise as you set the iso higher. Like getting a faint signal on the radio and turning it up.

The shadow area has more noise. Less light. So you can see more of the noise.

The exposure system is good. Really just a robot. You're going to get some kind of clipping in a high contrast scene. At some point, there has be a black on one end and white on the other end. If your sensor has a lot of lattitute, the black point and white point is going to be farther apart.

 

[rke21]

Nikons what... 20,00 point system and built in logic may work ok.
Don't think Pentax's is that advanced though.

You have to get used to the algorithm of the camera you're using. One system isn't really better than the next. it's different for everyone.