Re: Calibrationism vs. By the Numbers
- From: David Azose <drazose@xxxxxxxxxxx>
- Date: Wed, 28 May 2008 11:29:04 -0700
Mike Russell wrote:
I seldom start threads in this group, because my interest is in helping
people with answers, and in my own learning when I run into a question that
I can't answer, and need to do a little research.
Recently there has been little discussion of "accurate colors", followed
by the usual recommendation that the person asking the question purchase a
screen calibration device. I think a few words here might save people some
money, or cause their money to be spent on a new display, lens, or other
equipment.
Color is an enormously complex field of human endeavor, and the only thing
that is certain is that we don't know everything about it yet. Photography,
particularly with the new advances in digital photography and profiling, is
in an explosive stage of complexity, and there are a large number of things
that you need to know to really navigate what happens to an image on its
way from the original subject to your screen or printer.
By the time colors reach your screen, there have been through a pipeline
consisting of several stages of capture and processing. Each stage has its
own flaws and idiosyncrasies. For example, the lighting, reflections
between objects and within the camera, the optical and detector system, and
potential processing in the camera or the raw processor each take their
toll on the accuracy of the final captured image. What matters is not
accurate colors, but colors that look good, and an image that brings out
the important colors and textures of the subject.
There are two philosophies for dealing with how to get good colors. One is
called "calibrationism", and is based on the notion that any bad colors,
such as poor skin tones or color casts, means that there is an error
somewhere in the pipeline, and fixing the colors is a matter of chasing
down where that error is. When confronted with poor colors, a
calibrationist will never ask about a specific color or object in the
image, but instead about which components were calibrated, and recommend
purchasing additional hardware or custom profiles. Final tuning, for a
calibrationist, is always a purely visual affair that relies 100 percent on
an accurately calibrated monitor. I have several dear friends who are
completely sold on this notion. I think it is a will o' the wisp, and many
millions of dollars have been wasted on calibration gadgets.
I myself am a big fan of calibrationism's toys and gadgets. I have always
liked histograms, ever since seeing one on a NASA photograph many years
ago. I also have a soft spot in my heart for measurement gadgets. I use
a screen calibration device, and find that it works quite well. I do not
hesitate to alter the monitor settings afterward, if I find that things are
not matching perfectly. I also have a spectrophotometer for measuring
prints and colored objects. It is not used for creating profiles, but for
experimentation and verification when I want to make sure something is
working correctly, figure out what is unusual about a printout or a colored
object such as a flower, or to compare colors in a quantitative way.
Although they are convenient, and even necessary in a production
environment, none of these gadgets is necessary to create excellent images,
as witnessed by the fact that many good images were being created, easily,
in Photoshop before calibration devices were available on the market.
The other philosophy was created by Dan Margulis and is called "by the
numbers". He acknowledges that calibration is important, but need not rely
on instruments. Each image can benefit from a certain amount of individual
treatment. This is done methodically, using tools and knowledge to home in
the colors of specific objects, such as grays or skin tones, and to assign
color and contrast to important objects in the image. "By the numbers"
says that it doesn't really matter how the colors got there, fix them, and
you'll have a good image. Learn to fix them well, and quickly, and you're
on the road to having better and better images for the rest of your life.
With "by the numbers", it is possible, though masochistic, to get good
colors using only a black and white monitor.
Naturally, others are welcome to chime in with their own thoughts and
ideas. I'm not prone to arguing, once I've said my say, but I will be
happy to answer questions, or to otherwise clarify what I have said here.
I agree generally with what you've written. However I'd like to add one more thought: color management is not the same as color correction. The aim of color correction is to achieve pleasing color, while the aim of color management is to achieve PREDICTABLE color, often between 2 or more computer systems.
So if you are making prints by yourself, for yourself and you are the only person that has to be pleased by the prints, then "by the numbers" without any fancy instrumentation is an acceptable way to go. However, if you send files to a lab to be printed, or for that matter to another computer and computer operator to be edited/printed, then color management is needed, the first step of that management being calibrating each monitor to a known standard.
David A.
.
- Prev by Date: Re: Thumbnail different from the image
- Next by Date: Re: Need some help on this
- Previous by thread: Re: Calibrationism vs. By the Numbers
- Next by thread: Kan dit wees?
- Index(es):
Relevant Pages
|
|
