Hello Simon. > just on a side note: the "filled" highlight in your screenshot looks > aweful to me - just filled with some artificial blue that can't be found > anywhere on the car hood. And if you look closer you can even see an > edge at the position where the blown highlight starts. This doesn't look > like "reconstruction" (which would mean there was some information at > all) but more like some "filling with what RT thinks might suit".
I guess we have different expectations and requirements as to what highlight reconstruction means and should do. Certainly there are quite a few people who need what RT offers or no one would have bothered implementing it. Yes, there is mach banding when I use very extreme settings in RT. But I can just adjust the 'wrong color guess' (the yellow tint) around the actual highlight (whose color is pretty spot on, I'd say), by adding come local color correction elsewhere in the pipe. The real issue is detail (and thus, highlight size). The reconstructed highlight in RT has details brought out that I simply cannot get back with DT's color reconstruction. At the end, the highlight in DT always has about twice the size of that you see in the screenshot of RT. Try yourself. .mm ------------------------------------------------------------------------------ Put Bad Developers to Shame Dominate Development with Jenkins Continuous Integration Continuously Automate Build, Test & Deployment Start a new project now. Try Jenkins in the cloud. http://p.sf.net/sfu/13600_Cloudbees _______________________________________________ darktable-devel mailing list darktable-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/darktable-devel
