You are over simplifying in a big way.
First there were averaging meters, they gave equal emphasis to every
part of the frame. They were very susceptible to giving poor exposure
in backlit and other non-even lighting situations.
Then came center-weighted meters. The gave much more emphasis to the
center of the frame and very little to the edges and corners. They were
considered a HUGE advance over the averaging meter for most situations.
Generally the subject is centered or while metering can be centered.
The meter will properly meter the subject with very little influence by
a dark or light background (only visible in the corners/edges). These
meters lasted as the most sophisticated for decades (1965-1990) and were
the basis of Nikon's excellent reputation (F2).
Evaluative metering can vary the emphasis on different parts of the
frame depending on what the meter sees. A set of parameters is stored
in the cameras memory to make these decisions. For example, the top two
segments see much more light than the bottom two. The camera assumes
that this is a landscape (top two segments see sky) and changes the
emphasis to pay more attention to the bottom two segments. This is a
simple example, but a good one. The more segments, the more patterns
that can be saved.
This is the current state of the art, with more segments/patterns being
assumed as better. One advantage to having more segments is that the
metering can be weighted on the focusing point so that you don't even
need to reposition the camera to meter.
Mr. Bill
Thomas Bantel wrote:
>
> Both, evaluative and center weighted metering work quite good with
> "normal" scenes, which are "medium bright" on average (18% gray).
> The evaluative metering tries to be intelligent and automatically
> compensate for tricky lighting (backlit scenes). The center weighted
> doesn't. It strictly calculates a center weighted average of the
> brightness of your scene and assumes this should be 18% gray.
*
****
*******
***********************************************************
* For list instructions, including unsubscribe, see:
* http://www.a1.nl/phomepag/markerink/eos_list.htm
***********************************************************
