Originally posted by audiobomber I chose Auto ISO, the camera chose ISO 100. The K20D can boost the ISO in flash mode, I've seen it. So why didn't it?
I don't know, but it must have thought it wouldn't need to. And indeed, had your subject not been too far away for the flash to reach, and not you not had highly reflective snow right in front of you that cut off the flash prematurely, that exposure probably would have worked just fine.
How do you respond to my observations about the distance of the subject and the reflectivity of the foreground snow? Do you not see how those two factors complicate the situation? If your flash was a laser beam that could have lit the bird only - and was strong enough to do so at the distance you were - then you'd expect good results, sure. But that flash probably reflected off the snow at your feet, which reached the camera sensor and made it say, "woah - too much light! too much light!" - thus cutting down the flash output far too much to reach the bird. The area at your feet that probably caused the bright reflection reflection doesn't show in your picture, so you aren't aware that this happened, but I'm betting if you had taken the same shot with a wide angle lens, you'd see exactly what I'm talking about - bright snow at your feet, getting progressively dimmer as you go further out. And since the flash would was insufficient to illuminate the bird at f/8 even at *full* power, no surprise that you get an underexposed shot in a case where the camera is scraming at the flash to cut down on its output because of the bright reflections in the foreground.
Quote: I wanted the camera to take the same exposure it metered without the flash, but I wanted to add fill flash. Instead it metered way below.
I think my explanation makes perfect sense and probably predicts exactly the results you are seeing, but since I'm not a flash expert, I'll bow out of any further attempts to try to explain *exactly* what was going on, since I don't know for sure. Maybe someone else can explain the exact procedure followed by the camera.
Instead, I'll suggest you look at it a different way. You say you wanted the exposure the same. But you didn't present the same scene, nor did you ask the camera to do the same thing. In the first case, you presented a scene lit by the sun 93 million miles away from every point in the scene and asked the camera to meter off that and set an ISO and shutter speed. That's a straightforward task. In the second you presented a scene that was lit evenly by the sun but also very unevenly by a flash unit about five feet away from the highly reflective foreground but 30 feet from the bird, and asked the camera to not only meter set ISO and shutter speed but also to set the flash output. So you gave it scenes that were lit *very* differently and then asked the camera to do different things, too. I don't find it that surprising that this led to different results.