Of color consistency
Of color consistency
- Subject: Of color consistency
- From: Henrik Holmegaard <email@hidden>
- Date: Sat, 1 Sep 2001 09:52:03 +0200
Larry Wilson <email@hidden>
Therefore, this spec is for
variation within a sheet, not over time, within a run, etc. I tried
to get specs for sheet to sheet delta E, but no one could supply me
with it.
We have had jobs rejected due to differences
between the proof and final run. Of course, these differences are
most noticeable in light tints.
For the past three years or so I've regularly posted here on the
list, explaining how to use LOGO ColorLab, and more recently the
MeasureTool in v3 to get the information you are looking for:
a. print a test chart multiple times through the day (use the timing
feature in the Mac OS PostScript driver),
b. pick every X print (every fifth, every tenth ...) and measure the patches,
c. use the Comparing function to get you the average and the outliers.
The history of it is that the averaging and comparing functions in
the developer tool, ColorLab, allowed the LOGO folks to create the
statistics for printer profiles bundled with LogoProof for Kodak, a
color server that eventually evolved into iQueue, and other OEM
projects.
Xerox's specs say Delta E of 4, as calculated by measuring
two points on a single sheet of paper.
Toner-based printing systems are not recommended for proofing, and
for two reasons:
d. proof-to-proof consistency is too poor, by comparison if I print
four 323 patch Eye-One CMYK test charts across the format of a
self-calibrating HP 5000, the average deviation is 0.5 dE.
e. the gamut volume of toner-based printing systems is not sufficient
for proofing, for instance, if you proof ISO semi-matte offset on an
Apple Color LaserWriter, a third of your offset greens, yellows, reds
and magentas are out of gamut for the studio printer, as you can
easily see in Steve's ColorThink, and the proof to print dE would be
a fair bit more than dE 4 -:).
Which of course is also what you are observing. So this is a case of
not doing your homework before shopping for a studio presentation and
proof printer :
f. when shopping, first ask for test chart prints and measure the
white point and black point using a quality spectrophotometer, one
test chart for each colorant and substrate combination, then
g. measure the patches, create profiles and study the gamut volume in
a 2D or preferably 3D gamut comparison tool, then
h. write the cheque or write the printer off -:)
Hope this helps