On this web page
<http://www.robgalbraith.com/bins/content_page.asp?cid=7-7889-8132>
there's an article on Bill Atkinson's new profiles for the widest of
Epson's new printers. But this quote indicates his work might also
benefit at least some of the rest of us too, at least when printing
in color. I don't understand it well enough, however, to know whether
this is relevant at all to B&W printing:
>[snip]
>
>The eight different gamut mappings come courtesy of the options
>available in the five programs he used to build the profiles:
>ColorSavvy ProfilePrinterDeluxe v5.3, ColorVision ProfilerPro,
>Fujifilm ColourKit 4.3.1b1, GretagMacbeth ProfileMaker Pro 5.0.5
>and X-Rite MonacoProfiler 4.7.2. Though not all of the makers of
>these packages permit the resulting profiles to be freely
>distributed normally, Atkinson secured the necessary permissions.
>
>All of the targets were printed on an Epson 9800 that was calibrated
>with Epson ColorBase first. As such, they should work equally well
>with a calibrated 7800. Plus, Atkinson noted in an email exchange
>that the profiles may also be applicable to a calibrated 4800, but
>not the R2400 (or older printers that don't use the Ultrachrome K3
>inkset). His comments on the level of compatibility with these
>printer models are based on recent feedback from Epson, rather than
>his own testing, however. In making three prints today
>on our ColorBase-calibrated 4800, the results with Atkinson's
>ProfileMaker 5.0.5-generated profiles were a close match to ones we
>had generated with the same software earlier this week. This
>suggests that 4800 users, in addition to those running a 7800 or
>9800, may find it worthwhile to check out Atkinson's new profile
>offerings.
>
>In fact, even if you don't own any of these printers, but you're
>considering the purchasing of one of these profiling packages,
>Atkinson's profile collections provide useful information about the
>quality of the profiles each package generates and whether one
>application's handling of the perceptual rendering intent is
>better-suited to the types of photos you print. You can of course
>compare the relative colorimetric intents in the packages too.
>[snip]
--
Sam
[Non-text portions of this message have been removed]