Re: Scanner and Printer Profile Editing for Fine Art Repro
Re: Scanner and Printer Profile Editing for Fine Art Repro
- Subject: Re: Scanner and Printer Profile Editing for Fine Art Repro
- From: Roger Breton <email@hidden>
- Date: Thu, 11 Nov 2004 12:52:14 -0500
Steve,
Let me start by saying I always print the synthetic Macbeth file from Bruce
in RelCol but with AbsCol. Since I'm looking for an absolute match of the
patches and not a relative match. Once I'm convinced that my absolute match
is OK then I move to test the RelCol transform.
> In PS, when I print the file I am using
> relative colormetric with BP checked,
So, in BEST, you don't do any conversions, only in Photoshop?
> reason being is this is how I normally
> print 98% of all my images
I understand.
> and never use absolute colormetric because there
> is to much clipping.
Right.
> I also check my printer profile using ALWAN Color
> Pursuit, which I have been getting consistent profile quality of 9 out of
> 10.
I should take a look at that utility one day...
> So I take my print, and evaluate it under my gti light box next to the
> color checker chart.
OK, side by side under the same light.
> Right off the bat, I notice some density issues and
> some gray balance issues.
The grays don't look as neutral?
> To verify this I break out my trusty i1 and using
> the measure tool, first look at the white square, the forth gray square and
> the black square.
OK.
> For density purposes I am just looking at the "L" value.
OK.
> Since I am printing on canvas, the white point of the canvas is around L94,
> after reading my print I am getting L91,
Normal. The white patch on the Macbeth is not 100. So you can't expect to be
mapped to L94!
> so I can see why my white patch is
> not white.
It ain't 'white' on the Macbeth in the first place.
> Then looking at the forth patch gray, The L should be 51 but mine
> is 55
That's because of RelCol.
> and on the black square I am getting L 28, but it should be around L
> 20.
20 in abs col. Although 28 is a little high.
> So this leads me to my first profile editing question.
> Should I try to correct this in the printer profile using either Profile
> Maker 5 editor, Monaco Profiler 4.6.1, Fuji ColorKit editor 4.0.7. The
> reason I mention all of those is that is what I have to use at my shop, and
> a side bar note.
Wow! You have the "embarrassment of the choice", we say in french.
> Is there any good documentation on any of these apps when
> it comes to editing a profile? I sure have never seen any.
Nada. Ain't that a shame?
> Being an ex drum scanner operator from the good old days, I was always did
> corrections in the following order, Hi light, Shadow, gray, then finally
> selective color. I notice all the profile editing programs seem to be set up
> the same way.
OK. Makes sense.
> Remember at this point all I am trying to accomplish is first getting the
> bottom row of the color checker chart to match my output. Seems simple
> enough.
Don't forget contributions of optical brightners.
> NOT. Am I asking to much of CM?
You raise big questions.
> Shouldn't I be able to at least be
> able to shoot a chart, weather it's gretag's, Fuji, Kodak, or Hutch and be
> able to get a good visual match?
The moment you say "shoot" you're bringing the camera into play. First,
neutralize the color once shot. And then we can talk. Also, you need to
neutralize the color of the printer. Then we talk some more.
> Seems reasonable. I realize that CM in my
> application is not the silver bullet, and there will always be a need for
> selective color correction in PS to get a good match. But how much?
Hmmh.
> Where I am at is basically I have had some limited success in not screwing
> with the printer profile, but instead making these adjustment in my scanner
> profile which I use on every piece of art I shoot.
I practice what I call "modular debugging", get each profile to perform well
independent of one another. Then I know what to expect once I start to
transform images in my chain.
> In Bruce's book real world color management, it mentions not to be tempted
> in going down the so called black whole of using the gretag chart for
> profile evaluation, but for what I am trying to accomplish with CM is trying
> to match a known source color, in this case either the gretag chart, or
> ultimately the piece of art I am reproducing for my client.
>
> Am I going about this all wrong?
No. I think that there are limits to the whole "precision" of the system.
But until you document all your transforms, put them down to numbers, assign
some deltaEs, you don't really have a target. I am saying this in a general
sense, I don't mean you personally don't do any of that.
> I wish I could afford to fly one of you great color management guru's to
> Maui Hawaii for one on one consult, but I can't.
WoW! That would be nice. If you pay for the room I'll see about absorbing
the cost of the plane tickets ;-)
> Kornreich Fine Art Prints
> Steven Kornreich
Regards,
Roger Breton | Laval, Canada | email@hidden
http://pages.infinit.net/graxx
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden