[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

sectioned scans redux

Dear List,

Thanks to all for the helpful suggestions for dealing with scanning of
large alternate process prints.

The summary answer is--new scanners are better than old ones.

More specifically, a cheap (200 bucks) modern scanner eats the shorts
off a ten year old scanner that cost nearly ten times that much back then.

The cheap scanner doesn't quite deliver 'perfect' repeatability.
Sometimes making the seam in a pure tone area like a sky there is
detectable difference (usually correctable by using a selection-area
curves adjustment) but it's quite minor and basically acceptable for the
task at hand, which is digitizing large prints for the web. I can only
guess at the capabilities of _expensive_ current tech scanners, but it
must be incredible.

The big thing is that the scanner, with very little manual assist,
delivers incredibly convincing RGB rendition of the actual color values
of the various platinum-print variations. This is remarkable, because
rendition of monochrome color is awfully difficult in any reproduction medium.

The main limitation of the cheap scanner is its size. The UMax 2200 I
got can only take 8.5 by 11.7 inch bites. In practice, this means that a
7x17 is an easy split with two scans. 12x20's take 4 bites. I've done
two of these so far, and find that the whole operation can be done in
about twenty five minutes. That's two minutes to scan the four sections,
and twenty three to make the joins. I wind up with over a dozen files by
the time the joining is over (never work on a scan, always on a
duplicate), but the results are quite presentable.

Anyone interested in seeing some humble first attempts with the 7x17's
is welcome to have a look at "Gallery Two, The Flat Land Series" on my
website, link below. A gallery of 12x20's from the South Dakota Badlands
will probably be up tomorrow.

---Carl

-- 
Workshop info and an online gallery, at:
http://home.earthlink.net/~cweese/