find_photog find_assist join_asmp

Separation Anxiety

For a print article on pages 20-22 of the ASMP Bulletin Spring 2008 issue, Ethan Salwen examined the logistics of color management and preparing images for reproduction in print. Here, we provide additional content on this important subject.


Lead by Example

Use before-and-after illustrations of an image to improve communication with clients — practical advice from Pulitzer Prize-winning photographer Brian Smith,


Ethan G. Salwen: Do many of your clients request unprocessed raw files?


Brian Smith: I’ve only delivered raw files perhaps four or five times ever, and last year I delivered zero files in raw format. To be honest, I don’t even get asked for raws anymore. This is probably because I generally get hired to deliver a specific look and post-production plays a big role in my style.


EGS: Why do you think you’re getting fewer requests in the first place?


BS: I know photographers who complain they get asked for raws for every job. I think part of the issue is simply that some photographers don’t have their act together. If you’re constantly getting requests for raws, you need to take a close look at the processed files you’re delivering to see if you’re the problem. If you’re not the problem, you can illustrate and communicate your capabilities with simple but powerful “before and after” images that can address every kind of postproduction work.


EGS: What do you suggest photographers should do when they get requests for raws?


BS: When clients ask for raw files, it’s important to ask why. Rather than making a logical argument, many photographers just become belligerent. Rather than saying, “No you can’t,” photographers should phase their response more along the lines of: “You might think you want raw, but here’s why you don’t.”


The last time I was asked for delivery of raw files, I began by asking, “Why?” In that instance it was because the photo editor said most photographers were not making the transition from film to digital very well and they’d been getting lots of bad files. That’s when I sent him this example showing both a zeroed-out raw and my final processed version. That was the end of that discussion.


I use this example (below) from a portrait session with Richard Branson, to illustrate to clients the difference between unprocessed raw files and my final processed files. This is one of my greatest weapons for clear communication with clients about file delivery issues.


© Brian Smith
© Brian Smith


EGS: Why do you think that clients think they want raws?


BS: When clients ask for the raws it’s generally for one of four reasons:


  1. They’re cheap and they figure that they can eliminate digital costs if they just ask for raw files.

  2. They’ve gotten really crappy digital files in the past from other photographers.

  3. They’re gotten JPEG files from a photographer who only shot in the JPEG mode and they want to make certain the photographer shoots in raw.

  4. They may want the freedom to choose the select at the last minute when the photographer might not be available to process it.


My response to the first reason is simply to tell clients that I will still charge a digital production fee regardless of who processes the final images.


With the second and third reasons, the solution is for photographers to know what they’re doing. Unfortunately a lot of photographers don’t, and this can create problems for the whole industry. Just talk to anyone in pre-press and they’ll tell you horror stories about receiving files that were over-sharpened or over-saturated beyond the limits of reproduction.


Addressing the fourth reason, where a client wants to make the final choice last minute, is no different than delivering color prints. I’d rather provide the client with three or four final selects with two days notice than to expect to find out about the final select an hour before deadline.


EGS: So, from your perspective, a photographer needs to take responsibility not only for shooting the images but for image processing as well.


BS: I’d definitely rather have control of the final look. Otherwise I don’t feel that I’ve finished the job I’ve been hired to do. I rarely get asked for raw files these days and when I do I just send them an illustration like the one I did of Branson. This is a slam-dunk for convincing every client every time.


Taking Full Advantage of Soft Proofing

Insights and examples from Baltimore, Maryland-based digital pre-press specialist Grace Zaccardi,


Ethan G. Salwen: Please tell me a little about the professional services you offer.


Grace Zaccardi: My business provides digital pre-press and post production for photographers and for their corporate, design and advertising clients. The majority of the work I do is destined for offset printing, but most of these images will additionally be used for the Web as well as for internal publications.


EGS: So you act as a middleman between photographers and clients, designers and printers?


