Yahoo Groups archive

QTR-Quadtone RIP

Index last updated: 2026-04-28 23:12 UTC

Thread

Lab values not in order error

Lab values not in order error

2005-11-03 by jdaalder

Hi

I've got an Epson 7800 and an Eye One unit and I'm trying to use QTR's
create ICC program (on Windows XP) to create an ICC profile for the
ABW mode.

I've printed the step wedges (51 grey random) with no colour management.

I've used measure tool to measure the patches (in reflective mode, no
spectral).

However when I drop the .txt file on to QTR-Create-ICC.exe, it always
gives me a .txt output file saying 'Lab values not in order, cannot
create profile'.

I'm sure it is something obvious but I can't figure out what is wrong,
nor can I find anything obvious in the eye one notes or the two yahoo
groups.

Thanks for any help!

Cheers
JD

Re: Lab values not in order error

2005-11-04 by wwodets

JD-

The problem you are having is because the LAB values are actually out 
of order (which QTR Create ICC after versions 2.3.0 will not 
tolerate).  If you look at the LAB values in the Measure Tool text 
file I think you will see that in the shadow range (L* 18 or lower) 
the numbers are not consecutively smaller as they would be if the 
patches were  consecutively darker.

I have not found this "reveral" of shadows with Epson papers, but I 
have had it on HPR and several other good papers printed through the 
ABW driver.

So your choices are:  linearize the printer with QTR; try different 
ABW settings (I find the light setting more linear than the darker 
setting); use a different paper (the Epson VFA, if you like the 
texture, produces a very linear response and the best DMAX, used with 
K3 inks, of any paper/ink combination around; or manually adjust the 
numbers in the text file to make them just linear (e.g. if they are 
15.35, 15.90 and 15.40, change them to 15.35, 15.34 and 15.33).  The 
latter option is a bit of a cheat and will reduce the shadow 
linearity and separation, though I'm not sure to an extent that is 
visually detectable.  Incidentally, I have *not* found that reducing 
the ink limit (in the ABW driver) corrects the reversal problem.

Walt



--- In QuadtoneRIP@yahoogroups.com, "jdaalder" <jdshop@h...> wrote:
>
> Hi
> 
> I've got an Epson 7800 and an Eye One unit and I'm trying to use 
QTR's
> create ICC program (on Windows XP) to create an ICC profile for the
> ABW mode.
> 
> I've printed the step wedges (51 grey random) with no colour 
management.
> 
> I've used measure tool to measure the patches (in reflective mode, 
no
> spectral).
> 
> However when I drop the .txt file on to QTR-Create-ICC.exe, it 
always
> gives me a .txt output file saying 'Lab values not in order, cannot
> create profile'.
> 
> I'm sure it is something obvious but I can't figure out what is 
wrong,
> nor can I find anything obvious in the eye one notes or the two 
yahoo
Show quoted textHide quoted text
> groups.
> 
> Thanks for any help!
> 
> Cheers
> JD
>

Re: Lab values not in order error

2005-11-04 by jdaalder

Walt

Thanks for your very helpful reply.  Of course I should have just
considered the message literally, but it seems almost too obvious
somehow!!  

The paper is indeed HPR.  I'm pretty committed to it.  Not
particualrly interested in the Epson papers, for a whole host of
reasons, so I would like to find a solution.  I also use Torchon
pretty extensively.

> So your choices are:  linearize the printer with QTR; 

By this, do you mean print with QTR and linearise its curves?  Happy
to do this but am waiting on curves for the 7800 which don't appear to
exist yet (QTR 2.3.1, PC).  I presume I can create my own curves but I
thought I'd leave it to the experts!  I suppose I can still create an
ICC profile with QTR-create-ICC once those curves are in place and
linearised, and then I have the best of both worlds perhaps.  I find
soft-proofing to be pretty much essential!

try different 
> ABW settings (I find the light setting more linear than the darker 
> setting)

I will try this, thanks - I take it the end points (ie dmax/dmin) stay
the same, it is simply the ramp (curve) inbetween that changes - ie I
won't lose DMAX by using 'light' over 'dark')

> or manually adjust the 
> numbers in the text file to make them just linear (e.g. if they are 
> 15.35, 15.90 and 15.40, change them to 15.35, 15.34 and 15.33).  The 
> latter option is a bit of a cheat and will reduce the shadow 
> linearity and separation, though I'm not sure to an extent that is 
> visually detectable.  Incidentally, I have *not* found that reducing 
> the ink limit (in the ABW driver) corrects the reversal problem.

It take it that others potentially have from the way you say this?



Thanks again for your helpful reply

JD
Show quoted textHide quoted text
> 
> 
> --- In QuadtoneRIP@yahoogroups.com, "jdaalder" <jdshop@h...> wrote:
> >
> > Hi
> > 
> > I've got an Epson 7800 and an Eye One unit and I'm trying to use 
> QTR's
> > create ICC program (on Windows XP) to create an ICC profile for the
> > ABW mode.
> > 
> > I've printed the step wedges (51 grey random) with no colour 
> management.
> > 
> > I've used measure tool to measure the patches (in reflective mode, 
> no
> > spectral).
> > 
> > However when I drop the .txt file on to QTR-Create-ICC.exe, it 
> always
> > gives me a .txt output file saying 'Lab values not in order, cannot
> > create profile'.
> > 
> > I'm sure it is something obvious but I can't figure out what is 
> wrong,
> > nor can I find anything obvious in the eye one notes or the two 
> yahoo
> > groups.
> > 
> > Thanks for any help!
> > 
> > Cheers
> > JD
> >
>

Re: Lab values not in order error

2005-11-04 by wwodets

> Thanks for your very helpful reply.  Of course I should have just
> considered the message literally, but it seems almost too obvious
> somehow!!  

