The picture looks "washed out" because the printer has neither the color gamut or contrast which exists on an uncompensated (for printing!) monitor.
BP is needed for Relative Colormetric, and not for Perceptual; it does no damage to leave it on when viewing Perceptual.
To get around the shock of seeing your picture go to hell when using soft proof, engage it and walk away for a while, or at least avert your eyes for a moment. Don’t compare.
I find soft proofs to be simply an indication. For instance, some show a blue shift in the paper color yet the paper is neutral to warm. So, if the image changes rather badly, then make a copy and tweak the image while in soft proof to bring it up. But be sure to make a test print first. It will be in between the uncomped screen and the soft proof.
A caveat: This advice is directed to inkjet printers. If you are planning an offset press run, consult with a knowledgeable person who does this regularly, or find a book and read it. Blatner’s "Real World Photoshop" is good for starters.
Your monitor is calibrated and the working space properly defined, isn’t it? 😉
Lawrence,
Thanks for the info.
Yes, I am very well calibrated using X-Rite colorimeter and ColorEyes Display One software, plus I had a custom profile made for my Epson 4000 for the paper that I use. My monitor is very, very close to my prints when viewed under 5000K lighting. I’m amazed how close I can get. Sometimes I see a very small difference, but it can be quickly adjusted with curves or selective color or the hue portion of hue/saturation.
So, Lawrence, what you’re saying is that for viewing the image and soft proofing on the monitor, using BP is correct, right? And, of course, use it for printing is a given.
And, if the soft proofed image does appear a bit more washed out on the monitor, then it wouldn’t hurt to make the proper adjustment in order to get richer blacks. Then print to see where I’m at. Right?
I will eventually be doing this for offset. Generally, there are some problems using an inkjet to proof the process because there is no roseate pattern. However, this doesn’t usually present a problem for me because I print using "stochastic" which doesn’t produce a fixed dot pattern–similar to an inkjet. But, I’d have to buy a RIP and somehow get SWOP Certified, from what I understand.
The other issue is that printers aren’t always that knowledgeable about their profiles and proofing. My goal would be to get things very close with my proofs and not have to pay the hundreds or thousands for them to proof it. My problem is that some printers don’t even calibrate their monitors, they eyeball it! So, you end up paying for proofs because they aren’t properly color managed. In fact, they seem to make more money by not being so precise because suckers will pay for hours of color correction and proofing. But, in the next few months I will be moving into this area.
Mike
You might want to set up a layer set/group for your adjustments for a given printer, so you can then turn them off as a unit when you switch to a different profile for soft proofing.
BP is necessary for Relative Color, but not for Perceptual, and it does no damage to leave it on in Perceptual.
Yes, if the print in soft proof looks plain wrong, go ahead and tweak it. My own method these days is to find Dmax for the blacks in the Info palette and be sure it’s where i want it, trading off with maintaining shadow detail in the actual print. Sot proof is damn good in that part of proofing. So many times, I have seen detail simply disappear when invoking soft proof. It the color is not to my liking by a minor degree, I go ahead and print anyway. The proof of the pudding….
So far as offset is concerned, since i do not have any certification or experience there, when I am asked for a photo, I provide an RGB file at a 2:1 ratio, resolution to dot, and a print from the inkjet which meets my standards. That’s their source, proof and subsequent problem!
I have a Canon i9900 on my desk, and an Epson 9600 available. The difference between them (and between papers on each of them) is almost staggering. Not that the color is off so that yellow is now greenish, but that the spectral ends are so different. With the i9900, Canon supplies two ways to work. You can let the printer pick the profile when you select the paper, or you can use their custom profiles. I was shocked by the difference that I see when switching from the custom profile for their Matte and the print obtained by allowing the printer Matte control. Neither output met my expectations. The difference between Perceptual and Relative Colormetric is astounding, which can be easily seen in the soft proof for that profile. Actually, I rather like it as I can jump start changes that I may need to make in soft proof. Sometimes the gutsiness of Perceptual is what i want, sometimes the Rel. Color pastel qualities.
I wish the differences between printers didn’t exist, as the Canon clearly has not the print life of the Epson, but yet the colors and punch of the Canon are quite marvelous, especially deep blacks.
