January 9, 2012 | 1
This is a guest post by Jim Perkins, a professor at the Rochester Institute of Technology’s medical illustration program. I met Jim through the Guild of Natural Science Illustrators, but before I met him in person, I was a big fan of his posts to the GNSI’s email discussion list. (For those of you not familiar with their discussion list, I highly recommend it. The wealth of knowledge that the diverse group of professionals bring to the table is priceless and the spirit of sharing can save you a lot of time and heartache exploring dead ends.) Though he posts sporadically, when he does, his emails are so informative and clearly written that they usually end up getting printed and pasted up on the wall behind my computer screen. As such, I asked him to contribute a few pieces to Symbiartic. What follows is the first installment of a 2-part post on color calibration and why it’s critical if you are a visual artist working with computers. Enjoy!
Guest post by Jim Perkins
As a professional illustrator, I’m often asked what kind of color printer I use for my work. Frequently, this question comes from other illustrators who are interested in my workflow for creating digital artwork. They assume that I have some sort of high-end color laser or “Giclee” style inkjet in order to proof my color work before sending it off to a client. They’re usually surprised to learn that I only have a cheap inkjet that came free with my last computer purchase. They’re even more surprised to learn that I never use my color printer to proof my work. If it wasn’t for the freebie, I might not own a color printer at all.
I can understand spending a lot of money on a good color printer if you are generating the final output of the artwork. For example, I know many natural science illustrators who make their own prints, postcards and greeting cards to sell at museums, craft fairs, and other venues. But if you’re submitting work to a client who will have the artwork printed professionally – in a textbook, journal, brochure, poster, etc. – there’s no reason to waste your own money on a high-end printer, let alone the outrageous cost of color ink and toner.
What Is “Real” In Digital Art?
So why do artists feel compelled to generate color prints of their artwork, even if the work will ultimately be printed by a publisher on an offset press?
As artists, we’re accustomed to thinking of art as existing in some tangible medium. Those of us over 40 remember a time when artwork was created entirely in traditional media on paper, canvas, or some other substrate. Even the Millennial Generation probably learned to draw with a crayon and paper long before they picked up a Wacom stylus. Therefore, we are conditioned to think of the tangible image – on paper, canvas, etc. – as being the “real” piece of art. Even when art is created digitally, many artists don’t feel it is “real” until it exists in a tangible printed form. Just as some of us prefer to own “real” books, as opposed to a Kindle or Nook, there’s a certain satisfaction that comes from holding a finished piece of art in your hand.
Prior to the desktop publishing revolution of the late 1980s, artwork was created entirely in hard-copy form using traditional media. In order to reproduce that artwork in print, it was necessary to photograph or scan the image to produce film negatives. The film, in turn, was used to create a high quality proof (such as 3M’s Match Print system) and also to create the printing plates used in offset lithography. At every stage of the process – from scanning to proofing to the final press run – the results could be compared to the original piece of artwork to ensure color accuracy.
In this digital age, however, we need to rethink our notion of what is “real”. When artwork is created digitally, it does not exist in a tangible form. The only thing that is “real” in a digital image is a collection of 1’s and 0’s stored electronically on a piece of magnetic media. The image you see on screen while creating the artwork is simply the monitor’s interpretation of the numerical color values stored in the digital file. If you change monitors or move the file to a different computer, the image will look slightly different. To demonstrate this, just look at the wall of televisions in your local big box electronics store. Every screen looks a little different even though they are all displaying the same digital signal.
Similarly, every color printer works just a bit differently. There are several different printing technologies including inkjet, laser, and dye sublimation. Even among inkjets, there are several variations. Some use the same four pigments found in offset lithography – cyan, magenta, yellow, and black (CMYK). Others add a light cyan and light magenta for a total of six colors and there are even eight-color inkjets on the market. Even with high-end devices, you’ll find significant variation in print quality. As with monitors, each color print is simply an interpretation of the “real” numerical data stored in the digital file.
If the image on screen and the printed output are merely interpretations of the digital file, subject to the idiosyncrasies of the specific hardware, how do you know which of these is closest to “reality”? In other words, how can you be sure that you are viewing an accurate interpretation of the numerical color values stored in the file? There is no original piece of artwork to compare it to. If you send the file off to a client, will you be unpleasantly surprised when you see the final output?
