Digital BW, The Print

Austin, and all!

I personally would be surprised if 100% black printed with just black on
Piezo. This is because I guess that the starting point for Piezo is a
theoretical range from Pure-black to pure paper-white (in terms of
half-toning / dithering patterns). The profile will then set 'end-points'
along the 'line of theoretically available tones' that could well (for
ink-limits or whatever) mean that more than 'just black' is used. Going back
several months, having a theoretical range of tones rather larger than 255
is great... because it does give the system flexibility to profile for
different media and still leave  - let's stay on the safe side >100
distinguishable tones.

Also, I think there may have been some evidence from the past where someone
had problems with just black ink (100%), but converting to 99% (presumably
introducing a second grey into the dither) helped get rid of the patterns
altogether. Of course... _that_ would imply that there was just black in
100% :) This may be a profile related.

I'll be interested to hear your results...

Unfortunately, I think your comparison test will be interesting, but not
representative of 1290 performance - in this respect, I believe the more
modern printers favour the development of 'non-Piezo' quad products that use
the Epson driver. I agree with you that printing line patterns is not the
best way to test a quad-printing 'technology'. That's not to say I don't
still Love Piezo, just that the newer printers give alternatives something
of a chanve to catch-up(!). OK I know that is a personal-taste related point
:)

FWIW, Piezo is FANTASTIC for smooth tone changes (IMO) and I can not speak
for MIS alternatives. However, we should all appreciate that some resampling
is going on, probably in both workflows, and that probably (possibly) both
are using some kind of bicubic interpolation (I think Piezo is described as
using 'stochastic error diffusion'. In other words, hard-edges will be
softened... it is inevitable, just print some text under Piezo and you will
see. In a black-on-white example, I believe the effect is bot great... but
works perfectly in con-tone images.

Final thought. Piezo only ever prints in graphics mode. It is _possible_ I
believe for the Epson driver to print in graphics and text modes at once.
This may effect results, but that's a big maybe. However, to me, a logical
conclusion to be drawn from 'interpolation' is that at some point, data is
'added' by Piezo... and probably the Epson driver too...
...we just don't know how much data can be 'added' successfully by each
system, or where it's effectiveness cuts-off (in the way it is _programmed_)
unless someone has internal details I don't know of.

Lighten up everyone! Can we keep this non-tribal if possible?

Best,
Nij



> -----Original Message-----
> From: Austin Franklin [mailto:darkroom@...]
>
> eMail me the square you made, and I'll print it out.  I have a 90x
> microscope, as well as an XRite 810 densitometer.  I can tell exactly what
> the printer is doing, and I'll let you know.  It will take me a
> day or so to
> do the test.  I happen to have one 3000 setup with Piezo inks,
> and one 3000
> setup with Epson inks, so I can make four test prints.
>
> Pure black should only print using black ink.  Actually, it'll be
> better, I
> think, to do this test using the Epson inks...since you'd see color if the
> Piezo driver is in fact using any of the other inks, right?