Digital BW, The Print

All I can say is that this list has constantly witnessed the frustration of
people using a B&W workflow which is not colour managed (because an
effective and colour managed workflow does not exist short of moving to a
VERY expensive RIP like Imageprint) and finding that their prints don't
match the image on the screen.  Well surprise surprise.  As you quite
rightly note a display is capable of producing a broader colour gamut than
today's inks on paper and some tonal compression must occur.  (And this
would be also be evident in a colour managed B&W workflow - if only because
ink and paper can't produce the dMax of the display.)  Until Carl (and
previously Tyler) came up with an easy method for soft-proofing, the answer
was trial and error. Now at least, for any selected curve/ink/paper/printer
combination one can get an accurate soft-proof.  It's not colur managed per
se but it achieves the same goal - readily and easily for anyone who has the
equipment to calibrate their monitor and printer.


> From: "David B. Brooks" <fotografx@mindspring.com>
> 
> The concept and function of Œsoft proofing¹ was to be able to emulate what a
> publication document would look like printed on an offset press before the
> fact on a computer screen. It really would have no purpose if a black and
> white digital file could be managed  for direct output in the same way color
> is managed. You would with a fully calibrated and profiled system
> automatically achieve screen to print matching through a Photoshop workflow.
> 

I don't believe this is true.  A soft proof would be required because the
(colour) gamut of the printer/ink combination is significantly less than
that of the display given today's inks and papers.  You would only get a
print matching your monitor if you had Proof Colours checked with the
correct ICC profile.  In colour prints this is less noticeable and we
typically have a better printer colour gamut in relation to the image at
hand. 

But it also leads to the discussion I began with Roy awhile back as to how
B&W "gamut compression" should be handled.  The classic complaint is that
(non soft-proofed) images print lighter than on screen.  A result of the
lower dMax of the ink/paper and a shift of the tonal composition of the
image by the RIP.  I am no expert in this area (and may be talking through a
hole in my head) but it seems to me that linearisation provides a distorted
smoothed compression into the tighter gamut (higher dMin, lower dMax) and
every pixel in an image gets shifted - except maybe the midpoint.  I often
wonder whether a clipping approach would be better (or some balance in
between).  I question whether raising the brightness of the image overall
(mid point may stay unchanged but the others are affected) to achieve "good
separation in the shadows" is the right approach (presumable the opposite is
occurring in the highlights). I find that when I soft proof an image for QTR
output I end up applying an S curve to try to get back my overall tonal
balance - accentuating the near blacks and near whites.  I may be wrong but
I suspect this is really undoing part of what was done at the linearisation
stage.

Regards

Steve