Originally posted by krs Thoughts?
With all due respect, I'm not sure what your two shots demonstrate or prove.
The K100D, like my *ist DS, has a 3200 setting. The K10D does not. I am inclined to think that the 3200 setting is "for real," meaning that it really does provide an extra stop of sensitivity. So far, this is pretty cut and dried.
But in real life, comparisons are not so easy.
I know for a fact--from shooting thousands and thousands of photos in school gyms in the last several years--that the light varies in these gyms from this spot to that spot 8 ft away. I shoot with my K10D in TAv mode a lot, and it's quite interesting to see how the camera changes the ISO from one shot to the next, even though I've often moved the camera only slightly. Your two photos are obviously not pointed at the same spot in the the gym. And to point out the obvious, the photo taken with the K100D at ISO 3200 is pointed at a lighted officials desk, which will undoubtedly make a difference to the exposure. A ton of other things can make a significant difference, including the fact that one photo has mostly white uniforms in it, and the other has a bunch of players in red uniforms, differences in the way you held the camera, and more. To be honest, I am sometimes rather surprised at the variability of the results I get in low light. One shot at ISO 1100, say, might be pretty good, while another shot of pretty much the same subject, taken just a few seconds later, with exactly the same settings but perhaps from just a slightly different perspective might be quite noisy.
If you want to do some comparative tests, you'll need to control the conditions much more carefully. I've done a lot of these tests myself, and the best I can say is that the smaller camera's performance at higher ISOs is only marginally better. My experience is that the K10D can produce usable photos at ISO 1600. Although I've not tested this on purpose in any controlled way, I am pretty sure that I get results from my K10D at 1600 that compare very favorably with the results I get from my *ist DS at 1600 or 3200, after I do a little post-processing.
My impression is that the K20D's superiority at high ISO is dramatically superior to the K10D's. The performance of the older, smaller cameras, however, is in my opinion not dramatically superior. When I shoot weddings and indoor sports where I can't use a flash, I take the K10D rather than the *ist DS. ISO isn't everything.
Will