QTR-Quadtone RIP

Just got a 2200 (used) and wanted to try out the QTR.  My monitor is 
calibrated with a Spyder 2.  First attempt on Epson Enhanced Matte and 
everything was printing very, very dark.  I had to slide the Gamma all 
the way to the left to get it even close.  I ran the calibration page 
at 100% and the Ink Limit looks to be around 40-45%.  Is this normal? 
I don't want to start the next step if I'm doing something wrong or if 
there is something up with the printer.  Thanks.

Dan

Dan wrote:
> Just got a 2200 (used) and wanted to try out the QTR.  My monitor is 
> calibrated with a Spyder 2.  First attempt on Epson Enhanced Matte and 
> everything was printing very, very dark.  I had to slide the Gamma all 
> the way to the left to get it even close.  I ran the calibration page 
> at 100% and the Ink Limit looks to be around 40-45%.  Is this normal? 
> I don't want to start the next step if I'm doing something wrong or if 
> there is something up with the printer.  Thanks.
> 
> Dan
> 

Dan

Most of the curves that come with qtr were designed to be used with a 
resolution of 1440 dpi. If you use 2880 dpi then the print will be too 
dark. I'm guessing you used 2880 dpi because you're ink limit at 40-45% 
would be in the right ball park for that resolution.

regards
Howard

Thank you Howard, you are correct.  I'm glad it was a simple answer 
and I'm very appreciative.  

I have very little working knowledge in printing so I feel like I'm 
just jumping in with both feet.  I was reading the QTR manual about 
making linearized curves.  It would seem that by adjusting the values 
there, you could really create some very customizeable curve tones.  I 
was wondering if there was any kind of software tool that anyone has 
created here that might help visually approximate how playing with the 
values would change the look of the 21 step chart.  Does that make 
sense?  You would create your basic curve settings with no 
linearization values, then measure the ink densities in each step to 
baseline it, then play to see how putting in different values would 
change the curve?  Does that make sense?  If nothing is out there like 
that, I may try to work on something.

Thanks again,
Dan
> Dan
> 
> Most of the curves that come with qtr were designed to be used with 
a 
> resolution of 1440 dpi. If you use 2880 dpi then the print will be 
too 
> dark. I'm guessing you used 2880 dpi because you're ink limit at 40-
45% 

> would be in the right ball park for that resolution.
> 
> regards
> Howard
>

Use the gamma 2.5 setting on your Spyder calibration if you are using 
Optical. A gamma 2.5 output device is what is expected by the PC for a 
linear-light output with no gamma power change. That will make your 
monitor mid-levels noticeably darker.

I'd like to think I'm wrong, but I don't believe there's any shortcut to 
curve creation.  For example even if you only change the shadow density 
on the grey tab, all of the ink curves alter, so unless you have access 
to the algorithms which generate the curves you will be at a bit of a 
loss in modeling how a printed step chart would react.  The problem is 
compounded by the fact that every change during curve creation also 
alters the maximum print density, so the each new curve iteration has to 
be compared against a new ideal density curve to see if it's close 
enough to warrant final linearization.  At least, that's my take on it.

I found that keeping careful notes and plotting measured densities 
against ideal densities in Excel for each curve creation iteration was 
the only reliable way (for me, anyway) to produce a final, linearized 
curve.  I also saved time by printing the step chart with several 
combinations of variables in one curve-creation "session".  This used up 
more ink and paper since I was, in effect, creating multiple threads of 
curve creation iterations, with only one being the best one.   It took a 
long time, but I'm hoping that my increasing experience will speed up 
the curve creation process.

What I would like to know is: just how close does a measured-density 
curve have to be to an ideal-density curve before it can be used in 
linearization?

Cheers,

Terence Lowe.