Digital BW, The Print

I'm like the atheist who prays to God - just in case.

I'm not a perfect-histogram monger, but I do like to know the deal, in case
I change my mind later.

Paul's RGB separation curves are a fairly radical adjustment for an 8-bit
image to undergo. Not unduly so, but if you've already stressed an 8-bit
file on the way, these curves won't help the case. So far, I haven't noticed
that working the file all the way through to print (including the
application of Paul's curves) in 16-bit gives any better a print, but that
is perhaps another topic. Those who participated in the PiezoBW 30 print
exchange know that Ron Landucci, Paul Roark, and Jerry Olsen, contributed
prints using Paul's VT method, and curves, that were at least the equal of
any prints in that exchange.

Thus, my question is somewhat theoretical, but I'm trying to get my head
around some of these concepts. Is a grayscale workflow, going through the
PiezoBW profiles, and the PiezoBW RIP, more gentle to the tonal data (bits)
of a file, relative to an RGB file, with RGB separation curves, and the
Epson RIP (driver)? Is Piezo somehow getting around separation curves, or
are they just contained in the profiles?

BTW, is that all profiles are - a curve set? Somehow they seem to be doing a
kind of "selective color" treatment, which seems a bit more complex than
curves, but that's just based upon my observation, I know nothing about
them.

Todd