Steve, et al: I've been following this thread and enjoyed the exchange of information. I have a few comments that, I believe, correspond to a few of things Tyler has been talking about. I’ll keep this as short as possible. On the desktop we use software to control the behavior of hardware. With displays, for example, the machine is calibrated to known standards and a profile is then built that describes the color gamut and gamma of the device. Inkjet printers are linearized and although calibration and linearization perform similar functions, for their respective devices, they aren't the same thing. Displays aren't linearized. The calibration includes a gamma setting. All image devices used in the digital darkroom—cameras, scanners, displays and, most important to this discussion, inkjet printers—include a gamma setting. As do analog film and printing paper. Gamma is built into imaging systems to account for human vision, which is anything but linear. In fact, you can think of gamma as an attempt to objectify our subjective experience of the world (which explains, in part, why we have different gamma settings to choose from in the digital domain). Linearization controls the amount of ink used in each ink channel. The basic idea is to accurately map the image file information, including whatever gamma you've selected, to the printed surface. One of the issues with current OEM print drivers, which cannot be linearized, is that the image file gamma has to be mapped to the driver gamma bias. People like Paul Roark have done an amazing job of working through this issue. With good software, linearization can be tweaked when necessary or desired (it just depends on how far you want to crack open the "black box"). Printer driver software, OEM or RIP, doesn't "care" what color space or profile is attached to an image file. It does care what gamma the image represents. It exists to manage the hardware functions of the machine and translate pixel information into droplets of ink. Years ago, Kodak developed a methodology for mapping image density to code values, called the "Jones Plot". It addresses the issue of how image information is translated from one form into another. It’s useful to us because there are, in fact, significant semantic issues currently at play in the digital darkroom. In time, these issues will resolve themselves (talking about the translation of image information is a little like talking about resolution, where the same terminology is used to describe different things). Long story short (Ha!): if possible, it is best to work with a linearized inkjet printer. If a system is properly color managed you should be able to work with any conventional gamma setting offered through Photoshop's working modes (or build your own, including any variation of the LAB model) and produce prints that appropriately match the image on the display. Bill Kennedy Associate Professor of Photocommunications St. Edward's University 512/448-8680 [Non-text portions of this message have been removed]
Message
Re: [Digital BW] Re: Tonal range and linearization
2004-12-08 by BKPhoto@aol.com
Attachments
- No local attachments were found for this message.