Filter
Top Products
Criteria for Defining Print Life
I
t is very important to know which degradation
mechanism is the limiting mechanism in any given
s
torage environment. It is also important to know
the print-life criteria being used in reporting print
life. The print-life criteria determine “how bad is
bad”—for example, when a consumer would take a
print down and throw it away.
Kodak uses the illustrative criteria stated in the
ANSI/ISO standard, which places dye-fade limits at
30 percent.
4
,5
Internal studies done by Kodak
indicate that 30-percent fade is a conservative limit.
T
able 1
c
ompares dye-fade limits, consumer
descriptors, and consumers’ perceptions or
reactions.
6
Table 1
Approximate Correlation of Color Descriptors to
Dye Loss from 1.0 Density
Depending on the scene content, fade levels below
a
bout 15 to 20 percent are usually not noticeable
without direct comparison to an unfaded reference
image. A fade level between 20 and 30 percent
could be considered noticeable without a non-faded
comparison, but typically would not be considered
o
bjectionable. Fade levels between 30 and 60
percent would not be objectionable; consumers
w
ould continue to value the image based on scene
content and emotional involvement.
Additional research continues to confirm that the
use of the 30 percent endpoint criteria is quite
conservative, and that levels of 60 percent or higher
can still be considered to fall within the acceptable
category. Generally, at fade levels above 70 percent,
the psychophysical studies behind this research
found that most prints would fall into the marginally
unacceptable category or worse.
7
Although it w
ould be possible to use a fade limit
greater than 30 percent and report longer print life,
Kodak takes a conservative approach with the 30-
percent upper limit. It is also important to note that
using a fade limit lower than 30 percent would
result in under-predictions of print life. For example,
a fade limit of 15 percent may be appropriate for
images in a museum, but is far too conservative for
typical consumer environments because a print at
this endpoint would only be slightly changed from
the original. Using a 15% endpoint would predict
print life that is only half of that predicted using a
30% criteria and would be very misleading. Many
c
onsumer
s would see little or no chang
e in a single
stimulus situation and calling this an endpoint could
cause needless concern and worry.
4
St
abilit
y of C
olor Phot
ogr
aphic Images—Methods for Measuring,
ANSI IT9.9-1996, and ISO 10977.
5
Stability of Colour Photographic Images—Methods for Measuring, ISO
Publication 18909-
2006
.
6
D. Oldfield and G. Pino, R. Segur, J. Twist, “Assessment of the Current Light-
Fade End-Point Metrics Used in the Determination of Print Life:Part I”,
Journal of Imaging Scienc
e and T
echnology
, 48 (6), 2004, pp
. 495-501.
Approximate
Correlated Visual Impact
Dye Loss
0 to 10% Minimal
10 to 15%
Very slight—noticeable to an expert
only in a paired comparison
15 to 20%
Slight—noticeable in a paired
comparison
Moderate—noticeable in single
20 to 30% stimulus to a person familiar with
original scene quality
30 to 60%
Notic
eable f
ade with a single
stimulus, but not objectionable
Greater than
V
ery noticeable fade with
60%
a single stimulus; possibly
objectionable based on use
5
7
D
. Oldfield and J
. T
wist, “
As
sessment of the Current Light-Fade Endpoint
Metrics Used in the Determination of Print Life:Part II”, Proceedings of
IS&T’s 2004 Conference on Archiving, pp. 36-42
.