GZ: Instead of “middleman” maybe a better word would be a bridge. I started primarily as a photographic printer more than thirty years ago, but my primary client has always been the commercial photographer. Most of the photographers I work for spend large amounts of time shooting on location. Since they aren’t always available during business hours, I deal directly with their clients. These clients often have serious time constraints and their questions need to be answered in a timely manner, and they frequently have tight deadlines as well. Because I’ve worked with some photographers for more than 30 years, I’m both intimately knowledgeable with their workflow and quite attuned to their particular photographic style as well. I like to think I provide a valuable service both for the photographer (who is almost always happier shooting than sitting in front of a computer) as well as for their client (who needs the images delivered quickly, yet also may have questions about image formats and workflow). Digital is a moving target, it has sometimes been difficult for the professionals I deal with to stay up to date.


EGS: There is a large debate about whose responsibility it is to handle RGB to CMYK conversions. What are your thoughts about this debate, and who is responsible for what in the RGB to CMYK process?


GZ: Yes, I’ve engaged in more than a few conversations about this issue. Back in my old photographic printing days, I had photographers who handed over shot film; I’d process it, send out the contacts and the designer would deal directly with me in making final selections and determining how the images should look. I used my creative sense to work within the vision of the photographer as well as that of the art director or designer, with very little feedback from either party.


Then, at the other extreme, I had photographers who were intimately involved at every step, giving me lots of input, marking their picks and actually going into the darkroom and either printing themselves or directing the printing. Both extremes were capable of producing great work — or screwing it up. In this regard, not much has changed with digital. If someone likes to have that level of involvement and they are skilled at it, by all means they should do it. That said, I now have few clients who are as skilled as I am in finishing their images. Obviously this is skewed by the fact that it is primarily a certain type of photographer who seeks me out to work with them in the first place.


Regardless of who handles the process — the photographer, a digital prepress specialist like me, or in-house staff at a design firm or offset house — besides the necessary technical skill, this person needs to have the creative ability to fully realize the finished image as well as the power and will to implement it. Working with a RAW camera image is very much like what it was like working from a negative. That RAW image can be widely interpreted; there is no “right” way. These decisions can’t be made entirely with image software. It takes human creativity, someone who can make decisions about how the image should look based on the photographer’s original vision. Besides this, one also has to account for what will happen to that vision once it gets shrunk down to the CMYK color space.


This is where many photographers fall short. They understandably want everything they view on their monitors to appear in the finished piece, and might lack the necessary skill to translate that image into the smaller CMYK color gamut. On the other hand, a pre-press operator in an offset printing house might be intimately familiar with the limitations of the press but woefully unaware of how that photographer prefers images to look. Most importantly, the pre-press operator does not have the autonomy to interpret the image creatively. Clients hire photographers for their vision, and a very specific look is often tied to this vision as well. The client’s investment can get easily lost if that strong image gets interpreted in a milquetoast fashion. The reason people hire me is because I’m not only willing to work with the constraints of the press, but I also have no problem pushing that image to reflect the best of my client’s particular vision. This is my primary skill: to give the photographer a result that is better than what they could achieve on their own, yet still remain true to what they envisioned. Whatever the process, the photographer needs to work with someone who cares as much (if not more) about how that final image looks as they do.


Unfortunately, this means not handing off your “children” to some future unknown person who may or may not know anything about how that picture is supposed to look. Some photographers have succumbed to their client’s demands to hand over RAW images without much thought about what happens after that. I haven’t lasted in this business by allowing clients to make mistakes like that without a concerted effort to make them aware of the consequences. Clients frequently give up quality for convenience and expediency. Hiring a photographer is a big investment and the value of that investment can all be lost in the last step if the person making the conversion isn’t both skilled and creative. That person might be the twenty something guy working pre-press in a design studio, it might be a digital pre-press specialist like myself or it might be the photographer, but it has to be someone with those two abilities, not just whomever happens to have a copy of Photoshop on their office computer.


EGS: Whether they convert themselves or not, many photographers insist that thinking in CMYK can be very advantageous when shooting images destined for offset printing. What are your thoughts about this practice, and do you have any suggestions to offer?


GZ: This is a little bit like what my teachers in art school would call “peeling it off the ground glass”, where the final print was the sum total of the photographer’s interaction with their subject matter. I guess one always has to work within the confines of the medium. The feedback you get from that medium will in turn make you restrict choices, but I can’t help thinking that progress is made by those of us who are always asking, “Why can’t I get this out of that!”? I’m always trying to push the constraints to the limit. I think that’s part of my personality; as soon as someone says, “Never do this” I want to do it. In that same vein, I’m always trying to challenge the CMYK space. I’m not ready to give in to it yet.


