cunim
Well-known member
You know, when I look at the cost and high technical caliber of devices such as the one Doug supplies I wonder why someone doesn't make an effort and do a new take on the true archival device. That would be an improved model of drum scanner or scanning table. Clearly there is a market for industrial-level imagers or Doug wouldn't be selling his instrument. How about a linear scanner using structured illumination to increase dynamic range? That could be close to a drum in dynamic range while having the speed and convenience of a flat bed.
After years of comparing spot, linear, and area detectors it always came down to the same conclusions. Area detectors have the advantage of speed and convenience. However, with fine detail there is a progressive loss of MTF with the area and linear scanners. This results from flare. There are also some strange behaviors arrays exhibit with highlights and when one part of the field is much brighter than another, but flare is the killer. It degrades MTF in density details and, more generally, anything beyond about 2.5D in optical density range. From what I could see the degree of loss depends on the proportion of the visual field being illuminated so linear detection is better than area.
Area detectors in devices such as Doug's will always have the edge in volume and, really, what practical alternatives are there at the moment? The biggest problem with scanners is that no one has designed one reflecting today's state of the art in optical engineering. Seems to me that's an opportunity!
After years of comparing spot, linear, and area detectors it always came down to the same conclusions. Area detectors have the advantage of speed and convenience. However, with fine detail there is a progressive loss of MTF with the area and linear scanners. This results from flare. There are also some strange behaviors arrays exhibit with highlights and when one part of the field is much brighter than another, but flare is the killer. It degrades MTF in density details and, more generally, anything beyond about 2.5D in optical density range. From what I could see the degree of loss depends on the proportion of the visual field being illuminated so linear detection is better than area.
Area detectors in devices such as Doug's will always have the edge in volume and, really, what practical alternatives are there at the moment? The biggest problem with scanners is that no one has designed one reflecting today's state of the art in optical engineering. Seems to me that's an opportunity!