Many thanks again Neil.
On 11/8/2019 5:29 PM, Neil Woodall wrote:
The critique on the methods they used is that it’s still not correct. The
viewing environment is important. Most importantly is the surround because
while prints don’t have a black nearly as good as a monitor they are typically
displayed with a white border. This causes the eye to adapt to the white of the
border and that causes it to be difficult to see that the black is not really
black. So if you are going to change the contrast of the monitor to match the
print contrast, you must also include a white border to match the white border
that the print will have. I’ve seen some recommend a border of mid level gray,
but that’s not correct as it’s not duplicating the print viewing environment.
ISO 3664:2009 recommends a 1-2cm white border and that’s much better.
The best way to think of this is next time you go to a movie, or think back to
when you’ve gone in the past. When the scene is mostly dark, black is not
really black, but a dark gray (typical Digital cinema is about 2,000:1 to
4,000:1 sequential contrast). When the scene has enough bright areas, black
looks black even though the black actually isn’t as black as before because of
reflections off the audience and walls.
There is also research that suggests that the human visual system doesn’t adapt
to a emissive display the same way as it does to a reflective print. That while
a print can have 100% chromatic adaptation, a display doesn’t so it always
tends to want to be adjusted to a cooler temperature than the ambient light
when the ambient light is less than 6500K. So I would make sure you have a
calibrated system (monitor, printer and paper) and then if you need to compare
an actual print to the display, try to find a setting for the display that
reproduces the tonal balance of the print. Trying to hand correct hue and
saturation between the display and print will be difficult because of the
different way the eye chromatically adapts. By setting, I mean the contrast
ratio (black level) and what type of border to add to the image when proofing.
Then set the print away and turn off the D50 light and go back to ambient light
of D65 and a monitor with D65 white point to make any adjustments to the actual
image file.
On Nov 7, 2019, at 1:57 AM, Yves Gauvreau <gauvreau-yves@xxxxxxxxxxx> wrote:
Neil,
First thank you for the info. I've come across a couple interesting links that
you and others might find interesting and pretty close to this subject.
The first, links you to an eBook on Fine art digital printing from Lightroom
which I find interesting.
The second is about a feature of the I1 Profiler software from X Rite and is
only useful if you have an I1 Pro or above I presume. The idea seem to be about
making your display behave like if it was the paper on which you intend to
print. They seem to do this by matching the contrast ratio of the print in
addition to whatever else is usually done with soft proofing. This approach
seem to be very limited since it can only be use if all the profiles where made
with the proper instrument. Interesting but practical, I'm nor sure?
Maybe there is a better solution for something similar and one that can be done
using Argyllcms tools?
https://fineartprintebook.com/case-studies/2019-10-07-x-rite-i1-studio-review-part-2.md-copy/
(follow the link "get it now")
https://www.youtube.com/watch?v=1vqzft1nHnc&feature=youtu.be
Yves
On 11/1/2019 1:04 PM, Neil Woodall wrote:
I’ve just been reading a lot of research in this area to come up with specific
recommendations on how you change the display to match the viewing environment
to minimize the differences. Most of these require OS support and specific
hardware. But if you can control the application (or the web page) and suggest
the correct viewing environment you can do a few things right now to fix the
issue.
View the display in a dark room and make sure the image on the display has a
little white and mostly gray around the representation of the print. The amount
should simulate the desired viewing environment for the standard. This is
usually around 10% of the peak white, maybe a little more because it’s not
going to cover the full field of view.
By doing the above you do two things:
1) The human visual system will chromatically adapt to the white point of
the display. That is, gray in the image will look gray to the user.
2) The gray scale will look correct because the background luminance level
is well controlled and simulates the standard viewing environment.
The biggest mistakes are:
when you view a wide gamut display in a reasonable ambient light viewing
environment. When this happens, individuals will actually see different colors
of gray because the ambient light is broad spectrum and the display has a
narrow spectrum (called user metamerism). If the ambient light isn’t D65, then
everyone will see the wrong shades of gray.
You view a display that is very bright in a dark environment, think viewing at
night with the lights off. You are now far away from the standard’s assumed
viewing environment. When this happens, the gamma of a calibrated display will
be too low for that condition and you will see more details in the shadows on
the display than you will on the print. The opposite happens when the display
is too dim for the ambient light level.
Neil
On Nov 1, 2019, at 3:57 AM, Yves Gauvreau <gauvreau-yves@xxxxxxxxxxx> wrote:
In the following context, mister X sees an image on a web site that he likes
and decide to buy a print version of it. When the print arrives, he is
disappointed because the print doesn't look the same as on his screen.
I'm sure this happens all the time and even if we have a "perfect" match or
almost where the image was printed, we have no way of knowing what will be the viewing
condition of everyone that could buy a print. Beside giving Mister X is money back this
time, is there something one can do to reduce the probability of this kind of situation
from happening?
Yves
