Digital BW, The Print

Robert wrote:

>Even the 100/100 warm IP5 profile uses a lot of photo cyan and photo
>magenta.

That will limit the warmth.  They probably have to in order to get a smooth
transition from the color-ink based highlights to the light-K midtones.

>IP5's success with the 2200 is attributed to their adding color
>to smooth out the highlights.  The key, is that unlike the Epson
>driver they don't add yellow...this is the key to the metamerism.

I'm not sure I agree that there is no yellow in there.  I don't see the
dots, but with a 1600 dpi scan, the eyedropper at 1 pixel does pick up
yellow.  I'm not sure, however, that if restricted to the highlights the
yellow would cause enough metamerism to matter.

>... I would really like to be able to get warmer without
>the dot gain...so I think your UC VM 4.3
>quad is absolutely the way to go for a metamerism free
>valuable tone solution.

S far is it is working fine.  The fade test seem to put it in the class with
the PiezoTone NW -- which is about the best there is.  I think we're at a
point with fade resistance where other things are almost certainly going to
be the limiting factors -- physical damage being the top of my list.

>  Can't wait to try it with Ink Jet Control...

A RIP is also going to be the way to go for the dual-K (Photo and Matte
black) systems that avoid the need to switch the inks.  The 7500 with that
system worked but the dither and dots were not acceptable.

>... what printers you know have RGB workflows for with the
>epson driver?

MIS has a fairly complete table at
http://www.inksupply.com/index.cfm?source=html/workflow_roark.html

I now also have curves for the 7500 -- using the 7000 driver.  The table is
also missing the 7000 cross-platform curves.

I might see if MIS can loan me a 1280 after the new inkset is finished and
out.  There seems to be too much trouble with the curves that are out there
for that printer.  I've never had one, and working through the mail or using
a scanner is just not very satisfactory.  Given the popularity of the
printer, I'd like to see if I can write a set of curves that is up to what
the 1270 and 1290 can do -- which is really terrific.

Paul
http://www.PaulRoark.com
__________________________________________________


On 3/5/03 6:18 PM, "Paul Roark" <paul.roark@...> wrote:

> Hi,
>
> I received some samples of 2200 IP5 output today (thanks Tom Fors) and was
> curious how they compared to the quads many of us use.
>
> In general, I think the 2200 output looks very good.  However, the dots in
> the highlights are apparent on some of the prints.  A 1600 dpi scan of the
> print makes it clear that much of this print is still a color print.  I
was
> surprised at how much color is still used even at 50%.  Then again, the
> UltraChrome light gray ink is quite warm.  So, even if for no other
reason,
> a fair amount of magenta and cyan has to be added to get the print
neutral.
>
> At any rate, looking at the highlight dots, I scanned a number of test
> strips at the 5% patch and used the Photoshop Histogram Standard Deviation
> as a measure of the relative roughness of the print at that point.  I
think
> this gives a reasonable measure that correlates with visual appearance.
(I
> have no authoritative citation for this, however.)
>
> At any rate, here are some measures of 5% patch standard deviations.  The
> lower the number, the smoother the print highlights.  I've also included
the
> 50% reading.  All samples are on EEM/EAM.  All were at 1440 as far as I
can
> tell.
>
> Printer/Driver-curve/Ink    5% Std. Dev.    50% Std. Dev.
>
> 2200/IP5/Neutral tint        5.54            5.03
>
> 870/neutral/MIS-VMS        2.15            3.56
>
> 3000/Neutral/VMS            4.80            7.91
>
> 1280/Neutral/MIS VM        2.56            3.38
>
> 1280/Piezo/FS-N            3.04            4.34
>
> 1280/Blk only/PT            7.46            10.2
>
>
> Paul
> http://www.PaulRoark.com