QTR-Quadtone RIP

I have several densitometers but they were in storage 1000mi. away at
the time I tried QTR out. Being the impatient sort,I used an epson
3170 and an IT8 chart to build the curves for a self mixed Quad set on
a 3000. It worked very well indeed. Scan the IT8 or any other
calibrated step scale along with your test scales.In photoshop, adjust
levels to match Lab L* on the 22nd(not the very darkest) and min
patches of the IT8. Then just read off the Lab L* values for your QTR
patches using the eyedropper tool or dragging the curser over them.
Yes,you may need to do a slight bit of fudging in the darkest values
if your paper/ink combination just happens to have a great d'max
capability, as mentioned below.Photoshop only shows 1% intervals for
Lab L* so high densitys can be very close or the same reading.You can
get around that by switching to rgb mode and looking at the patches to
see a bit more separation to determine which is darker. Then just
nudge your L* figure on that patch a bit lower and use it as input to
QTR for the linearisation.The same thing is true for the lightest
patches(generaly only the last 2) where it is little problem. If they
both read something like 99 L*, just call the last one maybe 99.1 L*
and go. Just be certain you don't give QTR any two readings that are
identical and you will be fine. Once you have generated the
linearisation curve, use it to reprint the scale. If it looks good and
shows reasonable separation of all patches it is quite useable to make
excellent prints. 

Having said all this, I want to make clear to all and sundry, that I
am not calling it a better or just as good way to generate the curves.
I am simply saying it is very doable and works. When I did get around
to dragging out one of my instruments this fall I took the readings
below from the linearised QTR stepscale.They tell me, to my own
satisfaction, that the above method didn't do a bad job.

Regards
Duane

VMCOOL 21STEP PREM DUOBRITE
 
 #   CIE-X  CIE-Y  CIE-Z   L*     a*     b*     dVIS' dRED' dGRN' dBLU'   
 
 1   1.310  1.211  1.219 10.637    4.49    -3.14 1.917 1.920 1.917 1.812  
 2   1.912  1.791  1.773 14.348    4.55    -3.27 1.747 1.717 1.753 1.651  
 3   2.562  2.457  2.372 17.724    3.84    -3.12 1.610 1.582 1.618 1.522  
 4   3.841  3.522  3.396 22.025    6.87    -3.48 1.454 1.410 1.455 1.375  
 5   5.387  5.103  4.782 27.025    5.71    -3.21 1.293 1.254 1.294 1.225  
 6   7.262  6.874  6.291 31.519    6.34    -2.87 1.163 1.126 1.159
1.109   
 7   9.167  8.856  7.931 35.706    5.35    -2.46 1.053 1.021 1.053
1.010   
 8  11.347 11.193  9.842 39.903    4.08    -2.06 0.952 0.929 0.951
0.920   
 9  14.671 14.518 12.559 44.967    4.16    -1.66 0.839 0.817 0.840
0.814   
10  18.311 18.366 15.673 49.937    3.21    -1.27 0.736 0.716 0.738
0.722   
11  21.736 21.838 18.560 53.854    3.23    -1.18 0.661 0.646 0.665 0.646 
12  25.361 25.590 21.716 57.646    2.94    -1.18 0.592 0.575 0.594
0.575   
13  29.113 29.294 25.069 61.040    3.39    -1.61 0.534 0.515 0.534
0.517   
14  33.145 33.079 28.488 64.225    4.49    -1.98 0.481 0.463 0.479
0.464   
15  38.691 38.771 33.089 68.585    4.23    -1.64 0.412 0.396 0.414
0.402   
16  44.043 44.253 38.264 72.397    4.08    -2.39 0.355 0.336 0.359
0.341   
17  47.960 48.784 41.697 75.317    2.58    -1.85 0.312 0.299 0.315
0.304   
18  55.113 55.905 47.888 79.559    3.09    -2.06 0.253 0.246 0.255
0.244   
19  62.013 63.700 54.457 83.809    1.41    -2.03 0.196 0.196 0.200
0.190   
20  69.493 70.753 61.323 87.365    2.79    -2.93 0.151 0.149 0.156
0.137   
21  84.687 84.531 72.242 93.681    6.11    -2.21 0.073 0.069 0.074
0.074   





--- In QuadtoneRIP@yahoogroups.com, "Tom Moore" <r.t.moore@r...> wrote:
>
> I have an X-Rite 810 that I bought off Ebay. I paid about $150 for
it about
> a year ago. That included several alignment step tablets. They may be
> cheaper today.
> 
> I also have an Epson 2400 scanner and a Fauste IT-8 calibration
chart. With
> Vuescan I can calibrate the scanner using the IT-8 chart.
> 
> I have tried making curves with the scanner and using the densitometer.
> Aside from the convenience of being able to import density values
directly
> into a spreadsheet from the densitometer (important if you do a lot and
> value your time), the scanner seems to lack precision near the ends
of the
> greyscale. The dark patches in particular don't measure the same when
> measured with the scanner as they do when measure with the
densitometer. 
> 
> Perhaps I didn't work at it enough. I know Paul Roark's web site
(see the
> BlackandWhitethePrint list) has a procedure for "calibrating" your
scanner
> that might be applicable to those using one to build QTR curves.
> 
> The possible benefits that Profile Prism might provide would be
> 1.	better calibration of the scanner
> 2.	automation of entry of step density values into a file in a form
> usable by QTR
> 
> As a densitometer, the X-Rite 810 only measures the L axis. It doesn't
> provide any useful a and b axis information (in Lab terms). Using PP
might
> be a cheap way to get tone information into ICC profiles created by
QTR - as
> opposed to the Eye-1 at $800.
> 
> Tom Moore
>