Color Management in Windows Vista vs. Mac OS X
I am a cross-platform kind of guy. I have my wonderful black MacBook as well a Dell desktop running Windows Vista Professional. When I teach a class on color-management or other aspects of photography, I always like to know which platform each photographer is using.
In most cases, the photographer’s operating system of choice has to do with experiential history—whichever platform the photographer first started using is what they have continued to use. Sometimes I hear about photographers who have switched from OS X to Windows and vice versa.
One reason I use both operating systems is because I frequently get questions about how Windows integrates color-management differently than Mac OS X. Different versions of Windows handle color management differently, So let’s focus first on Windows Vista.
Storing Profiles in Windows Vista: In Vista, the ICC profiles for your displays, scanners, cameras, and printer/media combinations are stored in the Windows/System32/Spool/Drivers/Color directory. If you haven’t done so yet, you might want to make a shortcut on your desktop to the contents of the directory.
One of my favorite color-management features of Windows is that if you acquire a profile from an outside source (either by downloading one from a website or having one sent to you via e-mail), you can easily copy the profile into the proper directory.
Simply right-click on the profile and select Install Profile. This simple process will copy the profile into the Windows/System32/Spool/Drivers/Color directory where it will be available for ICC-aware applications such as Photoshop and Lightroom. Mac OS X does not have this feature. If you are wondering where profiles reside in OS X they are located in you Library/ColorSync/Profiles directory for either the root or user level.
Display Calibration: When you calibrate and profile your display for color-accurate editing, you should know that one of the trickier things about Windows is that the video card driver must support a software rewrite of the default gamma tables in the video card LUT (Look Up Table).
X-Rite has a tool to test this. Click here to check it out.
If your video card does not support modifiable video LUT's then you may have to upgrade or downgrade your video card driver, or even possibly purchase a new video card. This is common on many older laptop PCs.
Mac OS X works on a totally different premise and ColorSync can use a table in the display profile to load the calibration data to the video card.
Windows Color Systems: One of the biggest changes in Microsoft color management was the introduction of WCS or Windows Color System for Vista. This was introduced during the development of Vista. This initiative was a collaboration with Canon, and uses Kyuanos technology developed by Canon. WCS is very powerful and has a host of totally new features designed to evolve color in the OS.
We’ll talk about some of these improvements in more detail in a separate post, but some of the benefits of the Windows Color System include a completely revamped color infrastructure and translation engine called CITE, an enhanced color-processing pipeline that supports greater bit depths and multiple color channels, and support for scRGB (a wide gamut RGB space developed by Microsoft and HP). Most notably, Windows Color System provides an improved user experience through a centralized color control panel.
Where Vista Still Lags Behind Mac OS X: The biggest issue with color management in Windows Vista is the lack of integration of the display profile into every application.
One of the beauties of Mac OS X is that every application uses the display profile to render color. This means that if an image has an embedded profile, OS X will convert the color data and give you the correct appearance on your screen. Your images will look the same whether you are viewing them in Photoshop, Preview, Safari, and Mail.
This is not true for Windows Vista. In Vista, applications need to be ICC aware to accurately display color images that have been created and saved in color spaces other than sRGB.
This means that images that you have edited and stored in the Adobe RGB or ProPhoto color spaces will look different in Internet Explorer, Outlook, or Microsoft Picture Viewer than when you edited them in Photoshop or Lightroom.
For more information on photography and color at Microsoft visit www.microsoft.com/prophoto and www.microsoft.com/whdc/device/display/color/WCS.mspx
Marc Aguilera is a part-time professional photographer and Photography and Digital Workflow instructor for University of California San Diego Extension. He is also an Apple Certified Professional in Color Management and Apple Certified Pro Trainer. He is a color expert for X-Rite's Color Services division and speaks on behalf of the creative community at AIGA's HOW Conference and at the PIA/GATF Color Conference each year. He has a BA degree in Visual Arts from Univeristy of California-San Diego. You can reach him through his blog at http://www.colorcritical.com.
Bit Depth Basics: More Than a Numbers Game
One of the most under-appreciated features of newer-model DSLRs is their ability to capture images with greater bit depth. Bit depth is a way of quantifying the amount of color information in each recorded pixel. It is also a key component of high-level image quality. In this post, we’ll look first at the numbers. Then, we’ll talk about what these numbers mean in terms of dynamic range, color fidelity, and highlight/shadow detail. These are key attributes that influence the quality of the final print.
Not long ago, most cameras functioned in 8-bit mode. Then the higher-end DSLRs became capable of using 12-bit capture. Now they are up to 14-bit, which is a huge benefit to photographers. If you don’t understand bit depth, these may sound like incremental improvements. But they are actually exponential improvements.
To see what I mean, let’s start with a brief review of the numbers. It is simple but non-intuitive:
· There are three primary colors (or color channels) in each captured digital image: red, green, and blue.
· Bit depth describes the number of tone gradations (or intensities of color) provided in each pixel. Most digital images are captured and/or stored in 8-bit, 12-bit, 14-bit, or 16-bit mode.
§ An 8-bit image has 256 tone gradations in each of the three color channels;
§ A 12-bit image has 4,096 tone gradations in each color channel;
§ A 14-bit image has 16,384 tone gradations in each color channel; and
§ A 16-bit image has 65,536 tone gradation in each color channel.
Another way to look at the depth of an image file: A 16-bit image file is twice as big as an 8-bit file.
The major benefit of working with high-bit images is increased dynamic range— the range of tones and detail that the camera can record from the darkest dark to lightest light.
One software company, DXO, now provides public access to its database on camera performance. Here are some examples of the differences between 12-bit, 14-bit, and 16-bit capture:
· The Nikon D2X captures images in 12-bit mode when shooting in RAW format; its dynamic range is rated at 10.9.
· The Nikon D3 captures in 14-bit mode; its dynamic range is rated at 12.2.
Dynamic range is measured like f/stops: an increase of one step is a doubling of dynamic range. That’s a big, big difference.
There is also a noticeable upside in image editing, and in the appearance of the final print.
Figure 1 above is a screen shot of a 16-bit image being edited. Note that the histogram is smooth and even, and shows no breaks or lines.
Figure 2 above shows a screen shot of an 8-bit version of the same image. Note the white lines running vertically in this histogram. These are sometimes called drop-outs, and they show information that is lost when the image is edited . These losses can result in color distortions, posterization, color aliasing, and more.
We lose information every time we edit or manipulate an image. So, the more image information we have to start with, the more information that is available to us as we progress in our workflow.
It pays to set up your workflow to protect as much image information as you can for as long in the process as possible. Most printer drivers can handle high-bit images without difficulty. In my own workflow, I only use 8-bit images for web publishing, e-mail, and the like.
Image-editing programs such as Adobe Photoshop Lightroom are now set up to use high-bit images. And Adobe Camera RAW and Photoshop provide pretty much the same editing tools for 8-bit images and high-bit images, making workflow choices easier for the photographer. This is all good news for those of us who love creating the most detailed and beautiful prints possible.
What’s re-complicating printing workflows right now is the fact that Adobe Photoshop Lightroom uses a bigger color space—ProPhoto RGB instead of Adobe RGB. But we’ll leave a discussion of printing through Lightroom to a future post.
If you have any specific questions about bit depth, I would welcome your comments.
Choosing the Right Color Space
One area of contention among digital enthusiasts is the choice of color space for doing editing work. Almost every camera gives you the option of sRGB or Adobe RGB at capture. sRGB is a good choice for JPEG capture, especially if you intend to place your images online. It’s a standard that is safe for most situations and displays. But is it appropriate for printing? That depends a bit on your printer and your workflow.Most of the better inkjet photo printers support a color space closer in size to Adobe RGB. Some of the newest printers are capable of printing colors outside of Adobe RGB as well, which makes ProPhoto a consideration.
How much difference is there? Figure 1 below shows both sRGB and Adobe RGB spaces, along with the ProPhoto color space. In this figure, the red area is the sRGB color space, green indicates Adobe RGB, and the outside line is ProPhoto.
Figure 1
It’s pretty easy to see that ProPhoto covers just about every imaginable color, so why shouldn’t you just select this color space and be done with it? The main reason you might not choose ProPhoto is that you can have colors in your image that will be significantly out of gamut for your output device.
Let’s take a look at Figure 2 below. This shows the ProPhoto color space as the large area with the new HP Baryte Satin Art Paper on an HP Designjet Z3200 printer as the inside area. The Baryte paper is an excellent media with a large color gamut, but it’s significantly smaller than ProPhoto. So, if you print to this paper you run the risk of having colors that are out of gamut and will need to be mapped to a new color.
Figure 2
But, if we work in Adobe RGB, you can see in Figure 3 below that Baryte actually has colors that can be reproduced outside of this range. You would end up having colors clipped and remapped that weren’t out of range of your printer in the yellow and cyan spectrum.
Figure 3
If you’re working in Photoshop Lightroom, this is a moot point because Lightroom uses ProPhoto as its working color space. Photoshop however gives you the option of selecting a working space.My recommendation to attendees at my workshops is to use the ProPhoto space if they’re doing their own printing. I’d rather have color in my image that is beyond the paper limits than have colors lost that I could have reproduced.
But, does it matter? After all, as I mentioned at the start of this post, your camera most likely only has sRGB and Adobe RGB as choices. If you’re shooting RAW though, a color space isn’t actually applied until you convert to an image format such as TIFF or PSD. If your image is already in one of these formats, going from Adobe RGB or sRGB to ProPhoto isn’t going to improve your output. It would be like moving a gallon of water from a one-gallon bottle to a five-gallon bottle. You still have a gallon of water.
If you’re not shooting RAW, you’re already giving up quite a bit of image data, so I’d suggest using Adobe RGB to keep as much color information as possible unless you don’t print, or you send your work to an online photo service such as Mpix or Snapfish to be printed. In that case, sRGB will save you a bit of time and give you results that will be as good as you can get from the source files.
Fine-Art Reproduction Part 2: What You Need to Get Started
Professional photographers are uniquely qualified to get into the fine-art-reproduction business because the element most critical to success is a quality image capture. Photographers not only have a discerning eye for color and detail, but also tend to have the best capture equipment. For fine-art reproduction, the lens, digital sensor, and software used to process the RAW digital image files have to be pro level, with no compromises.Specialized Training
Before you invest in any additional equipment, it might be wise to invest first in some training—so you can see for yourself how the fine-art reproduction workflow differs from printing your own photographs. During a good training session, you’ll see why it’s not smart to try to cut corners when it comes to buying quality equipment.
For example, the training should cover specific capture, color control, and printing techniques that have proven to be successful in faithfully reproducing the colors used in the original artwork. This usually requires one to two full days of work, with lessons focused on image capture setup, color management, media selection, image editing and print prep. You also need to understand how the build good relationships with artists and some of the printmaking traditions specific to the art market. Once you’ve been trained, you’ll better understand the rationale behind all the other elements listed below.
The Right Lighting Setup
Digital image capture for fine-art reproduction is based on good old-fashioned copy work. It requires two to four color-corrected lights (strobes or continuous), diffusers or softboxes, a sturdy tripod or studio stand, and a stable copy stand that can support artwork in a range of sizes. You could use continuous lighting using tungsten bulbs, but I don’t recommend it. Even though we can correct color pretty well, a light source that isn’t color-balanced can lead to a number of problems, including excessive time in post-production trying to correct color distortion. It’s better to use color-corrected halogen or fluorescent lights.
A Good Camera and Lens
I use a Hasselblad H-series camera and a Phase One digital back. I really don’t see any reason to use film for fine-art reproduction unless your client wants to archive an image on film for some reason. With my equipment, I have gotten excellent results with images up to 40 x 60 in. in one frame, and panoramas up to 10 ft. long. It is quite feasible to shoot even larger pieces in one frame. Or, you could shoot the image in quarters and stitch them together in Photoshop. The recently released Phase One/Mamiya medium format combination is interesting, and costs less than competing products. A new Mamiya kit with an 80 mm lens has been reported selling for $10,000. You could use a high-end DSLR, like the Canon Mark II and III-series, or the Nikon D3 series on many pieces of artwork, but successfully reproducing a large painting usually requires the resolution that only a medium-format digital camera can provide. In any case, the cost of acquiring technology is lower and dropping every month. Used equipment is frequently an option as photographers trade in and trade up.
