[font color=\'#000000\']When researching a scanner purchase the tendency is to concentrate on detail resolution, which is certainly important. But I've found that resolution is only one of four factors that are equally important. The other three are dynamic range, scanner crud, and ease-of-use.
Dynamic range. Sergio rightly flagged this as important in his initial post. As we all know the Dmax figures advertised for each scanner are nothing more than theoretical maximums for the bit depth of the Analog-to-digital converter component used in that scanner. The only valid way to know the real DR of a scanner is to see the resulting image file when a high contrast film frame is scanned. Even then, the results are totally dependent on the operator's knowledge of how to get optimal results from a given scaner. I feel the reviews on imaging-resource.com are invaluable because they publish such results. Unfortunately, not for flatbeds! These leaves you scouring the net for usually amateur reviews or trying to test yourself in your local retail store.
Scanner crud. This is my own colourful term for a colourful problem. Thanks to the wonders of grain-aliasing, pepper grain, and digital "noise", scanned image files are far from as clean and pure as are decent digital camera image files. Some manufacturers deliberately introduce a bit of defocus to lessen this stuff. Again, it's hard to evaluate on-line test images for this unless results from a variety of scaners against the same test film can be compared. However, one can gain a fair idea of a given scanner's crudiness by averaging a number of anecdotal user reports.
Ease-of-use. Here I'm thinking especially of: quality of the provided film holders, infra-red cleaning, and quality of the provided software interface. These things may seem relatively minor, but a severe weakness anywhere in the chain between developed film and Photoshop, can render a scanner essentially useless.
Paper shims. The film you are scanning needs to be at a certain precise distance from the CCD imager in order to achieve maximum resolution of detail. Owners have reported on-line that due to variations from one 2450 to the next one can achieve sharper scans by changing the height of the film above the glass. According to this theory, the particular 2450 you purchase may already be pre-set so laying your film directly on the glass gives best results, if so you need do nothing. If not, you need to prop the film up off the glass by a certain amount (determined by experiment on your part).
It may also be that this was a manufacturing problem that has since been fixed.
2450 links:
Kerry Owen's reviewNorman Koren's review[/font]
