Digital BW, The Print

Hi Paul


> From: Paul Roark <paul.roark@...>

 

> I don't think I can.  How many stops does your monitor deliver?  That is,
> what is the 100% black and 0% white reading range?
> 


I am not sure that Eye One will disclose this to me (or at least I don't
know how to find the data) - it is of course measured.  What I can say is
that I calibrate my screen for a 5000 Kelvin white point.  Actually, I would
like some input on this choice.  There is a discussion of white point on x
rite's website here:

http://www.xrite.com/support_doc.aspx?Line=32&SupportID=3005

I chose D50 because I understand that this is the standard illumination when
reading a printed printer calibration chart and because of a discussion on
this forum that D50 is a better representation of daylight at sea level than
D65.  In any event, I know the ambient light in my work room is a lot lower
than even D50 and I have not stumped up for a lightbox (and so I know there
is a mismatch between the look of a print in my workroom vs display but I am
prepared to live with this if D50 is a representative average of the viewing
conditions my prints are likely to be in.  Back to your question....

What I can also tell from my Eye One is that my display luminance is set at
a target 120cd/m2 (an actual of 120.6cd/m2) and the minimum luminance of my
display is 0.4cd/m2.  So presumably from those two points one can calculate
the dynamic range, 120.2cd/m2, and convert it to stops.  I need help in
doing this however.

> For grayscale, gray gamma 2.2 seems widely accepted and displays my GS files
> the same in that mode as when they are converted to Adobe RGB for printing.

My understanding is that Gray Gamma 2.2 is a subset of Adobe RGB.  The more
important issue is that you, by the nature of your workflow, define the
mapping from the workspace to the print space - you in effect hard code this
when designing your curves.  Therefore you dictate to those using your
curves the workspace they should use.  This is of course fine but needs to
be made clear, particularly given all the press on this forum re QTR Grey
Lab.  If I work in QTR Grey Lab and then try to print with your curves I
will not get the outcome you intended (my 95% step wedge is L=5 whereas for
you it is, I believe, L=2).  (With QTR it doesn't matter because we now have
a more "dynamic" mechanism for going from the workspace to the print space:
perceptual intent rendering.  I can work in Gray Gamma 2.2 or Gray Lab it
doesn't matter.  Roy will no doubt correct me if I am wrong on this.)





As you think about all this bear in mind that you are a lot further down the
track than you may think.  Before Roy's Matte/Photo Paper ICC profiles QTR
suffered from two issues.  Firstly, as I described in an earlier post, the
print space, while "linear", was such that when combined with the one-to-one
mapping that was being done produced flat prints out of the box.  We all had
to dabble with "s curves" to get a more attractive use of the narrower print
space.  Secondly, we had to deal with the soft proofing issue that you are
dealing with now.  (Actually Carl's use of Eye one to generate an ICC
profile soft proof overcame this second issue before Roy came up with his
paper profiles and colour managed workflow.) You have hard coded by design
the mapping from Gray Gamma 2.2 to the print space for each of your
paper/ink/printer "curves" such that it generates "pleasing" results.  You
have dealt with the first issue above.  Now you just need to be able to
provide yourself (and other people not in a position to do it themselves)
with a soft proof.  If you had an Eye One this could be done very easily
with Carl's technique.  I would think it is worthwhile seeing if that
technique can be replicated with your existing measurement device.  Carl
essentially swapped out one of the selectable profile charts in Eye One to
one suited for B&W (leaving the filename the same).  Print the chart and if
you can then select it so that Eye One is expecting the right data then you
can generate an ICC profile for soft proofing from it.  It's a fair bet that
you could do the same with your existing device/software.

Black Only printing, on the other hand, is still grappling with both of
these issues.  Firstly, I believe Clayton still recommends setting gamma to
1.8 rather than 2.2 which means the print space is even further removed from
a Gray Gamma 2.2 workspace (the mismatch is made all the more worse).  In
addition, the print space is not the best it could be and Same as Source or
one-to-one mapping is still used to go from Gray Gamma 2.2 to the print
space.  Again, see the input/output relationships that I measured and have
made available here:

http://homepage.mac.com/stevekale/stevekale2/FileSharing37.html

An immediate refinement of BO printing (in addition to selecting 2.2 rather
than 1.8 as the gamma in the Epson driver) would be to profile the BO 2.2
workspace using Carl's soft proof technique and distribute this to those who
use BO but don't have access to an Eye One so that they can soft proof there
images with reasonable accuracy (emphasis on reasonable).  Then their s
curves can be done with a preview in view.  Better yet, if possible, the
technique/skill that was used by Roy to generate his matte/photo paper
profiles could be employed so that Black Only printing became a colour
managed (rather than Same as Source) workflow.  Then perceptual rendering
could be utilised to curve the image into the BO print space.  One wouldn't
have to fiddle with s curves again - just like QTR today.

Hope this makes sense/helps.

Steve

PS:  If I get a chance I will try to provide a BO1.8 and BO2.2 soft proof
profile for Eboni on EEM on a 2100 using Carl's technique.