Nature Photographers Network
This is the preserved archive of the original Nature Photographers Network. The community continues at naturephotographers.network.

Printing Issues

Discussion

18 replies
Posted by Nate Palmer on July 8, 2008 at 1:00 PM
I have a question regarding printing. The current place that I get my pics printed (I won't say the name, because I don't really want to give them a bad reputation because they provide great service) typically prints my pics darker than they appear on the screen, thus desaturating my pics. They say this is due to the Fuji paper that they use. It doesn't appear to be an issue of color calibration, because it looks good on multiple screens, not just at home. Therefore, I need to edit my pics for both the screen (minor sharpening, saturation, etc.) and then do another adjustment for printing (significant changes to brightness and saturation that makes the screen version look fake, but the printed version is accurate to the original pic... if that makes sense). Is this just part of the game or is there a place/printing process/archival paper that doesn't neccesitate this extra work. Just trying to streamline things. Thanks in advance for your help.

Nate
Posted by Todd Caudle on July 8, 2008 at 1:23 PM
One of the reasons to calibrate your monitor is to calibrate it to an industry standard. It seems to me that if your image looks right on your monitor and on monitors at the print lab, they should come out of any properly calibrated machine looking right. Otherwise you're forced into a guessing game that is machine-specific and nearly impossible to figure out. I think the problem is on their end and you should try another print lab that doesn't use ridiculous excuses like "it's the Fuji paper's fault." It's not the Fuji paper, it's their machine, or at least the way it's set up.
Posted by Todd Caudle on July 8, 2008 at 1:24 PM
Let me offer one other possible solution. See if the lab has a printer profile for that specific printer. If you tag your image with that ICC profile, it should come out looking just like it does on your monitor.
Posted by Mel Ladewig on July 8, 2008 at 1:33 PM
I have another thing to consider.

Are you doing your editing work in Adobe RGB or Pro Photo RGB?

Does your printer accept only sRGB and not convert your image from ARGB to sRGB? This doesn't fix the brightness issue, but greatly affects saturation.

But generally speaking, I agree with Todd.
Posted by Jack Brauer on July 8, 2008 at 1:36 PM
My advice: calibrate your monitor for sure. Then, try out inkjet prints from an Epson 9800 or similar - the colors are always spot on and brilliant. You'll also then have the option of using a variety of fine art papers like Silver Rag or Hahnemuhle Photo Rag.
Posted by Les Barstow on July 8, 2008 at 3:18 PM
I've had great results at a lab using Fuji Crystal Archive in the past. As Todd already noted, if you're color-calibrated on the monitor, make sure you use that accuracy by getting the printer's color profile, and saving your files exactly the way the lab wants them. Saving AdobeRGB and putting it on a printer that expects sRGB with no profile will do Bad Things to your print.

Like almost every output medium available, traditional photographic prints cannot fully replicate the range of colors you can capture from your camera, or produce in Photoshop. Having a fully color-managed workflow (capture, edit, proof, and print) is vital to understanding what you'll get from the printer.
Posted by Dave Rodenbaugh on July 8, 2008 at 3:19 PM
Calibrated monitor is only one piece of the puzzle...Sounds like you didn't proof during your workflow on the target printer. If you're referring to one of the Colorado labs, both have ICC profiles available on their websites and you absolutely need to check your print on that profile ("soft proofing") before you burn the TIFF to a CD and take it to the lab.

Dry Creek Photo has a great description of the full workflow on their website. Google them and you'll find it.
Posted by Nate Palmer on July 8, 2008 at 3:23 PM
Todd,

They did give me the ICC for the paper, but that still leaves me with 2 final versions... one for screen (website) and one for paper. If I just did a version for print (by calibrating the monitor to the ICC), it would look overly saturated on the website on other computers. Who do you use in Denver to print your photos (or do you do it yourself)? Jack, sounds like you do it yourself? If not, same question applies. Mel, I will have to look into the sRGB vs aRGB question as I don't know that off the top of my head. I thought taking good pics was the crux, but it appears it might be the processing. Thanks.

Nate
Posted by Nate Palmer on July 8, 2008 at 3:28 PM
Dave, you are right. The first time I did not soft proof, but now I am and the results on print are good. I'm just trying to find a way to not have to do a "screen version" and a "print version". Maybe I just have to do both as it sounds like the soft proofing process is just a part of the workflow. Not trying to be lazy, just would rather spend time shooting pics than clicking the mouse.
Posted by Nat Coalson on July 8, 2008 at 3:33 PM
As the others pointed out a calibrated display is essential for getting predictable color. I recommend the X-Rite Eye One (i1).

And as Dave mentioned, to get the most accurate results also requires soft-proofing: simulating on-screen what the print will look like and making any necessary adjustments... this takes some practice and currently can only be done in the full versions of Photoshop.

It's normal to produce different versions of a master file with adjustments for different types of output.

When having LightJet-style prints made (Fuji Crystal Archive etc.) it's always best to send the lab files in the sRGB color space.

You can use ProPhoto or Adobe98 for your master files then convert to sRGB before sending to the lab.