Procedures and Tools for Controlling Color
Reproducing artwork requires processes for controlling color at every step in the process, from capture to output. The more careful you are in controlling color, the less time and materials you’ll waste trying to get your print to match the original. If you’re serious about your digital photography, you probably have already invested in a high-quality monitor and tools for keeping it calibrated so you can accurately preview and edit your images on screen.
The most straightforward way to control color while shooting is to:
Ensure that the target artwork is evenly illuminated from corner to corner and from side to side, within 1/10 of an f-stop;
Shoot at lowest ISO available;
Use RAW capture; and
Include a grey card or white/grey/black target in your shot. (This will be an enormous help when processing the image on your computer.)
You can use a handheld spectrophotometer (such as the ones made by X-Rite), or an online service (which can be pricey). Or, you can buy a wide-format inkjet printer that has the custom-profiling functionality built-in (such as the HP Designjet Z3100). Being able to obtain accurate color profiles is essential, but that’s only part of the game. In order to efficiently be able to produce a print that looks just like the print you output six months ago, your profiles must be re-created or updated periodically.
A Pro-Model, Pigment-Ink Printer
You’ll need access to a wide-format, wide-gamut printer that can handle a variety of media types, media thicknesses, and roll widths. Most people in the fine-art reproduction business have a printer that can print up to 44 in. wide. Many printmakers use devices that can print up to 60 and 64-in. wide. These wider printers can be used not only to make larger prints, but also to efficiently print higher volumes of smaller prints. If you don’t yet own a wide-format printer, some studios will rent you access to their printers for a day or half-day. In other areas of the country, you may want to purchase one for yourself. To produce the wide color gamut needed to accurately reproduce fine art, your printer should have at least eight ink channels. In my opinion, having 12 ink channels is better because the color palette and the control provided over color and density is noticeably superior.
It’s also important to be sure that the printer uses pigment inks, instead of dye inks. When used with reputable brands of art papers and canvases, pigment inks can create prints that will last well over 100 years without noticeable fading if they are properly protected and displayed. After you’ve equipped yourself to go into the fine-art reproduction business, the next task is to attract customers. I’ll share a few tips on marketing in my next post.
Improving Print Accuracy
As photographers, we’re always concerned about how our images are reproduced, either on screen or in print. Sure, we learn about color management and how important it is to calibrate our displays and to use the correct printer profiles for output. But, how many of you have actually checked the accuracy of that output? Are you positive that your printer is giving you the best possible print in any given situation?
Most printers come with quality profiles for the paper that is sold by that company. HP is one of the few that I’ve seen that also offers profiles for popular third party papers as well. For the rest of your output needs, you’ll either need to find profiles, hopefully from the paper maker, or from a user group (Yahoo has groups devoted to almost every brand of printer). Or, if you’re the owner of an HP Designjet Z series printer, you can use the built-in spectrophotometer to make your own. The final option is to spend another $500 to $5,000 to buy the hardware and software needed to create your own profiles.
So, you’ve got the correct profile for your printer and paper, you’ve done your edits in Photoshop on your calibrated display. It’s as good as it’s going to get right? Maybe, maybe not.
Anytime I try a new paper, I go to the trouble of printing a test print to verify the quality of the profile for my needs. Many people will use one of their own images, sort of a benchmark, to do this. That’s fine and it gives you a good idea of how the printer does relative to other papers. But I find it useful to use a dedicated test file instead.
The advantage of using a test file is that it stresses all of the critical areas you need to be aware of when printing.
Test-file charts are available from a number of sources, but the two that I’ve found to be the most useful are from Uwe Steinmueller’s Outback Photo site and Scott Martin’s color and black & white charts. You can download these charts for your own use.
The advantage of using a standard chart is in having a known set of values. For example, 
you can evaluate how well your profile and printer produce gray ramps from white to black, color bars of different hue and intensity, as well as common subjects such as sky, skin tones. If you see problems, you can make adjustments prior to printing to get more accurate results.
Sure, it takes a little time, plus some ink and paper, but the overall time and cost savings can add up if the chart helps you find that your printer isn’t reproducing a particular range of colors as well as it could be.
