The RNP Arrays often get you close enough to be "in-the-ballpark" to the proper blocking colour, sometimes it nails it, but, because all the Arrays go up in steps there's a good possibility they can be a little bit off -- too dense usually -- when picking the first white square. I usually print the initially selected colour out to the 101 Step wedge and then look for and count the number of white blocks/steps until I see a "contrast change" if any, and usually there is.
For example, in the illustration below a 101-Step print of a Cyanotype was printed using an initial negative blocking colour of RGB 38,255,38. The white steps however go on from 0 to about step 12 before any change in contrast can be noticed.
The technique I use to determine "change" in contrast is to open "Levels" in Photoshop, and, while holding down the Alt/Opt key, slide (or "scrub" to use the proper editing term) either the White Point or Black Point sliders to the left or right. This shows an animated high contrast version of the image and allows you to visualise the 101-Step "in motion". On this one the first 12 steps fade in and out at the same time for the most part denoting they are of the same contrast. Sometimes prints are slightly stained by the emulsion which can cause novices to mistake this colour for density when in fact it really "veiling" so be careful, a print which hasn't cleared properly can fool you!
I go back to the original 101-Step Photoshop file that was colourised with the original blocking colour (38,255,38) and I measure the colour of step 12, using the eye-dropper tool, to "refine" the blocking colour.
The sampled square should read 64,255,64 which if you find that square on the HSL Array model it's a step and a bit above the first colour choice. Travelling up the Array vertically toward the top, the "less dense" area.
I refill a new negative 101-Step with the new colour (i.e. 64,255,64) sampled from step 12. The technique insures your curves will always consistently start at 0,0 instead of 0,12 (where the curve might have started had I not done this refinement) It would have produced what I refer to as a "lifted toe" curve.
The change between the two above 101-Steps seems hardly noticeable -- they both look pretty green to me! But subtle or not the change in blocking colour will have a profound effect on the starting point of the curve. It should be said that if this refinement is not performed you can still produce a valid curve if you punch in the numbers correctly by hand or use ChartThrob, BUT, it's my experience that the curve will be slightly more prone to the introduction of errors as the curve has to now compensate for the removal of excess colour. In addition to the curve starting "up the Y-axis" you also tend to see a very low slope in the early portions of the curve. Why make life any harder than it already is. Have fun curving.
13 July 2007.
There is always more than one way to skin a cat. This colour refinement technique uses values in the same colour "hue" to refine the initial colour choice. This tweak travels vertically up the HSL colour model changing the Lightness (or Brightness) of the colour, which, if you're not hung up on having to use colours only on the equator is fine. Upon reflection however you could almost as easily take a "horizontal" tuning approach where the "hue" became the variable. You would have to print a step wedge which showed fine granularity of the colours changing similar to the RGB Box wedge I created on the Arrays page. Just how fine a wedge might depend on the process but given that there are 256 step available I would divide that number either 16, 8, 4 or 2 to produce 17, 33, 65 or 129 step step-wedges. 256 steps would probably be overkill. Hint: making a 65 step wedge that goes from pure red to pure green to blue varying one colour at a time in steps of 4 would give you 195 colour choices. Lightness (i.e. saturation and brightness) could be set to 100 per cent or could be fixed in the lighter or darker areas of the HSL model (100,50 or 50, 100 or somewhere in between, depending on where the density border was located. Using the Image|Adjustments|Hue/Saturation menu in Photoshop brings up a handy "Lightness" slider than can be changed from -100 to +100 if you want to experiment.
Density "Box" wedge contains 33 step (I'm talking about the area between the black lines) representing pure RGB colours of the HSL Model.