I've just sent sets of inkjet prints to Graywolf and to Shel. One pair of prints is a 6x9 portrait of a man on one sheet, and on a second sheet an enlarged detail of the same file that's equivalent to a 14x20-inch print.
This is to illustrate some really interesting properties of digital inkjet prints. Looking at the 6x9 print under a loupe, you can see ink dots and a certain amount of detail. However, unlike with photographic grain, the ink dots don't increase in size at all between the 6x9 print and the 14x20 print. This in turn leads to a truly weird property of digital prints, one that we traditional photographers have a hard time adjusting to--namely that, up to a point, the LARGER the print, the MORE detail it has. That's because in a small print, you only have a certain density of ink dots to describe the detail recorded in the original file. As you ADD area to the print, you have proportionately MORE ink dots to describe the detail that's there in the original. So you can literally see more image detail in a larger print that you can in a smaller one. This is what digital printers mean when they talk about "throwing away detail" with smaller prints. You literally can't describe all the detail in the file on the paper if the print is too small. It's also the reason why a higher megapixel count isn't necessarily a good thing. For instance, if you're already throwing away detail from a 3-mp file when you print, it doesn't do you any good to ADD detail to the file by making it, say, a 5-mp file. Rather, what the higher pixel count will do for you is allow you to make a print with the SAME amount of detail, only larger. A good illustration of this is preparing illustration files for the web. Typically, monitors can only display 72 dpi. So what we do is take larger files and downsize them to 72 dpi. Since you're seeing "real pixels" on the computer monitor, all you do by packing the file with more pixels is to make it larger on the recipient's monitor. So a lot of this talk about "needing" 6 megapixels (I think somebody claimed to need 30 mp!!!) to "equal" the information in a 35mm slide is more or less a classic apples and oranges situation. Because digital prints don't behave like photographic negatives when you try to enlarge them. The "grain" of a digital print is set by the capability of the printer; and once it is set, the "grain" (ink dot) pattern will always be of the same fineness or coarseness, regardless of whether the print is 2 inches across or 20 inches across. That's counterintuitive to traditional photographers. The fact that the digital print will have MORE detail until it reaches the size at which information is not longer being "thrown away" is also counterintuitive to traditional photographers. My buddy Nick, for example, in his black-and-white inkjets from scanned 35mm negatives, settled on a print size of 16 inches wide. It's larger than he's _ever_ made a traditional print. But it's necessary to show all the detail his scanner gets from the negative. Before you decide what number of pixels you want your camera to have, it's a good idea to calculate the size of the prints you want to make. And if you have enough information in the file to make that size of print, you don't necessarily need a camera with a higher pixel count. The other interesting thing about the prints I prepared is that BOTH are made with 3-mp cameras, but one has a CCD that's approximately 5 times the size of the other. The one made with the smaller sensor doesn't yield image quality as good as the one with the larger sensor. Again, it ain't all pixel count that matters. --Mike P.S. If anybody thinks I have any of this wrong I'd love to be corrected. To say I'm still learning is an understatement. - This message is from the Pentax-Discuss Mail List. To unsubscribe, go to http://www.pdml.net and follow the directions. Don't forget to visit the Pentax Users' Gallery at http://pug.komkon.org .