EGS: Regardless of whether they think in CMYK while shooting, or whether they make CMYK conversions themselves, many photographers depend on the power of soft proofing in Photoshop. Can you explain what soft proofing is?


GZ: Yes, I depend on this too. Softproofing is a function where you load the printing profile and switch between RGB space and a screen approximation of your CMYK space. Since the RGB space is larger than CMYK, one can always limit the larger color space to show where the clipping will occur. Clipping occurs when the RGB file has information that cannot be reproduced in the destination CMYK color space. Sometimes this is not very troubling, but at other times it seems as if all the fire in the image gets destroyed. This often occurs with high saturation and fluorescent-type colors — colors that can be reproduced in RGB but not with a four-color press.


I say softproofing is an approximation primarily because, while viewing this on screen, it’s very useful to observe where certain colors will not map correctly (hues jump from one hue family to another) and where clipping will occur. This isn’t the same experience as seeing your image in print, primarily because a print is reflective and a monitor is transmissive, similar to the difference between a slide and a print. I softproof on the monitor to make sure I’m staying within the destination color space. I want to avoid the need to move specific color areas back into gamut, which can be very time consuming. Then I usually follow up with paper proofs, but not always. If you follow up with a printed proof, make sure you have a professionally created and accurate color profile for your particular printer/paper combination. The cost for this will be reimbursed in saved paper and ink almost immediately.


EGS: Can you provide some specific strategies and tips for soft proofing?


GZ: First, calibrate your monitor to the latest specs. Make sure you know what kind of paper will it be printed on and load that CMYK profile before you make any lossy change in the image and color. There is no sense in softproofing for a four-color press using glossy coated paper if your designer is going to use a rough textured, warm toned, uncoated stock. If you are working on images that print to uncoated paper with a color base it is important to desaturate the monitor (you do this in color settings) so to account for the lower contrast and the color being soaked up by the surface. In the old darkroom days, we’d have to do prints for newspaper, which is probably the very worst-case scenario even today. We’d take the finished print, turn it around so we were looking through the back. If we could still “read” the image, it was considered ready for the press. Oddly enough, this doesn’t mean you don’t work as hard at getting the image to read, you might just choose a very different contrast.


EGS: One photographer basically told me, “If you have soft proofed, you have essentially made the conversion, so just go one step further and convert the image.” Another photographer explained that he didn’t want to make the final conversion, but he did “soft proof and adjust colors accordingly” to ensure that his RGB files are more “CMYK friendly.” What are your thoughts about these two perspectives? What is your recommendation?


GZ: I agree that if you have softproofed you’ve done the conversion. That said, every image I convert is saved in RGB before I do the final conversion. I frequently adjust plates after a conversion because I’m not happy with, say, the black plate, so I want to be able to get back to the RGB. I routinely deliver both RGB and CMYK because in this day and age publications end up online as well as in print and my client’s clients have this annoying habit of changing the mode on CMYK images and for use on their Web site with disastrous results.


EGS: You’ve provided sample illustrations of adjustments you made to an image using soft proofing techniques. Please explain what we are seeing in each image and how you worked with them?


GZ: The first image [above] shows the “default settings”. I brought in a RAW file using the default settings in Adobe Camera RAW and then converted to CMYK with no soft proofing. Let’s say you deliver RAW files to your less-than-savvy client. This is how the image will look if you are lucky. If you are unlucky, the client will have fooled around and saved the default settings as something bizarre. With the default settings, the image is flat and a bit dull overall, but the auto color balance does approximate the slate gray sky pretty well so it’s a good place to start.


In the next “out of gamut” image I adjusted it on my monitor to have lots of punch and saturation. The red hat really pops off that neutral gray sky and the yellow grass is a nice offset to both, but I’ve raised the saturation to a point where the yellow grass and the detail in the red hat is starting to be lost in the CMYK conversion.


In the third “adjusted” image I soft proofed for the destination space and made sure that the image never got out of gamut. I compromised my need for punch by raising the local contrast instead of raising the saturation on the red hat and yellow grass.


EGS: What is the single most important aspect of soft proofing?