Calibration Is The Key
When my students submit prints of their digital artwork, they often complain that their color printer did a lousy job of reproducing the beautiful image that they created on screen. I ask them, how do they know that the printer is to blame? Perhaps their monitor was way off and the print is actually a better representation of the numerical data stored in the digital file. The only way to know for sure is to calibrate one (or both) of the devices to be sure that either the monitor or the printer is giving an accurate interpretation of the “real” image.
In theory, calibrating a printer is pretty simple. You print a digital target image that contains known color values. You then measure the colors in the printed image to see if they match the known values in the digital file. If there are discrepancies, software can generate a printer profile that will adjust the colors to compensate for these differences. (The more accurate term for this process is Printer Profiling, but we’ll stick with calibration for simplicity).
In reality, however, printer calibration is fairly complex and requires knowledge of things like Rendering Intent, Black Point Compensation, Under Color Removal, and other concepts too numerous to discuss here. Suffice to say that it’s not as straightforward as it sounds in theory. It can also be expensive. A spectrocolorimeter, the device for measuring color values in a printed piece, can cost well over a thousand dollars, although prices have dropped dramatically in recent years. Datacolor (www.datacolor.com) now offers the economical SpyderPrint3SR printer profiling system for only $359, including a mini spectro and profiling software. However, this is still more than twice the price of a decent monitor calibration system.
Furthermore, calibrating the printer only lets you see what the “real” image looks like after you’re done creating it. If your monitor is way out of whack, the image might look great on screen; only after you print a copy do you realize how ugly it really is. It’s very difficult (and costly) to create a good-looking image if you have to keep printing copies to see what it really looks like.
In my opinion, monitor calibration is the way to go. This ensures that the image you see on screen is an accurate representation of the color values in your digital file. You can paint with confidence, knowing that the image you see is a true reflection of the “real” digital image.
More importantly, you can be confident that you are sending your client a digital file with the colors looking just the way you want them. Of course, it’s up to the client to ensure that the colors look just as good when they print the file. Since I work with some of the world’s top medical publishers, I know they are using high quality equipment and a reliable workflow. With a carefully calibrated monitor, I no longer get unpleasant surprises when my clients send me copies of the final printed piece.
There are several good monitor calibration devices on the market. I use an older Colorvision (now Datacolor) Spyder2. Datacolor currently offers three different versions of its Spyder3 calibration device. They all use a similar measuring device; the differences are primarily in the software and the number of user-configurable settings. Other options are available from Pantone and X-Rite:
Datacolor Spyder3Express ($89)
Datacolor Spyder3Pro ($169)
Datacolor Spyder3Elite ($229)
Pantone Huey Pro ($99)
Pantone/X-Rite ColorMunki Display ($189)
X-Rite i1Display Pro ($269)
Prices quoted are full retail, direct from the manufacturer’s website. Most models can be found cheaper through online retailers such as Amazon, NewEgg, or CDW.
In my next guest blog, I’ll discuss the calibration process in detail and help you choose which of these systems is right for you.
Regardless of which you choose, every digital artist should invest in a monitor calibration system. It’s a tiny price to pay for knowing that your digital files are properly constructed and the image you see on screen is accurate. It’s a better investment than any color printer and is a far better choice for ensuring the accuracy of your color images. This is why I never bothered to get a good color printer. With proper calibration, I trust what I see on screen and have no need to generate printed color proofs.
Jim Perkins is a Professor in the Medical Illustration program at Rochester Institute of Technology, where he teaches courses in human gross anatomy, scientific visualization, and computer graphics. He is also a practicing illustrator, creating artwork for several best-selling medical textbooks, mostly in the areas of pathology and physiology. For 20 years, he has been the sole illustrator of the Robbins and Cotran series of pathology texts. He is also part of a team of illustrators who carry on the work of the late Dr. Frank H. Netter, considered by many to be the greatest medical artist of the 20th Century. To see examples of Jim’s work, visit the following links:
Give a 1 year subscription as low as $14.99X