[Leica] Canon i950 review, part 2

[Martin Howard <mvhoward@mac.com>]

I spent one particularly frustrating weekend refining 4x6" wedge prints 
by
manipulating the manual colour controls in the printer driver until I'd
managed to cancel out the green in the shadows, only to discover the
following morning when I opened the blinds to the light from a northern 
sky
that the appearance of an inkjet printout can *radically* change under
different lighting conditions (I've later learnt that this is called
"metamerism").  What was a neutral print under tungsten lighting became 
an
overtly magenta print in natural light.  While a green cast is decidely
unpleasant, a strong magenta cast is downright ghastly.

Likewise, I'd printed quadtones during the evening using the setting "Bl
CG10 CG4 WMG3" (an incantation written on the back of many of my prints
from that evening).  These were a nice duotone with neutral shadows and
slightly warm midtones and highlights with good detail in each under
tungsten lighting.  In natural daylight, the print had blue shadows with
green midtones and highlights.

Unless you've experienced this phenomenon first hand, it's difficult to
comprehend just how large the tonal shift can be in inkjet prints under
different lighting.  And these are both continuous spectra light 
sources,
the only difference being one was a 3200K 60 watt lightbulb, the other
5000K midday sunlight from a northern window.  Under a partial spectra, 
I
could understand, but this just left me baffled, frustrated, and short 
of
paper.

I kept notes during this and noticed that there are a bunch of other
factors that also affect the tonal balance of an image.  The paper 
settings
in the printer driver appear to have some effect on the colour.  I'm
guessing this is because it has to do with how much ink gets squirted 
out,
and the times (and absorbtion rates) between squirting out the ink.

Print size appeared to affect the colour too, although I'm not entirely
sure that this isn't an optical illusion.  Smaller prints appear more
saturated than larger ones, but I suspect that this is an error on my 
part
by not standardizing viewing distances so that prints have the same
apparent size.

Dry down is definately in effect in inkjet prints.  It differs for
different papers, but I've found that at the very minimum you need to 
wait
one hour before assessing tonal qualities and waiting overnight is 
better
yet (6+ hours).

Interestingly enough, different papers paper exhibit different amounts 
of
metamerism too.  The papers I've used are:

	* Canon Photo Paper Pro, 4x6"
	* Canon Photo Paper Pro, 8.5x11"
	* Epson Borderless Premium Glossy Photo Paper, 5x7"
	* Epson Premium Semigloss Photo Paper, 8.5x11"
	* Epson Matte Paper Heavyweight, 8.5x11"
	* Epson Enhanced (formerly Archival) Matte Paper, 8.5x11"
	* Ilford Inket Fine Art Matte Paper, 8.5x11"

Of these, the matte papers show less colour shift in differing light 
than
do semiglossy or glossy ones, and of the matte papers, the uncoated Fine
Art exhibits least shift of all (it looks almost identical in tungsten 
and
daylight).

Printing exactly the same image with exactly the same settings at 
exactly
the same size on these different papers is an interesting exercise.  I 
have
prints from the same file (a slightly warmtoned RGB image from a scanned
Tri-X neg) on all letter sized papers and they vary from green-toned 
(Matte
Heavyweight) through to warm neutral (Ilford Fine Art).  Same 1s and 0s
going to the printer.

Surface wise, I like the Epson Semigloss best of the "RC" papers and the
Epson Enhanced best of the "FB" papers.  Unfortunately, the i950 does 
not
like Epson Enhanced Matte at all and produces hideously colour cast 
prints
on it, which differ *radically* from what I see on the screen.  To the
eye, virgin Epson Enhanced Matte and Epson Matte Heavyweight are
indistinguishable.  Go figure.

So, I've more or less decided that the solution to this is as follows: 
Pick
one or two papers (I've settled on the Epson Semigloss and Ilford Fine 
Art
Matte), decide on what settings to use in the printer (BJ colour
management, no manual colour control) and Photoshop and just use those.

In Photoshop, I've found that, for me, the following works: After I've
finished working on my image and I want to print it, I target it for the
i950 and the Fine Art or Semigloss paper.

	 * Create a Hue/Saturation layer, checking the "colourize" box.
	 * Make the layer blend mode "Soft Light"
	 * Set layer opacity to 100%
	 * Set "Hue" to 44, "Saturation" to 25, and "Lightness" to +30.

When I print, I use Photoshop's "Print with Preview" and in the colour
management section ("Show more options...") I select "Document" as the
source space, "BJ Color Printer Profile 2000" as the print space,
"Relative Colorimetric" as the "intent" and "use blackpoint
compensation".  In the printer, I select paper type, check the "high
quality photo" option, use "Canon BJ colour management" at the "photo"
setting, and leave manual controls alone.

This gives warm/neutral prints on both papers, although slightly warmer
highlights on the Ilford paper.  I should think that this is mostly due 
to
the Epson Semigloss being whiter than the Ilford Fine Art Matte.

(Why do I still use the BJ printer driver colour management?  Well, I've
discovered that the variation between different types of paper appears
less when it is checked than when it is not.  I'm guessing that the
firmware in the printer does colour correction compensations to 
compensate
[in turn] for the adjustments in printing method depending upon paper
type.  Or maybe not.  In any case, it seems to result in less 
variation.)


EPILOGUE

So, will I be shelling out on the $195 Piezography BW set for the Canon
i950 when it arrives?  Probably, yes.  The reason is metamerism.  I'm
guessing that quadblack inks exhibit considerably less of this (if any 
at
all) simply because the inks only contain black.  I'm guessing that
metamerism in inkjet printers chiefly comes from the fact that the 
various
dyes or pigments reflect light of different wavelengths differently from
each other.  Using the same (black) ink at various dilutions shouldn't
present this problem.  While it doesn't bother me all that much, it 
makes
it harder to send prints to other people, not knowing what light they 
are
going to view them under.

Until then, and for my 5x7" work prints, my current solution is pretty
good.  And while Ilford Fine Art is only available in the 8.5x11" size, 
a
$25 paper trimmer will allow you to chop it down to 5.5x8.5", which is
actually a pretty attractive size that fits an enlarged 35mm negative a
lot better than 5x7" does.

But in some ways, I think Jim Shulman nailed it when he said, "Digital
printing is the TV dinner of photography."

- --
To unsubscribe, see http://mejac.palo-alto.ca.us/leica-users/unsub.html