Digital BW, The Print

Hi Roy,

Thanks for the reply.

THe file that I get from a scanned BW negative have RGB values that 
don't usually vary by more than about 10, and usually much less. 
When I do the two hue/sat adjustment layers trick to desaturate etc, 
the values are pretty well R=G=B.

I will try the conversion and see what I get. If that doesn't work, 
I'll print the usual target on that ink combi and try to work it 
that way.

Many thanks,

Ellie

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Roy 
Harrington" <roy@h...> wrote:
> Hi Ellie,
> 
> I'm wondering if the RGB file you send to the printer is basically 
gray
> where R=G=B or at least close.   Or do the 3 RGB channels 
separately
> control different inks to significantly control the density.
> 
> The main "new" thing was to use the RGB profiling that i1Match
> already does to a Grayscale.  The reference file is actually still 
an
> RGB definition but only values with R=G=B are used.  When the idea
> first came up,  we actually used the ordinary RGB targets but I 
think
> that had two difficulties.  First was how to drive a real 
grayscale print
> method with an RGB target file.  Converting to Lab and keeping L
> worked pretty well but its not exactly the same thing.  The other
> problem was that the Eye-One reader scans patches that must vary
> enough for the read to recognize transitions.  Reducing the target 
and
> reference to just 21 gray values solved both issues.
> 
> So if your RGB files are essentially gray you should be able to 
just
> convert the targets to RGB and print, and do the same proofing that
> is described.
> 
> However, if your RGB files somehow encode different inks I would
> think you might have to go back to the regular RGB references and
> target -- just use i1Match as normal.  The difficulty will 
probably be
> reading the patches, although you can do it without scanning.
> The other issue is how well the i1Match software can deal with
> colors other than real RGB.
> 
> I suspect you'll need to do various experiments and see what works
> and how well.
> 
> Roy
>