Sigh!
"Generally, there are some problems using an inkjet to proof the process because there is no roseate pattern. However, this doesn’t usually present a problem for me because I print using "stochastic" which doesn’t produce a fixed dot pattern–similar to an inkjet. But, I’d have to buy a RIP and somehow get SWOP Certified, from what I understand. "
Not having a dot pattern in inkjet proofs is almost never a problem. The only time it really matters is if you have image content that might cause a moiré on press, and most of the time, you can easily recognize those times and take preemptive action. As far as needing a RIP to have some sort of certified proof, it just isn’t necessary. With the right custom profiles for both your press output and your inkjet, you can produce very accurate inkjet proofs directly from Photoshop. A RIP just automates the process, but it’s very simple to do it yourself. And since most presses do not run to SWOP, or anythng near it, you want to target a specific press or at least that printers’ proofing system.
"My problem is that some printers don’t even calibrate their monitors, they eyeball it! So, you end up paying for proofs because they aren’t properly color managed. In fact, they seem to make more money by not being so precise because suckers will pay for hours of color correction and proofing. But, in the next few months I will be moving into this area. "
Doesn’t matter if the printer calibrates a monitor or not. All they need to do is output your CMYK numbers, and if you’ve run a custom profile of their proofing system, what do you care what it looks like on their systems. I do this all the time, and the two most common responces are something like "that’ll never work, it’s a waste of time" to "how did you nail the color on the first round of proofs?" Most of the time, my Epson proofs are closer to the actual output than their proofing systems.
my Epson proofs are closer to the actual output than their proofing systems.
Right, and since the printers in town are plenty, I quit trying to do the cmyk. The process chages so fast that last year’s custom profile is this year’s garbage.
I never bothered with seps in the old, analog days and I won’t now.
Lawrence,
Does that mean you’re leaving it up to the printer to do it for you? If you are, that’s really scary. Most of them that I deal with are not color managed and have no idea how to get a good proof without going through three or four rounds, which not only is expensive, it take a lot of time.
That’s right, Peter, it was the moire’ pattern that the printer was concerning me about. Exactly! That’s good to know that it shouldn’t be a problem. Thanks. They seem to make a mountain out of a mole hill.
My goal was to not have to even get a single proof from the printer. Is this possible? And the "custom profile" that you refer to, do you actually have a profile made of their system or are they providing it for you?
When you do your own proof using their profile, you then use Photoshop to perform a print preview to simulate the press on your printer?
Lawrence,
I agree with Peter about letting the printer do the work because most printers are not color managed. The problem with going this route, that I’ve found, is that if you want CRITICAL COLOR, then you’re going to pay for a million rounds of proofs because they are just guessing all the time–AND CHARGING YOU FOR IT!!!
The last time I went to press with my line of art cards, I found a printer who would charge me a flat fee no matter how many proofs they made. They did a good job, too. However, I could have done it better.
I provided them with prints from an ancient Epson Printer along with CMYK files. If I would have allowed them to convert to CMYK, it might have been much, much worse because the color conversion whacked much of the colors that I wanted. So, I took it upon myself to do the conversion myself and to correct the colors to get the most out of the image.
There is a huge difference between "pleasing color" and "critical color." Anyone can do the former.
This is why, next time, I will be doing everything possible to provide the printer with real proofs and final files. I will save a lot of money this way and I’ll have complete control of the process.
Depending on your level of experience, it becomes pretty easy to spot where moiré is likely to be a problem. Repetitive patterns in fabrics, building, even in a sole of a shoe, images with television screens or computer monitors, etc. After printing thousands of images, you get an innate sense of what level pre-emptive action to take. Usually it’s some selective blurring and adding of noise, but other times it’s sizing the images and placing them at 50 percent in the layout program. This effectively makes the image double the normal resolution, which can be very effective, especially in conjunction with the former suggestions.
I make my own custom profiles. I’m not sure how many places I’ve profiled in Southern California, but out of the fifty plus, I think there have only been two that had their own profiles, and even those didn’t have the ink limits or black generation I wanted. I find it much easier to send them a target, read it and then make the profile that is right for the types of images and paper I’m printing on.
