Paul, I agree with you concerning using the matrix metering. I think Canon used to, and maybe still does, call their equivalent metering "evaluative metering". I don't use matrix much because Nikon says "matrix" is based on algorithms derived from "thousands" of pictures. But all these thousands of pictures are almost certainly topside pictures. And since I don't know the thinking Nikon used, I cannot predict what the camera might do consistently enough to really count on it in my underwater photography.
I do use center-weighted metering. A lot. Especially when photographing a subject, like a fish, that is moving around in the water column. Because of this movement, the color of the water behind the fish varies continuously. In this type of situation, I find it almost impossible to keep changing the camera exposure settings (aperture or shutter speed) based a manual spot meter setting which would keep the water color consistent. And still keep my attention on composition and focus. If the fish moves 15 degrees to up or down in the water column, the water color brightness behind the fish can change by at least a full aperture. If the fish moves 90 degrees or more to either side, the angle relative to the sun changes dramatically and the water color change can be two or three apertures. In this type of situation I find that using the center-weight metering in combination with shutter-priority auto-exposure gives me a much more consistent "background" exposure. And TTL flash allows the fill flash to be about right and compensates as the fish moves closer or farther away. This technique also has the advantage of maintaining a good ambient light exposure component if a cloud passes in front of the sun and the overall light level drops. The camera just opens the aperture setting to compensate. And TTL flash adjusts to the wider aperture. This technique has worked so well that I now use it on virtually all of my shots where water color is important in the picture. My thinking on this is that if I can delegate this technical aspect, exposure, to the camera, I can give more of my attention to focus and framing the subject.
I will attach a few shots from earlier this month in Bali where I used this shutter-priority exposure technique.
If I've done it right, there will be five pictures. There are two pairs of shots - a sweetlips with butterflyfish and a cleaning station with surgeonfish. The camera settings on both pairs of shots were the about same. If you look at the sweetlips shots, you will see that one shot is virtually head-on while the other is angling upward from below. If I had used manual spot metering, I would have had to manually readjust the shutter speed or aperture to keep the water color such a consistent blue. Shutter-priorityauto exposure did it for me. The shots of the surgeonfish cleaning station are from totally different angles. And in one I am much closer. But the proportion of blue water in both is about the same, so auto-exposure kept the overall scene very similar and the TTL flash compensated for the difference in distance. My white balance in the two surgeonfish shots did change noticably because in the closer shot, the flash had been filtered by less water and was warmer when it got to the fish. But that would also have been the case if you were shooting everything on manual. The fifth shot is just a crinoid shot I liked, again shot on shutter priority with TTL flash.
Paul, I guess we will always disagree on whether housing color makes a difference. All I can say it that topside wildlife photographers use camou clothing, and even camou tape on their tripods and lenses to get closer to their subjects. Should we believe that marine creatures are so unobservant and so unwary that it wouldn't make a difference with them, also? I believe it does and so I try to minimize bright things that can grab their attention. I believe a darker, less reflective color is better than a bright color. And believe that a camou or disruptive color pattern is still better.
Edited by divegypsy, 20 May 2012 - 05:23 PM.