Digital BW, The Print

----- Original Message -----
From: "Anthony Atkielski" <atkielski@...>
To: <DigitalBlackandWhiteThePrint@yahoogroups.com>
Sent: Friday, August 02, 2002 10:14 AM
Subject: Re: [Digital BW] Storage of digital images


> Ernst writes:
>
> > Anthony, substitute 'essentialy opaque' for
> > 'almost transparant'. Then start all over again.
>
> The inks I'm looking at on inkjet prints are hardly transparent at all.

The 2000p is a bad example of an Epson printer with the first Epson pigment
inks available.
Though more opaque they still are transparent.
Put your prints on a light table, check the composite greys at 60/70 percent
and you will see that they are not opaque and overlap so they have to be
transparent in a way. Your offset prints will show the same but with a bit
more white in between.

> > And that also means that the ppi/dpi relation
> > is of another order than the ppi/lpi relation.
>
> The essential point is that the true resolution of an inkjet printer is
> nowhere near the machine-dot resolution.  You don't get 2880 pixels in one
> inch on an inkjet printer advertised as 2880-dpi.

The real resolution is usually 720 dpi while some go up to 1440. The 2880
and on the older models 1440 dpi is a shift of the dotrow by half a dot.
Printer dpi should be compared with image setter dpi, the last can go up to
4000 dpi though. However printer dpi isn't the only way to get differences
in tone as I wrote, droplet sizes count as well. On Epsons like the 1270
there are 3 sizes in a chosen resolution.

> > Tone differences and by that colourmixing can be
> > achieved in several ways: dot size variation, dot
> > frequency, ink layer thickness.
>
> I haven't seen evidence of ink-layer thickness being varied in inkjet
> prints, but perhaps some of the fanciest inkjets do this.  To do it,
they'd
> need much more transparent inks than the inks used on the Epson prints
I've
> seen.

The Iris prints have inklayer thickness differences. Their resolution wasn't
higher than 300 dpi but the droplet counted per spot was between 1 and 13 I
believe.

>> They all rely on transparent inks.
>
> You don't need transparent inks to vary dot size or frequency.  Indeed, if
> you have truly transparent inks, you don't need to use adjacent dots at
all;
> you can just overlay one dot on top of another, blending colors to produce
> one pixel per machine dot.  This is exactly what dye-sublimation does, and
> if you compare a dye-sub print to an inkjet print under a loupe, the
> difference is impossible to ignore.

Using CMY (K) inks requires subtractive mixing, like subtractive mixing
requires CMY (K) inks. In an ideal world they should be printed on top of
one another and be completely transparent. Like they are used in film an
photo material, and less ideal in dye sub and rotogravure. It is the
limitation of the offset process, paper specs, ink transparency that makes
it difficult so 4 raster screens have to be printed at angles, where the
lighter dots are separated from another and the dots will start to overlap
above 50%. Not an ideal colourmixing process. But with enough development in
150 years it is now a very good process.

> > Easiest method is getting a loupe and examples
> > of different printing systems.
>
> I have, and it supports my previous explanation.  While inkjet printing is
> excellent, I've not seen inkjet prints that exceed what is achieved by
> high-end offset in terms of resolution.  I haven't examined any high-end,
> small-sized inkjet prints, though (I've looked at large, high-end prints,
> but perhaps their resolution is deliberately lower).

You should examine the small sized  inkjet prints and the latests wide
format models of Epson and then the conclusion can only be that they exceed
offset quality. They did that already on colour gamut so inkjet proofing for
offset printing became possible.

 > In any case, inkjet still leaves something to be desired in a number of
> ways, but it does do a very nice job with the right printer under the
right
> circumstances.  My 2000P cannot match the quality from my MD-2300 dye-sub,
> but it's good enough for most purposes, and using the latter was such a
pain
> that it finally wore me down, which is why I went to inkjet.

Your 2000p is one of few mistakes in printer models that Epson launched. It
would perform far better with dye inks. The pigment inks that it uses are
'metamerism' prone and to suppress that the UCR is extended so black goes
further. That shouldn't be a problem if the droplet size was smaller, it
isn't. To overcome that problem in the newer models Epson gave them smaller
droplet sizes, another ink and an extra grey =light black (transparent of
course).

Ernst