Yes, it also took me a while to actually read the error message when 
this first happened.
> 
> The paper is indeed HPR.  I'm pretty committed to it.  Not
> particualrly interested in the Epson papers, for a whole host of
> reasons, so I would like to find a solution.  I also use Torchon
> pretty extensively.

This avoidance of Epson papers in quite common and I'm curious about 
your reasons.
> 
> > So your choices are:  linearize the printer with QTR; 
> 
> By this, do you mean print with QTR and linearise its curves?  Happy
> to do this but am waiting on curves for the 7800 which don't appear 
to
> exist yet (QTR 2.3.1, PC).  I presume I can create my own curves 
but I
> thought I'd leave it to the experts!  I suppose I can still create 
an
> ICC profile with QTR-create-ICC once those curves are in place and
> linearised, and then I have the best of both worlds perhaps.  I find
> soft-proofing to be pretty much essential!

Yes, I meant using QTR curves and printing through QTR.  Although an 
ICC profile does some "linearization," I don't think this is the way 
to correct large problems.  Linearizing with QTR and then using the 
ICC profile would be ideal, and I think Steve Kale (on the BW Print 
board) is doing that.  Fortunately, the ABW driver is overall quite 
linear, though it tends to compress the deep shadows on the "darker" 
setting.

I too find soft proofing essential, but with my workflow there is 
very little difference in the two screen views.  With the HPR you 
will have to experiment with "Paper white/ink black" in the soft 
proof.  With the VFA I find that the preview is better without that, 
with EEM or HPR I find it better with those options checked.
> 
> try different 
> > ABW settings (I find the light setting more linear than the 
darker 
> > setting)
> 
> I will try this, thanks - I take it the end points (ie dmax/dmin) 
stay
> the same, it is simply the ramp (curve) inbetween that changes - ie 
I
> won't lose DMAX by using 'light' over 'dark')

The various ABW settings (darker, dark, normal, light) adjust gamma, 
so the end points should be about the same.  The light setting 
produces a lighter gamma, turning a 90% gray from a L* 18 or so into 
22 or so.  On HPR this lineaized the curve for me, right down into 
the deep shadows.
> 
> > or manually adjust the 
> > numbers in the text file to make them just linear (e.g. if they 
are 
> > 15.35, 15.90 and 15.40, change them to 15.35, 15.34 and 15.33).  
The 
> > latter option is a bit of a cheat and will reduce the shadow 
> > linearity and separation, though I'm not sure to an extent that 
is 
> > visually detectable.  Incidentally, I have *not* found that 
reducing 
> > the ink limit (in the ABW driver) corrects the reversal problem.
> 
> It take it that others potentially have from the way you say this?
> 

I have manually rewritten the numbers and it works.  IF your question 
is about adjusting ink flow in the ABW driver, I am not aware of 
anyone who has linearized the shadows with this.  Clayton Jones feels 
(by eye) that his 2400 puts down too much ink and uses, I believe, a 
minus 5% ink flow for all his printing.  I haven't found an advantage 
in this with VFA, but Clayton is a very good printer.

One thing I didn't mention is selection of paper type in the ABW 
driver.  Though Hahnemuhle recommends a VFA setting, it might be 
worth trying something like EEM.  I think there is too much ink on 
the HPR (and the measured reversals support this) and that the sheen 
of the ink is making the dmax look even worse than it is.  
Regardless, in my experiments, it doesn't look or measure very well 
compared to the VFA. 

My work for the last six months has been on a 2400.  Today I'm 
supposed to receive a 4800, so I'll see if there's a difference in 
all this.  I prefer the HPR on everything but image quality (surface, 
color, panache, etc.) and wish I could use it instead of the VFA.  
But a side-by-side comparison of the two papers is, to me, too 
powerful.  I had hoped the 4800 would correct some of this, but it 
sounds from your experience (and Steve Kale's) as if it won't.    

Best,
Walt

Re: Lab values not in order error

2005-11-04 by lours51

--- In QuadtoneRIP@yahoogroups.com, "wwodets" <odets@c...> wrote:
>
> JD-
> 
> The problem you are having is because the LAB values are actually 
out 
> of order (which QTR Create ICC after versions 2.3.0 will not 
> tolerate).  ......... manually adjust the 
> numbers in the text file to make them just linear (e.g. if they are 
> 15.35, 15.90 and 15.40, change them to 15.35, 15.34 and 
15.33)..........
  
> > Walt
> 

In a previous post I exposed this same linearization problem I met. I 
have tried with different versions of QTR (2.3 & 2.3.1), I have try 
changing the numbers in the Txt file.....it doesn't work no matter 
what.
I spent two weeks on this....until I get sick.
(WindowsXP - Epson4000 - UT7 inksets)

Michel

Move to quarantaine

This moves the raw source file on disk only. The archive index is not changed automatically, so you still need to run a manual refresh afterward.