"Michael" <michael3442@...> wrote:
>
> I'm trying to work out the kinks in my workflow ...
> When I finish editing an image and move on to soft proof,
> the image shifts ...
Ideally, you'd edit your image in the working space that would match
the print so that there would be no shift.
On the other hand, since I use more than one printer workflow I now
edit in gray working space Gray Gamma 2.2. (Set in Photoshop at
Edit>Color Settings>Workins Spaces>Gray.) I wanted to use the most
standardized space. For color, Adobe RGB (1998) appears to be the most
widely used standard, and Gray Gamma 2.2 appears to be a subset of
this.
If a printing workflow is "color managed" such that the printer
software reads the image's embedded profile or workspace, then on a
calibrated system, the print should match the monitor. I think the use
of QTR's Create ICC to accomplish this is about the best and easiest
B&W workflow we have now because is solves this long-standing problem
of the print not matching the monitor, at least with respect to the
relative densities. Any Epson driver based B&W workflow should be able
to use these ICCs. For example, if one uses the Epson "Advanced B&W"
workflow, these ICCs can be made for the specific ABW settings you use
and then inserted in the PS Print preview when printing.
Most of our B&W workflows are not color managed. So, there are a
variety of techniques poeple use to get a better match. My particular
workflows either use QTR's Create ICC or QTR Rip. The Rip has what I
call a straight-line characteristic curve, as opposed to the compressed
shadows of Gray Gamma 2.2. To correct for this, I have a Photoshop
curve (often just put on a layer) that does the compensation/conversion
(GG22-to-QTR.acv).
(See http://www.paulroark.com/BW-Info/GG22-to-QTR.zip)
By using this curve I don't have to re-edit the underlying file. If
the file is 16 bit, there should be plenty of room to do this without
the loss of any significant information.
This is just one of any number of solutions to this problem.
Hope this helps.
Paul
www.PaulRoark.com