In my experience the algorithms have never properly learned to handle situations like this one in a consistent way. Most likely, the camera will have suggested that you use fill-flash instead. Do you remember if there was a little blinking zigzag icon in the viewfinder? The way the multi-segment meterings respond to different light distributions is less transparent compared to centre-weight and spot, so it could take more time to figure it out, but they are every bit as predictable as the others.
I remember being in your situation when coming to the Z-1 from the centre-weight meter in the P50. It confused the hell out of me until I learned to know it. :-) IMHO, the real strength of multi-segment metering is that when it goes amiss, it usually does so by a lesser amount than other metering methods. Thus, it is better than the others at supporting autopilot photography. It's more of a shot-saver than a perfect tool. At least to the exposure connoiseur. :-) Cheers, Jostein Quoting Shel Belinkoff <[EMAIL PROTECTED]>: > I saw that and thought that might have something to do with it. But look > at the left side of the subject where the lower shot has a much larger > bright area on the left than the upper shot. I guess the intelligent, > multi-segment meter decided that bright area was of no importance <LOL> > > BTW, had I adjusted the camera to use the way I'd usually use an in camera > meter, I may have used spot metering, or just used the thing in complete > manual mode and determined my own exposures. However, the point was to > learn how the camera reacts to different variables. > That said, the results of this "learning experience" prove, or certainly > lends credence to, what I've always said about built-in meters, and that is > that often small changes in the frame will have a large effect upon the way > a scene is rendered, and that for the most a built-in meter cannot be > trusted to provide consistent and accurate results. > > Further, I was always of the idea the multi-segment or matrix metering > relied upon highly complex algorithms that consider all the details in the > way a frame is light and apply that information in such a way as to give > consistently more accurate results. Feh! I guess I bought into another > marketing maven's fairy tale. > > Shel > > ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program.
