Optimum Dynamic Range
2003-01-25 by Paul Roark
Is there an optimum dynamic range for a print? A Kodak paper states the following: "The range of lightness levels that the eye can perceive at any one time is, at most, several hundred to one. ..." See, <http://www.kodak.com/US/en/corp/researchDevelopment/technologyFeatures/oled 2002Lum.shtml> If we are trying to get the maximum impact from a piece of paper/print, increasing the dmax has been a goal. However, I think we are seeing decreasing utility from the increases at about a dmax of 2.0. That is, there seems to be little reason to go above log 2.0. Does this possibly relate to the 100:1 range the eye can see at a single time? If I understand my log scales, each number is a factor of 10. So, perhaps 2.0 is where our limit is. My lightly-selenium-toned silver prints, in the real world of darkroom printing, usually have a dmax measuring 1.9. No one ever suggested that the black tones of these prints were not deep enough. While I can certainly see the differences between 2.0 and 2.5 dmax prints when I compare their blacks side-by-side, I'm not sure looking at the image as a whole there is any difference in impact, or appearance of full dynamic range. In the real world when prints are lying side-by-side on my desk, the 1.66 of a high-load, fairly neutral matte black actually looks deeper than the 2.0+ RC "luster" black due to the latter's reflection or "sheen." (Someone suggested that a light satin or matte spray might take this off, perhaps at a small cost of dmax.) My 1.6 dmax matte black prints also usually seem to have a higher dynamic range than my old silver prints. Reflections and lighting conditions of the display seem to determine which of these prints looks best. So, the reflections are part of the formula to get to any optimum dmax, and I don't think our densitometers are measuring this. They work from optimum lighting that does not exist where my prints are usually displayed. In addition to limited benefits from higher dmax and dynamic range, I think there might be a "cost" (lower apparently quality prints) to high dmax that offsets the benefits as some point. I think a higher dmax shows more digital artifacts. I assume that the higher the dynamic range of the print, the more tonal variations the eye can see on close inspection of print details. While the eye can only see a range of 100:1 at any one time, upon close inspection the limiting factor probably changes to how small a density variation can be detected at reasonable lighting levels. The bottom line is that I think I can see more defects in the high dynamic range 2.2 dmax prints than in lower dmax prints. If the initial impact of the print as a whole is not improved by the very high dmax, I'd rather not have it and hide the inevitable defects in my files and systems. There must be an optimum dynamic range that compromises the benefits of increasing the impact of the print as a whole, and the costs of making visible the limits of digital or any other printing process. With some of my older files and systems, this is probably well below 2.0. Paul http://www.PaulRoark.com