"When you do your own proof using their profile, you then use Photoshop to perform a print preview to simulate the press on your printer? "
You convert to CMYK using the custom profile, and then convert back to the Epson RGB profile using either Relative Colorimetric, if the printing paper is close to the Epson in whiteness, or using Absolute Colorimetric if there is a substantial difference in paper whites. Since we humans are so adept at white point adaptation, the Relative Colorimetric often produces a more pleasing if slightly less accurate proof, because it retains the paper white of the Epson paper rather than attempting to simulate the press stock.
Well, if the printer doesn’t know how to make or get made the seps, then why would I? I don’t run the press.
Actually, the majority of my clients include my work with others, like a current annual report which includes b&w, color, reproductions of paintings drawings etc. So, I’m not about to get into that loop.
If I did go with process color on an offset, I still would let the work commence with my output as rgb. Can I do cmyk? Yes, and still the printer/designer team do their own thing.
The high end print shops in Portland know what they are doing. My only complaint is that they actually thought that because I didn’t use a Mac, my files would be useless! Now, it’s a pdf and I’m done with it.
If you all know the process, great. How long, Peter, did it take to get the experience recognizing just the moire pattern, and the proper screen angle to minimize it?
Some people insist on dealing with weenies. I deal with professionals. That’s part of my "job description".:-)
Off to do some Epson 9600 prints. 🙂 🙂
Peter,
How do you go about having the printing company print out the target for you. Doesn’t that require a press run?
Mike
"How do you go about having the printing company print out the target for you. Doesn’t that require a press run? "
Typically, you would profile their proofing system, assuming it is well calibrated, doesn’t drift, and reflects their actual printing conditions. Some proofing systems are much better to profile than others. The best of the digital proofers that I have experience with are the Kodak Approval, the DuPont Waterproof, the Imation Digital Matchprint. I’m not too crazy about profiling on top of some of the inkjet proofing systems that use an Epson to output. In those cases, you end up profiling on top of another profile, as invariably, the printing house doesn’t know how or where to access the profiles that their RIP is using for the Epson.
I am about to run custom profiles for one printer here in LA that we use a lot. I’ve profiled their Oris proofing system, and while the results are good, I can see deficiencies in the profiles that I think will clear up by profiling an optimized press run. It should cost about $500 for the press run, but it should also improve our output on difficult images. Profiling the press itself will require running several targets across the width of the paper and averaging, after running the press to their target densities. Should be interesting.
Big assumptions, Peter.
When you say "averaging", is this being done with a spectrophotometer or by eye?
Even with inkjets, variations creep in. This last run today with the Epson 9600 vs the Canon i9900 is an eye opener. Previously, the match between the 9600 and the older Canon s9000 on the same paper (Epson Enhanced Matte) was pretty good. Now, they are miles apart, while the Canon, on either Epson Enhanced Matte or the Canon Matte now provides a better match to the Epson on Ilford Pearl.
Go figure!
So, I have to find a paper/ink combination that will print well on the Epson and match the i9900. This canon makes gonzo prints, and it’s pretty hard to come up with that look on an Epson running pigments, while the Canon runs dyes.
I may have to eat my words about soft proofs. The Canon is actually doing really well.
"Big assumptions, Peter."
Not really. Why do you think so?
"When you say "averaging", is this being done with a spectrophotometer or by eye? "
You run several targets across the press sheet, measure each of them on the Spectrolino, then average those readings using Gretag’s Measure Tool. It’s all pretty easy.
"Even with inkjets, variations creep in. This last run today with the Epson 9600 vs the Canon i9900 is an eye opener. Previously, the match between the 9600 and the older Canon s9000 on the same paper (Epson Enhanced Matte) was pretty good. Now, they are miles apart, while the Canon, on either Epson Enhanced Matte or the Canon Matte now provides a better match to the Epson on Ilford Pearl. "
Were you using current custom profiles for both printers and papers? It’s been my experience that the Epsons are extremely repeatable over a fairly long time. I find I need to reprofile maybe every 9 months or so. Sometimes there are slight variations in ink batches, but paper batches can vary even more. If you notice there is some sort of color anomaloy, it’s time to make a new profile. You’re not using canned Epson profiles with the Canon or Ilford paper are you?
