Digital BW, The Print

Hi Peter,

I've actually spent quite a bit of time on Lindbloom's website but I've
never been in contact with him.  Anyway, it sure is chock full of information.
There's quite a few formulas there and other places I've investigated.
If you have the formulas it's not difficult to calculate most conversions,
but deciding what you want is the hard part.

For QTR linearization the idea is to make correction curves such that 
linearly spaced grayscale values result in linearly spaced L values on
the paper -- producing linearly spaced perception by our eyes.
Incidentally, I've seen the aims values that IJC/OPM uses for their
linearization and they exactly coincide with QTR.  The stepwedges
I get seem reasonably even by my eyes.

On the other hand when I calibrate my screen and look at a stepwedge
they don't seem nearly as even on the screen.  So I can do a lot better
by making a Proof Setup that brings the screen vs print much closer.
Part of the issue is that the screen has a much larger contrast range
than the print can possibly have.  

I'm not very familiar with internals of ICC profiles, but it is interesting stuff.

Roy

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, Peter Miles <P.Miles@m...> 
wrote:
>  "Roy Harrington" <roy@h...> wrote:
> 
> 
> Hi Roy
> 
> I don't know my self about this either. However I think a person who would
> know would be Bruce Lindbloom at <brucelindbloom.com>. He is very
> approachable.
> 
> You may already know but Bruce has a comapanding Calculator on his web site
> for converting between L*, Y, Density & Gamma values. Could be that
> photoshop uses one of these other functions for monitor display?
> 
> The calculator is in the "Calc" section of his website. There is also a very
> interesting blurb under "more info" within his companding calculator.
> 
> 
> After I Have liniarised QTR I create an ICC profile using ColorSynergy
> software. This solves for me the screen mismatch problem.
> The software effectively makes its Tone reproduction curve. Looking at the
> Tone Reproduction curves in the ICC profile they all look pretty gentle and
> very similar to each other in magnitude for the inkset I'm working with.
> 
> Hope this helps.
> 
> Kind Regards
> Peter Miles
> 
> 
> 
> 
> > Message: 15
> >  Date: Wed, 12 Nov 2003 20:45:13 -0000
> >  From: "Roy Harrington" <roy@h...>
> > Subject: Re: QTR printing lighter than on screen...
> > 
> > --- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Steve Kale"
> > <stevekale@b...> 
> > wrote:
> >> I recently calibrated my monitor and Epson 2100 with Gretag Macbeth's Eye-
One
> >> Photo system.  When I use QTR with Carl's curves my prints are about 1/2 a
> >> stop or so 
> >> lighter than on screen.  Is this just the difference in printer production
> >> tolerances (ie I
> >> should really construct my own QTR curves for my own printer as I have done
> >> for 
> >> colour work) or something else?  I am assuming that in a well calibrated
> >> environment I should be getting exactly what I see on screen.  Is the QTR/B&W
> >> world 
> >> the same as colour in this regard?
> >> 
> >> Cheers
> >> 
> >> Steve
> > 
> > Steve,
> > 
> > I've also experienced the same thing.  I don't know the reason -- I'd
> > also think if everything is calibrated it all ought to match.  Fortunately
> > its quite easy to create a Proof Setup so you can compare the screen
> > and the print.  Tyler Boley has a nice write up for doing this in the Files
> > section of this group.  Check out:
> > 
> > Image Processing>Matching Your Monitor view to Your Prints.pdf
> > 
> > This is well worth the effort.
> > 
> > Roy