Digital BW, The Print

With the piezo inks, I have a dot-gain curve set up to approximate the way
the inks will look on paper (this is taken from the piezo manual).

How do you go about proofing images which are to be printed using the Roark
curves etc?
-- 
John Brownlow

http://www.pinkheadedbug.com

ICQ: 109343205

on 8/21/01 9:21 AM, Johnny Deadman wrote:

> With the piezo inks, I have a dot-gain curve set up to approximate the way
> the inks will look on paper (this is taken from the piezo manual).
> 
> How do you go about proofing images which are to be printed using the Roark
> curves etc?

Good question John, and one I don't have an answer for yet. Unfortunately
sometimes I'd rather just stumble my way through a process rather than put
the time into testing and refinement.

First off let me state I'm still using out my Piezo/generations blend, which
Paul's first VT process was based upon. MIS users may feel otherwise.

I do find that with Paul's method I get a print that looks like it was given
a reverse S-curve relative to my screen image, i.e., darker highlights and
lighter shadows, and probably a tad deeper in the midtones overall.

I've come to just add a compensatory adjustment layer along with Paul's
curves at print time.

But on this list last week Dan Culbertson was passing along some wisdom
which I have not been able to test thoroughly. One possibility was to create
some custom gamma spaces which could be applied in the print driver at print
time. The other is to create a print profile through which one can preview
the image as they tweak it. I like this profile preview option best, 'cause
it will indicate your hue effects (your tone choice), and would be very
useful in aiding your separation curve scripting, like when you need to
write a curve set (as in profile) for a new paper.

My results were mixed, but then I used Profiler RGB with my Umax Astra
scanner, and I've never gotten good results with that combo for color work
either. I found the profile useless when used with Paul's existing curves,
but extremely helpful to allow me to script a decent set of curves on my
first try. But, my curve set was not as good as Paul's and that's where I've
left off at the moment, back to just slapping adjustment curves on, and
printing as I always have, by making successive iterations based upon the
result of my last print. I think there will always be some of that, as
reality begins where ink hits the paper.

However, now that the community of MIS VT users is growing, and I will be
joining those ranks shortly, I look forward to us refining and expanding
this process as a group, and sharing curves and profiles amongst each other.
Such is the beauty of an open system.

Todd

--- In DigitalBlackandWhiteThePrint@y..., Todd Flashner <tflash@e...> wrote:
snip
> But on this list last week Dan Culbertson was passing along some wisdom
> which I have not been able to test thoroughly. One possibility was to create
> some custom gamma spaces which could be applied in the print driver at print
> time. The other is to create a print profile through which one can preview
> the image as they tweak it.
snip
> My results were mixed, but then I used Profiler RGB with my Umax Astra
> scanner, and I've never gotten good results with that combo for color work
> either.

I think the scanner is the problem. I just made an RGB preview profile with Dan's method using Profiler Pro and a 
Spectrocam to double check the method in preparation for doing RGB quad curves for a friend. The profile worked very well 
and allowed me to create curves in about 10 minutes that look like they will need only minor tweaking.
Keep in mind that the profile is useless to print through, and is used to create curves with, or preview a quad file for 
tweaking.
Tyler