GZ: To never get out of gamut in the first place. Getting back in after you’ve gotten out is a little like translating a poem from French to English and then back to French; you lose a little in each translation.


I think one thing that doesn’t get enough attention is how much the original conversion from RAW to RGB matters in the entire scheme of things. If that original RAW to RGB conversion is out of gamut (in terms of the destination CMYK space) then the most that the person converting from RGB to CMYK can do is to make the best of a bad situation. In my opinion, the RAW to RGB process is critical, far more critical than RGB to CMYK.


It’s important to understand that there are certain colors that cannot be re-created in CMYK; the best you can do is make them look approximately like they do in real life. The most troublesome aspect to converting an image is when colors jump to a different hue. In these cases I frequently adjust the hue separately on the specific colors that can’t be accurately reproduced in CMYK. The original hue and tone might not be possible to achieve, but it’s crucial to shift from a situation where the highlights are one hue and the midtones and shadows another.


The typical problem is a bright yellow color where the highlights turn green. Yuck! You have to switch that highlight back to yellow, even if it has to be a lighter or less saturated yellow for it to look right. It won’t be “right” but it will look right.


EGS: Would you say that soft proofing is underused by photographers in general?


GZ: There are lots of tools at our disposal that are underused. It’s not like photographers don’t care or are careless; its just that post-production is not usually their core strength. Even a full-time post production pre-press specialist like myself has a hard time keeping up with the quantity and the scope of tools at our disposal. It’s overwhelming at times. I think people use tools like they do their brains — they only use a small part of the potential. We engage in a high degree of specialization and this is what makes us a wealthy society. I don’t see this as a flaw. Specialization is a strength, however, the downside is that this can doom you to obsolescence. Plus, every day someone invents a new software tool that performs a function that formerly required great skill to master and a long learning curve; therefore, we all have to carefully pick what skills and tools we want to spend our limited time and energy learning to use.


EGS: Do you have any further thoughts about thinking in CMYK and soft proofing that photographers might benefit from, in relation to the greater issues surrounding CMYK conversions?


GZ: I think that the biggest obstacle isn’t that of skill, it’s mental. Photographers are afraid — afraid to take on the additional liability even though no one knows more than they do about how their image should look. They are afraid of that call, the one where the client calls to say that the piece was printed and it came out awful and you’ll never get another job from them. Offset printers have sort of fed into the fear that unexpected things will happen! I still live in fear of that call, but I’ve learned to push the fear to the back of my mind and dive in anyway because, well, someone has to do it and I doubt there is anyone else out there who cares more than I do about how images look in print. I’ve devoted my life’s work to it.


RGB to CMYK: A Complete Twelve-Step Program

Richard Anderson ( is a commercial photographer and the principal author of the Universal Photographic Digital Standards (UPDIG,


Ethan G. Salwen: You encourage all photographers to practice making CMYK conversions, insisting that the process is much easier than many photographers believe. What are the most important things for photographers to consider when embarking on learning or improving their CMYK conversion techniques?


Richard Anderson: The most important thing is to evaluate whether you have the proper tools. These include Photoshop CS3, a high quality LCD monitor profiled to 5500K, and an inkjet printer with good paper profiles.


EGS: You have provided a concise, 12-step guide for making CMYK conversions. What should photographers keep in mind when following your guidelines?


RA: Most of the battle is being able to create a good, solid RGB rendition of the picture before starting. In general, if your RGB image is high quality and has a good tonal range, the conversion should be reasonably easy.


EGS: Are there any problem areas that photographers typically run into and anything they should keep in mind to avoid these?


RA: Some very typical problems are dealing with pure, bright, saturated colors. These may change fairly dramatically in the conversion to the CMYK color space. In many cases, these colors may be improved, but not completely achieved. So you need to use your knowledge of color to manage both your own and your client’s expectations.


EGS: How would you suggest that photographers approach making conversions when they are unable to work directly with representatives at specific printing houses?


RA: I think it’s important to know that differences between different varieties of CMYK tend to be less than the differences between different varieties of RGB. Because of this, the real danger to your work comes from a printer who strips out your RGB profile, rather than one who ignores your CMYK profile. If you make a conversion to CMYK space optimizing your RGB file to fit the US SWOP coated v2 CMYK profile, you will definitely be in the ballpark with most printers/presses.