Finally, as Les pointed out, you really need to work in a end-to-end, color managed environment. But if the lab doesn't have their color management locked down, you're still out of luck...
Posted by Nate Palmer on July 8, 2008 at 4:13 PM
Nat, sounds like different versions for the final output is normal so I will keep making multiple versions.

All, thanks for the input. I always learn a lot when you guys ring in on my questions.

Cheers,
Nate
Posted by Dave Rodenbaugh on July 8, 2008 at 5:36 PM
I use Reed and PhotoCraft both, Nate. Both have their color workflow locked down...what I see on the screen is what I get on a print (after soft-proofing). Like Nathaniel said, you have a master file and then a lab-specific ICC-converted file for output at their lab, and you often need to tweak things a bit to get what you want. Standard operation procedure, I'm afraid. :biggrin:
Posted by Jon LaBass on July 8, 2008 at 6:10 PM
Nate,

I've had pretty good success with processing in the ProPhoto RGB color space and sending that to the lab (McKenna Pro). However, for web images, I convert the color space to sRGB.
Posted by Les Barstow on July 8, 2008 at 6:27 PM
Nate, I used Photocraft in the past, and when I started they required me to send a file converted to their Lightjet Fuji ICC profile. Then they upgraded their imager and removed the pre-conversion requirements - I believe they will now accept any file with an attached ICC profile (or sRGB without an attached profile), provided the size is properly set and the resolution is within their parameters. It's up to you to soft-proof it to their ICC profile to ensure color accuracy before sending it.

The long and short of it is, though, that even with my own printer I save a separate file from my master file. If I ever need to send the image elsewhere, I don't want to send my Epson R1800 file with the blues desaturated slightly, the yellows punched, and the shadows opened a bit when needed to match the printer's gamut. For another printer those adjustments might be totally unnecessary, and the Epson's strong points might be overblown...
Posted by Kalin Wilson on July 8, 2008 at 6:48 PM
Nate,
You've gotten good feedback so far, so I'll try not to duplicate it. But one of your comments struck me as indicating a misunderstanding on your part
They did give me the ICC for the paper, but that still leaves me with 2 final versions... one for screen (website) and one for paper. If I just did a version for print (by calibrating the monitor to the ICC), it would look overly saturated on the website on other computers.

You don't calibrate your monitor to a printer profile. Or put another way, don't assign the printer profile to your monitor in your OS's color management settings.
You should profile your monitor to a set of viewing standards (wp:6500K, lum:110) and then do you editing in a color managed app like Photoshop. Because different output mediums (printer, web, etc) have different capabilities and different interpretations of color values, you will always have to generate different versions for each. Printer versions will vary by paper and ink lots. Your printer may update their profiles when they swap inks. Softproofing against the output profile is key and you will have to make adjustments based on the softproof to get the output you want.

Here is how I manage this:
I work on an image in Photoshop and create a develop master. This has all of the layers, masks, etc that makes up the final image. This can become a large file with lots of layers. Once I have this master image the way I want it based on my color managed and profiled monitor I save it with a 'DMSTR' in the name.
Next I make a dup of that image in PS and flatten the dup, then I save this file with 'PMSTR' in the name. This file is visually the same as the develop master but much smaller and easier to work with. This is my Print master.
With the print master file, I soft proof against a target printer and paper and use curve adjustment layers or whatever else is needed to get the soft proofed image to match my original as closely as possible. One good technique for this is to dup the print master and display both images tiled in PS with the original not in soft proof mode. That way you can evaluate changes side by side without having to turn proofing on and off. Once I have the proofed version looking like the original, I save the adjustments in a layer group named for the profile used to proof. You can do this for multiple outputs and keep the adjustments in groups that you turn on and off as needed.
So when I print, I select the adjustment group for the target output, do a quick proof check, dup the image (for protection of the original), resize to output size, sharpen for printing, and print.

This has served me pretty well. If my develop master changes, I can usually make a new print master and drag the output adjustment groups to the new print master. The only time this doesn't work is if I have layer masks or other adjustments tied to visual elements that may have changed, but recreating the adjustment layers is usually pretty easy.

Still management of all of this can be involved. Good luck!
Posted by John Mumaw on July 8, 2008 at 6:58 PM
Kalin caught the same sentence that caught my eye.
How are you calibrating your monitor? I've never heard of 'calibrating to a profile', so this may be root of the problem.
Posted by Grant Collier on July 9, 2008 at 2:18 PM
I'll just add one thing...if you don't want to spend much time making alterations to your images to prepare them for print vs. web, you might try a lab that uses a Chromira printer, such as whcc.com or the Slideprinter. As long as you have your monitor calibrated, the images will usually come out pretty close to what you see on your monitor, even w/out soft-proofing. That being said, you should still always soft-proof.
Posted by Nate Palmer on July 9, 2008 at 3:36 PM
Kalin,

Thanks for the clarification. In reading the responses and doing some research online, I now understand the difference and my previous misunderstanding. All, thanks a ton for the info!

Nate
Posted by Ken Wyatt on July 11, 2008 at 10:45 AM
Simple solution, let Nat Coalson do the printing for you...

A satisfied customer, Ken