|[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]|
I think what you're dealing with here is an inaccuracy in the LS-2000's profile. Remember, there are two kinds of profiles, device profiles and non-device profiles. A non-device profile describes an abstract color space (e.g., sRGB, Adobe RGB) as nothing more than a set of primary colors, a white point, and a gamma value. This kind of profile fits into a few hundred bytes. A device profile is a much bigger file (typically hundreds of kilobytes) that contains lookup tables that can describe any arbitrary relationship between numbers and colors, and is used for compensating for nonlinearities in a physical device. The problem is likely that the device profile for the LS-2000 was built for some prototype on the test bench at Nikon, not for your particular LS-2000, and the electronics undoubtedly vary from unit to unit. Ultimately, the best solution is to get an IT8 target slide and some profiling software, and build your own profile. Your web page gives me the impression that you're pretty serious about this stuff, so I'd recommend investing a few hundred bucks in a package that can profile your scanner, as well as a colorimeter for calibrating your monitor. Printer profiling is the next step beyond that, but to do that right, you'll need a spectrophotometer, which is another rather large jump in price. Barring custom profiles, you'll just have to scan with color management turned off, and do it all by hand. Hopefully, you'll only feel the need to do this for a minority of your scans. -- Ciao, Paul D. DeRocco Paul mailto:firstname.lastname@example.org > From: Jerry Lodriguss > > While doing scans of a transparency, I ran across a curious problem with > the Nikon LS2000 losing shadow detail that was captured in the scan, but > not retained when color management was used. > > I've detailed it with pictures and descriptions of my methodology at: > > http://www.astropix.com/CURIOUS/CURIOUS.HTM > > If anyone with any expertise would like to comment on this I would > appreciate hearing an explanation of what is going on here. - Turn off HTML mail features. Keep quoted material short. Use accurate subject lines. http://www.leben.com/lists for list instructions.