I am a myself new in the field of B&W digital printing and it took me about
a month, which I can afford being retired, to get accustomed to the various
aspects of not getting the print you are expecting. Color Management,
profiling, curves and proofing, everything was new to me. So I did a lot of
reading and trying, but to be able to print every subtle change you are
planning was not possible at first. And, to my opinion, printing B&W is much
more delicate than color printing. It can make or ruin the picture. If you
want the look of old fashioned barite, you will have to master everything
that is involved in printing.
Of course you can get a lot of advice from Paul Roark and others, but I
found it several times not completely correct. The results depended too much
on a lot of settings in Photoshop for instance. Tiny tweaks that gave an
enormous difference. So I developed my own system that, of course, is much
better.
As follows;
I use Epson R220 en MIS inks in the combination warm tone/cool tone. That
works marvelous.
The paper is INNOVA ifa-14 (matte) but for the future that will be ifa-09
(white gloss).
That was the easy part.
How to profile the paper? I noticed that the downloaded profile was not
correct (I suppose that every printer needs its own profile). Making a 21
grey-step was easy and the print was scanned with my scanner (Epson V700).
Maybe not needed for this stage, but keep working in gray 8-bit and
gamma=2,2.
My scanner does not give L-values, but with the use of a graph made with the
values for RGB and L which you see when you make the grey-step, you can read
the L-values from the graph when you scan the printed 21-step.
Make the txt-file needed for QTR-Create-ICC-RGB.exe and do not forget that
you have to use Tab's. Otherwise it won't work. And do not use the value 0
(zero), but use 0,1. And there is always a positive increment needed. Twice
the same value for L and it won't work.
With the resulting profile the grey-step printed nicely (printed from
Photoshop with preview,color management and profile QTR-INNOVA etc.)
But then the really hard stuff. How to get on the screen what is coming out
of the printer!
I tried curves, I tried custom dot-gain, I tried adjustment layers and I
tried adjusting the monitor, but I did not get a consistent result. And I
kept forgetting about no colormanagement, assigning profile, with or without
proof (you see, I am not a trained expert), RGB or Gray-mode and so on.
I gave up and found something better.
Deciding on the settings for your screen is completely visual. You can never
get a complete likeness between a picture on screen and what is on the
print. It depends on your position for the screen, the balance between
highlights and shadows in the picture. It depends on the light in the room
and, I noticed, it depends on your being tired, your mood. If you get it
right (or almost) for one picture, it will be wrong for another.
I use a grey-step (AdobeRGB, Gray gamma=2,2) with 21 steps + the steps in
between. Arranged so that the steps in-between are placed not in between but
besides. You print that special grey step using the profile for the paper
you are using. Than relax and sit back a bit. In Photoshop you have that
grey-step on screen. No colormanagement and proof is off. Than you compare
the grey's on the print with the grey's on the screen and you decide on a
L-value (they are printed on the steps). Do so for all the 21 steps on the
print but do not follow a sequence. That can be misleading! No, it certainly
is! Every decision for a L-value has to be a new start. So start with step
4, than step 16, than 8, than 20 etc. Repeat it several times and make a
graph for step's 1 to 21 against L. You can see how much difference there is
between each 'measuring' of L. But you can draw a nice graph and use those
values for making, as above, of a profile (Let's say screen-INNOVA.icc).
Place it in the right folder.
Now you can use that screen profile as proof (in Photoshop: view / custom
profile etc.)
Open the grey-step in Photoshop (no colormanagement and proof = screen
profile) and compare your screen with the print. It starts to get close. How
close depends on your ability to guess the L-value (above), but if it
differs it can be changed quite easily. Make a new profile for which you
change the L-value for the step that is wrong. L a bit higher and it becomes
a bit brighter. I succeeded so it must not be too difficult.
Now you have a good screen profile and a working paper profile so the prints
must be good? Wrong. As abovementioned you can not get a good setting for
every picture. You will have to adapt it just before the final printing. But
there is the graphic card too. I suppose that every graphic card has the
possibility to adjust color settings. Mine has. Brightness, contrast and
gamma. I keep the program for adjusting the monitor open. Than I make a
print of the picture I want to work on. Adjust the settings for brightness,
contrast and gamma (minor adjustments) and know now that what is on the
screen will be on the print.
But I have noticed that you need to repeat that just before the final print.
I am the instability in the system who is changing all the time.
This is the way I am working now and it works fine. I am completely in
control of every minor adjustment needed to get a good print. Mainly
landscape in cloudy, misty or rainy atmosphere, and they do need a lot of
work to get it lively.
In include my grey-step's for those who want to have a look at it.
Greetings,
Anne Zwaan
[Non-text portions of this message have been removed]