Digital BW, The Print

Clayton,

Sorry to take so long to get back to your post on this topic. Some
unfortunate incidents last week put me out of work for a few days.

I understand what you're saying when you say no one curve can be
applied to any image to make a black-only print. I also recognize that
each image needs the careful attention of the printer's craft to best
represent the aesthetic that the photographer intends. There is no
one-stop solution for this, and the curve I made is certainly not
going to help the printer who is looking for such a panacea.

What I found when I tried BO printing was that the straight printing
method was giving me very, very different looking images from the FS
prints I had been making. They had a very sharp toe and very long,
drawn out midtones and highlights. If I wanted my print to look like
that, it should look like that on-screen.

I do not have a perfect WYSIWYG workflow for FS printing but it's very
good. BO printing with the traditional "just click print" method was
not giving me what I wanted, especially when I'm trying to compare BO
printing to FS printing.

So, for the purpose of (1) WYSIWYG workflow and (2) swapping between
QTR-FS/Roark-FS/BO and trying to get similar results, I made that
curve. I'm not saying I have achieved both of these goals yet, but I
am convinced that profiling curves are necessary to get there.

Regard,
-Pieris

digital darkroom blog: http://www.pmb.net/darkroom

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Clayton Jones"
<cj@c...> wrote:
> 
> Hello Pieris,
> 
> >Apparently there wasn't one before, or it's not readily available.
> >
> >Because this is my first attempt at profiling and I prefer not to
> >work in isolation, I'd appreciate feedback on this curve. It should
> >provide superior tonal separation in midtones and deep blacks.
> 
> After quite extensive experimenting with BO printing I found that
> canned curves aren't necessary for BO, and in fact are largely
> ineffective in the sense that a curve worked out for one image is
> usually not appropriate for another.  Any fixed curve that is
applied
> universally to every image will likely be a compromise in some way. 
> That's why you don't find any BO curves published anywhere.
> 
> The reason for this is that curves/profiles are needed with ink
> systems that blend multiple inks because they are controlling the
> places where the different inks cut in and out.  Since different
> papers react differently to the inks, a different curve is needed
for
> each ink/paper combination.  
> 
> With BO printing only one ink is used so the problem of partitioning
> doesn't exist.  The basic workflow is to use a common paper for all
> proofing, such as EEM, and work up an image to its best appearance
> (using levels, curves, whatever) on that paper, and then save the
> image. This image will always look its best on EEM as is, and the
> proof then serves as the model for final prints on fine art papers.
> 
> For final prints, depending on the paper used, a contrast curve may
be
> needed to adjust the image to match the proof, because the different
> papers have different contrasts, very much like silver gelatin
papers.
>  This curve will usually be different for each paper, and a curve
for
> an image on Condor BW, for example, will most likely not be
> appropriate for a different image on that paper.  So even these
> paper-specific contrast curves can't be applied to every image. 
These
> curves are usually quick and easy to create, and can be saved either
> as an adjustment layer in the image or in a curves file (I use the
> former method, even though it adds to file size, because it's more
> convenient).
> 
> This is one reason why BO printing is so easy - we don't need canned
> curves and aren't dependent on someone with a densitometer or other
> form of expertise to provide them.  There's more to good BO printing
> than just the curves, of course.  A more detailed explanation is
found
> in a series of articles at the link below.
> 
> Regards,
> Clayton
> 
> 
> Info on black and white digital printing at    
> http://www.cjcom.net/digiprnarts.htm