Digital BW, The Print

Alessandro,

>...all I need is the density of each
>of the three grays relative to the black ink,

I'm not sure you need this if you are just writing curves for a 24-bit,
partitioning workflow.

> so I can cope with scanner
>inconsistency as long as the introduced variation is linear (i.e. equally
>applies to each of the inks, and not to one more than to others).

That is the problem -- the inconsistencies are not linear.  The gamma
differences push up or down the 50% density reading relative to the 100% and
0%.

But again, if your aim is a good looking 21-step test strip from your
workflow, why do you need the ink densities?  Just use trial and error.  You
can just assume that the FS cyan is about 8x%, and magenta about 3x%, and
the yellow 2x%.  All you really need to know is what the approximate
densities are -- and you have them.  Just start trying curves and see what
happens.  The vmqw and PzoRGB4 give you a good starting point.

>I hope
>what I wrote makes sense, the idea is clear in my mind but I'm afraid my
>english is not good enough :-)

Your English is great.

Paul