Digital BW, The Print

Clayton-- I've mulled it over and thank you for the detailed response.

You did not say if your monitor is hardware calibrated and to what 
settings.

I am always using "No COlor Mangement/Same as Source."

What you say makes perfect sense and I've tried most of these 
approaches.  The dot gain 20% is, incidentally, what Epson recommends 
for the workspace/profile with the ABW with the "darker" setting in 
the driver.  The overall density matches pretty well (screen to 
print) with this setting, but the lower end of the tonal scale is 
quite different and the driver is compressing the blacks.  I believe 
that Epson is inclined to this in their drivers because of all the 
sensitivity about D-max and the desire for prints "with punch."

I went to the "lighter" setting in the Epson driver because it opens 
up the shadow areas very nicely (as compared to any of the other 
settings) and the correspondence (in terms of *spacing* or 
linearization) to a step wedge (screen to print) is quite good.  This 
however leaves the screen image darker than the print in tems of 
overall density; and the coorespondence in shadow separation is not 
exact.  A dot gain workspace would, of course, only agggravate this 
problem.

So I end up, like you, using a curve at the top of the file.  I am 
simply using mine to match the screen view to the print (and I turn 
it off for printing) and I then use a curve to get what I want on the 
screen (which I do not turn off to print).  You are using your curve 
to match the print to the screen view (or to previous prints on the 
smaller printer or BO).  What I'd like is a system that did not 
require this "special" curves layer for each file (the screen 
matching one in my case), and I have no need to match to a different 
printer or print.

I suppose this means profiling the ABW mode and using that profile 
for Soft Proof.  (Or conversely, creating a profile for printing that 
matches the print to the screen.) What would be necessary for this 
profiling?  Eye-One Photo?  Or just the i1 photospectrometer and the 
software mentioned earlier to read desnity rather than LAB?

Thanks again.

Walt





--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Clayton Jones" 
<cj@c...> wrote:
> Martin and Walt,
> 
> >>>>>
> >I have found that, if I set my color setting (PS CS) working 
> >spaces "grey," to Gamma 2.2 and my ABW prints, either warm or 
> >neutral to a fine adjustment of "dark," I get a pretty good 
> >match to my screen.   I do not know if this is unversal, but 
> >it does work for me.
> >>>>>
> 
> >>>>>
> I am also using the ABW driver with "light" (smoother step wedge)
> and "neutral" settings. I have not reduced the ink limit. Do you
> have any correspondence between screen and print (i.e. something 
like
> a soft proof) and if so, what are you using for screen view
> (workspace, monitor calibration, etc.)? I am using a workspace of
> gamma 2.2 and a monitor calibrated to 5K and 2.2. The screen view is
> much darker than the print and the shadow compression is somewhat
> different between the two; so I use a PS curve for viewing to fairly
> closely match the print. But I thought you might be doing something
> better than this.
> >>>>>
> 
> I'll try to explain what I'm doing, but first please understand that
> these various settings push and pull an image at both ends, so there
> are various combinations that can end up with virtually the same
> results.  This explanation is in no way suggesting that it is 
superior
> or that I'm suggesting others should do the same.   It's just what
> works well for me.
> 
> This method originates with the BO technique, which uses "Same As
> Source" as the printer profile setting (I use PS CS, I think in CS2
> this is called "No Color Management").  This means that the front 
end
> (image) profile does not affect the print, it only affects how the
> image looks on screen (for a full explanation of how this works 
please
> see my article #4 at the link below).
> 
> Since the Epson BO driver puts out a fairly light print, the image
> itself has to be a bit darker to get the proper density.  Because 
the
> image is darker it looks darker on screen, so in order to get good
> WYSIWYG (make the screen image match the print), I use a front end
> profile of Dot Gain 18 or 20%, which makes the image look
> lighter on screen than GG2.2.  
> 
> This has proven to be a very reliable and consistent approach.  I 
have
> set the PS default gray space to use a custom DG18 curve, so every 
new
> grayscale image gets this to start with.  Depending on the image, I
> may change this to match the first test print, anywhere from DG15 to
> DG20.  This gives me excellent WYSIWYG all through the entire 
process.
> 
> I use EEM for proofing.  When it's time for a final print on some
> other paper, I usually have to tack on an adjustment curve to tweak
> the contrast and/or density to match the proof.  I save this curve
> with the image, usually naming it with the paper name.
> 
> Suppose now I open an existing image to print to the 2400.  Since 
the
> image is darker, I've found that setting ABW/Tone to Lighter and
> reducing the ink limit by 5% produces a print of nearly identical
> density to the BO version (which probably means I'm using less ink 
as
> well).  All that's required is an adjustment curve to tweak the
> contrast to match the proof, depending on what paper and ABW color
> settings are used - essentially the same workflow as I've been using
> all along.
> 
> Having this consistency is paying off, because I can work up an 
image
> using BO on the R200 (which saves a lot of expensive K3 ink) using 
my
> tried and true methods.  If I decide the final image will look 
better
> as a K3 print, I can then send it to the 2400 using those saved 
driver
> settings and be right in the ball park first time, only needing 
minor
> tweaks on the paper curve .
> 
> This was put to the test last week when I made a formal portrait of
> some friends at church.  I worked up the proof in BO and then sent 
it
> to the 2400.  No base adjustments were needed to the image, just the
> usual curve tacked on to make the final image match the proof.
> 
> I was real worried that my entire workflow would have to change with
> this printer, so needless to say I was VERY happy to find I can
> continue to work this way and don't have to change any of my 
existing
> images.
> 
> If you have been using "Same As Source" for the back end and GG2.2 
for
> the front end, then your images are probably lighter and setting 
Tone
> to Lighter and/or Ink Limit to -5% may not be a good thing.  But 
there
> are many possible combinations of these things, so you just have to
> find what works best for you.  But at least maybe this will shed 
some
> light on it.
> 
> 
> Regards,
> Clayton
> 
> 
> Info on black and white digital printing at    
> http://www.cjcom.net/digiprnarts.htm