Converting RGB Files to CMYK: A Complete Twelve-Step Program

by Richard Anderson,


Step One: Determine the Right CMYK Profile

Determine your target CMYK profile for the offset printer on which your image will be reproduced. If you are unable to determine the specific CMYK profile, the best practice is to use Photoshop’s default CMYK profile, which is U.S. Web Coated (SWOP) v2 profile.


When possible, it’s always a good idea to contact the printer and determine if they require a profile different than the U.S. Web Coated (SWOP) v2 profile. Many of the better printers are switching to the newer standards, namely GRACol profiles for sheetfed presses and the updated SWOP profiles for web presses. A few printers will be using custom CMYK profiles, and if they are accurate and kept up-to date, they can work very well; however, I feel that standardized GRACol and SWOP profiles are a better concept in the long run.


Printers are generally very happy to email their standard or specific CMYK profiles, which can be loaded into the appropriate place on your system OS and makes them available when you choose a CMYK working space or a customized “soft proofing” set-up in Photoshop (View>Proof Setup>Custom>Device to Simulate).


Once you have determined the best CMYK profile to use, go to the Photoshop>Edit>Color Settings, and choose that specific CMYK color space for your “working” CMYK color space.


If the printer suggests using the legacy Photoshop function for Custom CMYK (found in the color settings menu) to create a color separation, and then provides a set of parameters such as dot gain percentage, separation type (GCR/UCR), Black Generation (light, medium, heavy), Black Ink Limit/Total Ink Limit, and UCA amount, then you will know that the shop that does not have a color managed system in place. Under these circumstances my recommendation would be to decline to convert to CMYK. However, it is still possible to make a cross-rendered print of the image (see step nine for a full description) using the U.S. Web Coated (SWOP) v2 profile for reference, and deliver the image in sRGB color space. This will increase the chances for the best conversion after the file leaves your hands.


Step Two: Prepare File for CYMK Conversion

Once over the hurdle of determining which CMYK space to use, the ideal starting point is to work with a 16-bit RGB TIFF or Photoshop file in a wide gamut color space such as Adobe RGB (1998). You will have already finalized making image adjustments, performing retouching and all color and tone corrections. This image file should be at least the same size or larger than the intended final size.


If the file has layers, you’ll need to flatten the file, or make a flattened copy to work with. Make sure that the image has a nice amount of contrast. Conversion to CMYK can often make a photo look a little flat, especially when using one of the uncoated CMYK profiles. One easy, non-destructive technique for appropriately boosting contrast — called “localized contrast enhancement” — is to use Filter>Sharpen>Unsharp Mask with these settings: Amount: 20, Radius: 50, and Threshold: 0.


Step Three: Limit the Output Range

Because offset presses have a reduced tonal range, it’s necessary to limit the output range of the image file. The simplest method to accomplish this is to go to Image>Adjust>Levels>Output Levels, and set the highlight end of the scale to 253. This will keep the highlights from blowing out to paper white — which is absolutely critical. In order to keep the shadow areas from going too deep and blocking up, set the dark end of the scale to 3.


Note: These best-practices numbers will serve almost all CMYK conversion needs. However, whenever possible, it is preferable to discuss the best highlight and shadow parameters with the specific printer who will be handling the job.


Step Four: Save File as RGB Pre-Press Master file

With your file ready for conversion, save it as an “RGB pre-press master file.” This is a version of the file that you can always go back to and convert for any different conversion needs in the future.


Step Five: Re-size Photograph for Specific Use

Based on the size that the image will be reproduced as well as the ppi requested by the printer, resize the image for the intended use. If you do not know what the printer wants, use the industry-standard default of 225–300 ppi (1.5–2.0 x 150 lpi).


Step Six: Sharpen the Photograph

This is the appropriate time to sharpen the image for best output results. Photoshop offers many tools, but I recommend using a third-party sharpening plug-in for both ease of use and best results. I use either Nik Sharp or Pixel Genius PhotoKit Output Sharpener. Both of these mini-programs have specific sharpening settings based on the paper types and linescreens that will be used in printing.


