Digital BW, The Print

This will be old news to many, but I thought I'd relate my initial
experiences with my recently acquired 7600.

 

The printer had OEM UltraChrome inks in it, so I did a little experimenting
with these to see what they could do.  My first surprise was how good the
driver with just "Color Controls" checked printed a grayscale file on EEM.
If only indoor lighting is involved, the metamerism was even less than I
expected.  Unfortunately (for Epson) outside light turned the B&W green
(metamerism).

 

The bad news with the 7600 and UltraChrome matte black is a really weak
dmax.  The only paper I tried that could break 1.6 with the Epson driver was
EEM.  All the others I tested were below my "acceptable" 1.6 line.
PremierArt 205 could only hit 1.5.  Sad.

 

The easiest way to get a better dmax appears to be to install Eboni.  With
that EEM occasionally hits 1.65.  That is the point at which I think there
is very little more to be gained in actual prints.  This also pulls
PermaJet, Photo Rag and Innova to just above the 1.6 line.  PremierArt
doesn't quite make the grade.

 

This is the first printer where I will be using a RIP.  With IJC I can hit
1.69 with my old batch of EEM.

 

With the OEM color inkset, the best dmax appears to be with a rip and cyan
added to the 100% point.  In actual IJC profiles it appears that the light
inks need to be turned off one point before the black point.  That is,
perhaps because of an artifact of the smoothing algorithm and interaction of
the jets, even if the light cyan ink is off at 100%, the dmax does not hit
its highest potential.  So, I put a second point as close to the black point
as possible and have it at 0 just like the black point.  It works.  That
allows the cyan and mk to reach the maximum dmax I see in the linearization
tests. 

 

I've now loaded the UT7 inkset into the 7600.  My first impression with this
inkset is that the combination prints a nearly perfect EEM B&W with a
grayscale file, Epson driver, Color Controls checked, and the sliders at 0.
With Eboni able to hit 1.65, this EEM printing is hard to beat if archival
paper is not an issue.

 

With the UT7 inkset also, a rip can significantly increase the dmax.  The
best dmax with UT7 is with a second Eboni in the yellow spot.  However, it's
not worth losing the slider printing.  Just by adding some dark cool gray
(magenta position ink) to the 100% spot, I'm getting within 0.01 unit of the
dual Eboni on all papers I've tested.

 

So, with UT7, the route to the best dmax I'll be taking with matte papers is
to just add M (dark cool gray) to the black point via a rip.

 

With glossy printing, the Epson driver with RGB curves does a great job.  A
Kirkland glossy with the 2200 UT7 Ilford curve is excellent, with an initial
dmax of 2.45 (with Eboni in the K spot but not used by the curves).  The
Kirkland print has almost no bronzing even without glop.

 

Paul

www.PaulRoark.com <http://www.paulroark.com/>  

 



[Non-text portions of this message have been removed]

> So, with UT7, the route to the best dmax I'll be taking with matte papers is
> to just add M (dark cool gray) to the black point via a rip.

How much do you add? Do you run it full strength? Does it work the same 
way on the 2100?

--
Daniel Staver
http://daniel.staver.no

> > So, with UT7, the route to the best dmax I'll be taking 
>>with matte papers is to just add M (dark cool gray) to the 
>>black point via a rip.
> 
> How much do you add?

If you mean dmax, it adds up to 0.06 density units where it is needed the
most.  So, PremierArt 205 which hits only 1.50 with UC MK, hit 1.55 with
Eboni and the Epson driver, and is up to 1.61 with the rip, Eboni and UT7 M.
PermaJet and Innova papers go from about 1.61 with Eboni and the Epson
driver to about 1.66.  With Photo Rag it appears to add almost nothing.  PR
gets 1.64 - 1.65 with Eboni with either the Epson driver or the rip, but
more experimenting might increase this a bit.

>Do you run it full strength?

It takes very little UT7-M to do the trick.  It varies, but usually backing
off the K ink load from 26 to 25 makes "room" for the UT7-M addition.

> Does it work the same way on the 2100?

The 2200 (and I assume the 2100 is the same) does not have the same dmax
problem that I'm seeing with the 7600 and 4000.  So, I have not experimented
that much with it -- unless I've intentionally crippled the machine.  To
"simulate" the 7600 I used the 2200 with the black ink load backed off from
26 to 20 in IJC.  This allowed a lot more experimentation that changing
carts on a 7600 could do.  However the simulation was not entirely accurate.
It is not just an ink load issue that differentiates the printers.  

The trick with the 7600 is to reach the dmax levels that most older printers
can reach with the Epson driver and no tricks at all.  For most of the older
printers, the EEM or Matte Paper Heavyweight ink loads put the papers at
about the most they can achieve with any combination of pigments as far as I
can see.  If you add more ink to these loads, the dmax usually drops.  That
is, you've reached the top of the hill.  

With the 7600 and 4000, the top of the hill is lower if only one ink is
firing.  The way to get that top higher seems to be to add a second ink,
even if the density of the second ink is less.  The second ink seems to be
filling in spots that were not well covered by the first. 

Note that the time the first ink has to sink in before the second hits the
paper is one of the variables.  With the 2200, I found that the dmax and
optimum load increased slightly as the second Eboni was moved from the LK to
the Y spot.  Y is the last jet to fire.  I speculate that part of what is
going on is the time it takes for the ink base to be absorbed.  

Especially with the UC MK, the cyan also seems to be doing some of what the
cyan does for the PKN.  It is probably due to the relative colors of the
inks.  UC MK is so warm that the cyan may be mopping up that warm color.
I'm obviously speculating a bit here, but the UC MK + C combination is the
most dramatic improvement I've seen.


Paul
www.PaulRoark.com