From: Greg T. <gd...@ir...> - 2007-09-24 00:34:41
|
I have been unable to get this image to render decently using any stock version of dcraw or ufraw. If I uncheck the "Use matrix" box, the colors look normal, but very muted; if I check it, the saturation comes back, but the sun and surrounding highlights are bright yellow (they should be largely white with a reddish or orange tinge, similar to the colors further away from the sun). Can you explain where your 'should' notion comes from? Is it against in-camera JPG or canon's raw converter, or your memory from taking the picture? I've been starting to think about what "correct" raw conversion with all settings at default means, and beyond thinking that it matches in-camera JPG with settings at "normal" (not "auto"), this has mostly resulted in my head hurting and not too much useful. Camera WB says 5276K, but if I go to Exposure Comp -2 and spot wb on sun, it says 2269K. Did you happen to measure incident sunlight with a spectrophotomer? The dynamic range is pretty extreme in this photograph. One thing seems relevant that I have never understood. The clip state is about 'positve EV', but there's highlight clipping with 0 EV. I don't understand how clipping behavior and whether EV is positive or negative are related at all. |