Step Seven: Duplicate for Comparison

Duplicate your image by choosing Image>Duplicate. Then place the two images side by side on your monitor with the copy on the left. This serves as the pre-adjusted RGB version that you will try to match in Step Eight when making adjustments to the CMYK version on the right.


Step Eight: Adjust your RGB Colors for CMYK

With the right-hand (non-duplicated) image file selected, go to View>Proof Setup>Dialog. Choose the desired target CMYK profile in the “Device to Simulate” drop down menu. Choose the rendering intent. I find that, 90 percent of the time, the default “Relative Colorimetric” works best. Next, choose the black point compensation. I don’t recommend using “Simulate Paper” or “Black Ink.”


The process of CMYK conversion involves toggling the preview button on and off so you can preview the changes that will occur when you will actually make your RGB-CMYK conversion. If a color is too far out of gamut for CMYK, you will see a loss of detail. To bring these colors back into gamut use Image>Adjust>Hue/Saturation control. Choose the slider that most represents the problem color and gradually back off the saturation until some detail comes back.


If a color cannot be reproduced in CMYK, it may turn into another color, such as a bright blue sky turning purple. Once again, just as with issues of detail, use the Hue/Saturation control to adjust the affected colors and to restore the look of the RGB original. Sometimes trying “Perceptual” or “Saturation Rendering Intent” controls will help you solve out-of-gamut problems. Increasing the “Master” saturation a few percentage points will often improve the overall appearance of the image when converted to the smaller CMYK color space.


Another effect of out-of-gamut colors is for a bright, saturated color to lose detail, becoming a “blob” of color. When this occurs, it is often effective to use Select>Color Range on the affected color, and then use Image>Adjustments>Hue/Saturation, and reduce the saturation of that color until some detail begins to come back. You’ll have to carefully choose between bringing back some detail, and making the color too desaturated. However, it will be you who controls the artistic compromise. Another technique is to “borrow” detail from a channel by copying that channel and pasting it into the channel with the missing detail. This needs to be done on a layer, so that the opacity of the pasted channel can be adjusted to the point where the color and detail are both maximized.


Step Nine: Make a Cross Rendered Print

Using a profiled printer, make a print using the “Proof” space instead of the “Document” space. Verify that the Document space is the correct CMYK color profile for your project. For “Color Handling” choose “Photoshop Manages Color,” and for “Printer Profile” choose the appropriate paper profile. I have had great success using Epson semi-matte proofing paper, which is available in sheets or rolls and matches many dull-coated white offset papers. Below the Printer Profile, you will find that Rendering Intent is grayed out. This is because whenever you are printing from “Proof” the rendering intent will automatically be Absolute Colorimetric. If you were generating a proof from a file already converted to CMYK, you would choose Absolute Colorimetric as the rendering intent since it will simulate the paper color. Below Rendering Intent, you will see the Proof Setup menu, which will be your working CMYK space, chosen based on your communication with the printer. One final box to check will be the “Simulate Paper Color”. Then hit print.


If the resulting cross-rendered print looks good, you are ready to convert. If the print shows that more tweaking is needed, go back to Step Eight and keep adjusting until you have preserved as closely as possible in CMYK the look of the RGB original.


Step Ten: Convert to CMYK

Convert the original file to CMYK using the Edit>Convert to Profile tool. Normally you should embed the CMYK profile you are working with. However, if you work with a printer on a regular basis and you have both agreed on a specific CMYK profile embedding the profile is optional.


Step Eleven: Addressing File Management

After converting to CMYK, save the file in a specific folder for your CMYK conversions for this job/client. You do not need to save the changes to the RGB master file, as they were specific to optimizing the file for a target CMYK profile. If you do elect to save the optimized RGB, be sure to identify it as a “version” for “X” printer by appending a description to the file name. Saving the unchanged original RGB master file will enable you to easily repurpose the file for other output at a later date.


Step Twelve. Using ReadMe Files and Metadata

When delivering the CMYK files, include a “readme” file the same folder. This should describe the CMYK color space and whether or not the image file has been sharpened for output. It should also provide your contact information and specify the rights usage terms for the images.


Also, make sure your critical metadata is intact. Metadata can be stripped through file manipulation such as pasting images into new canvasses or running them through other image manipulation software.