Digital BW, The Print

"jrandall1149" <jrandall@...> writes:

> David:
> 
> Take any of the 21-step wedge test files available and crop to just 
> the steps (eliminate any continious ramp, surrounding background 
> space, and seperator lines in the step area) and then use the 
> histogram function of your image editing program.  You will see the 
> spikes to be equally spaced between 0-255 or 0-100%.  This indicates 
> to me that if you print out the step wedge and then scan it and check 
> the image as above, you should end up with evenly spaced spikes 
> (assuming as you said a good scanner).
> 
> This is the logic I used when I created the RGB partitioned curves 
> for the 1160 (see MIS web site or Files here). During scanning I set 
> the blackest black of step wedge test print to 100% (or 0 value) and 
> the whitest white (=color of paper) to 0% (255)--that is I expanded 
> the maximum range on my paper to the "theoretical" 0-100 range.
>  
> I have also used this approach to tweak Woolf's lumped (non-
> partitioned) single curve work flow.  I haven't tried Jerry Nivin's 
> workflow, but I suspect it is the same or similar.  

I find that when I set up a transfer curve (for use in Jerry Nevin's
workflow) using the "equally spaced spikes" approach, I get results
that look nice, but *not* results that match my screen at all (so the
"nice" only happens through adjust-print-evaluate-repeat loops).  The
curves I get tend to be *below* the diagonal (in the Photoshop
transfer function window).  

> So my suggestion (don't know if it is technically correct or not, but 
> it seems to work for me) is to shoot for equally spaced spikes.

I  like it in a lot of ways; I do definitely have more separable
tones, for example.
-- 
David Dyer-Bennet, dd-b@...  /  Ghugle: the Fannish Ghod of Queries
 John Dyer-Bennet 1915-2002 Memorial Site http://john.dyer-bennet.net
        Book log: http://www.dd-b.net/dd-b/Ouroboros/booknotes/
                 Photos: http://dd-b.lighthunters.net/

--- In DigitalBlackandWhiteThePrint@y..., David Dyer-Bennet <dd-
b@d...> wrote:

<snip>
> 
> I find that when I set up a transfer curve (for use in Jerry Nevin's
> workflow) using the "equally spaced spikes" approach, I get results
> that look nice, but *not* results that match my screen at all (so 
the
> "nice" only happens through adjust-print-evaluate-repeat loops).  
The
> curves I get tend to be *below* the diagonal (in the Photoshop
> transfer function window).  
> 

David:

Seems like you are using a transfer curve to get your printer to 
print what you see on your monitor.  

I think I take a different approach(?).  I start with a fully 
calibrated monitor (black and white points set for the lighting 
conditions of the room and gamma set to 2.2).  I use Adobe RGB 1998 
(a gamma 2.2 color space) as my working space.  I use No Color 
Adjustment in the Epson Color Management portion of the printer 
dialog box.  Once I get the profile to create step wedges with 
equally spaced peaks, the gray scale I see on my monitor matches 
pretty closely what I see on the prints.

Jeff Randall