Matching the proof

Ok, first off I apologize for asking this as I'm sure it's probably in a thread somewhere that I just have not been able to find. BUT -- here's the question.

I am looking to print some shots using EZPRINTS. I have DL'd their ICC so I can see the proof colorization based on their color space. I know I have to calibrate my monitor to get the colors to physically match the print; but my question is:

Is there a single photoshop action that can take a standard sRGB processed image and bump it's color space and brightness so that when you print that image it very closely resembles what you had developed in sRGB. I attempted to manually manipulate colors etc in both PS and LR and failed at coming up with a good match. The idea is so that I can process for web displays save it as a 72PPI JPG as well as save it as a 300DPI TIFF for printing, then run a PS action on the TIFF's so they are closer to matching the web versions when I print. I know that I will never get a 1:1 exact match because of reflected light vs the colors being lit. BUT - I know I can get closer than what I did.

In case I poorly explained the flow here's what I'm trying to do.

A: Take pic, load it to LR.
B: Edit Pic until it looks how I want.
C: Export a TIFF at 300PPI in one folder & Export a JPEG at 72PPI (smaller image as well) into a second folder.
D: I plan to print the tiffs, so open PS and run a batch that applies an action that changes those files so that when they print they MORE closely resemble the WEB versions - aka bumping up brightness, saturation, etc. etc. so that they look best on PAPER not on a monitor. I plan on exclusively printing with EZPRINT right now, so I would do this to every shot I plan to print. (Obviously only the TIFF's for printing)

IF for some reason I don't want to use EZPRINT, then I would leave it as is with no action ran and manually match the proof as best I can on a case by case basis.


Can anyone point me in the right direction? Know where there are actions pre-written that do this? Or are there too many other variables that I am unaware of like perhaps depending on the primary color cast of an image the change from one color space to another is dynamically different? I'm not sure here, just trying to find the best way to quickly make my prints more closely resemble my web versions.

Thanks for any incite!

I think this is the primary problem here - not the overall color cast, but the individual colors that will translate into printer colors differently. One image might be a portrait with colors that are all within sRGB and will need no compression in order to be rendered in the printer space while another image will have highly saturated colors (flowers, for instance) that will be greatly changed in the conversion to the printer space. Just how far out of gamut the OOG colors are is also a factor, colors way out will change differently than colors that are close to the border. And the renderering intent, Perceptual or Rel. Col, can also change the print. For some images you may want one RI and for other images the other RI may work better. If you are going to the bother of soft proofing, you will need to give each image individual attention.

Well, that may very well be the case. It seemed to me that since the ICC is a static file that the adjustments would have been static as well, just that they may be more perceivable in wider gamuts or higher saturation's. But, obviously I'm not sure, and haven't succeeded at creating an action that matches the colors; so you may be 100% dead on the money and it just simply cannot be done as I was hoping. Oh well, thanks for the info! Now I know ....

Bassun wrote in post #15178318
It seemed to me that since the ICC is a static file that the adjustments would have been static as well

Only if the colors in the image were "static" as well

LOL, well, that would make it easy .

I guess there just isn't a quick fix to help your print versions look more like your web versions. Thanks again both of you. I appreciate you taking the time to keep me from trying to accomplish a fools errand.

Bassun wrote in post #15181948
LOL, well, that would make it easy .

I guess there just isn't a quick fix to help your print versions look more like your web versions. Thanks again both of you. I appreciate you taking the time to keep me from trying to accomplish a fools errand.

Another thing to bear in mind is the two natures of the "dynamics" of the print compared to the monitor. A print is "reflected" light whereas a monitor is "projected" light. How much of a difference this makes would be, well, relative depending on the image and also the viewing "conditions".

And then, of course, is the fact that monitors and printers each have "limitations" in how they can display various colors. The sRGB color space is the "safest", but, well, we are still dealing with "imperfect" devices...

Here is my process for "calibrating" a lab for my workflow.
I select a group of images that represent the kinds of images I plan to print. Color range, subtle tonal transitions, skin tones, etc.

I then make two JPGS in Lightroom for printing. A) has sRGB assigned. B) has the lab's ICC profile assigned. I then send the images to the lab, specifically stating that they are to be printed without the lab doing any color management. (Some labs have a "professional" service where they will "adjust" the color based on their own workflow.)

Naturally, if there are multiple paper services and printers involved this has to be done for each.

My experience, using this testing process, is that its nearly impossible to tell the difference between sRGB and the lab's ICC profile for any C Print on any paper. Specialty papers and printers, i.e. canvas, is also most often virtually indistinguishable.

Using this technique, I've been able to use my local Costco for proofing 24x36" canvas prints from a completely different lab. That is, a 16x20" print done by Costco on their Epson printer was used as a proof for a family portrait that was printed on canvas by a different lab.

DMWard -- That is some fascinating information. Not necessarily applicable to EZprints per se as it seems they don't want us to assign the profile to their images, rather just use their profile for screen proofing. BUT -- there are other print houses other scenarios where that could be very useful.

Thanks for taking the time to enlighten me with your findings!

I soft proof using the EZPrints profile and require zero adjustment for color. I do need to use a bit of Curves to primarily adjust contrast and maybe tweak brightness a bit. What are your soft proof settings? Here's mine:

Bassun wrote in post #15189005
DMWard -- That is some fascinating information. Not necessarily applicable to EZprints per se as it seems they don't want us to assign the profile to their images, rather just use their profile for screen proofing. BUT -- there are other print houses other scenarios where that could be very useful.

Thanks for taking the time to enlighten me with your findings!

Soft proofing is a useful tool. I have found that I can get a good result by having calibrated monitors, controlled lighting for my work environment and then using lightroom and/or Photoshop to assign the appropriate profile for output. This was useful for a variety of photo labs and also for submitting TIFF files to a printer for a book we published.

What's important to remember is that the output from a printer is always flatter and with less dMAX than a screen image.

Reading the EZPRINTS page about soft proofing I find it interesting that they have a single ICC profile built for use soft proofing. They apparently spend considerable time and effort to make sure that all their printers (they list 4 manufacturers) are calibrated so images offered to them will be within acceptable tolerances, providing they are soft proofed with a single ICC profile.

Considering that soft proofing is intended primarily to ensure that out of gamut colors are rendered properly, it probably means that if you use my "calibrating" technique with EZPRINTS you will end up with a source for printed products that look essentially the same regardless of the printer used to create the product.

I'm going to have to run them through my test process.

BOB:
My settings are identical in PS. Below are mine for LR and PS:

I just noticed that my LR and PS Proofs are set up differently, lol. I think i actually changed to perceptual in LR on accident when I brought up the screen, as I'm pretty sure it was relative as well...

I wouldn't bother with soft proofing without a calibrated monitor. Once you do that and get things close get some test prints made then adjust your monitor brightness based on those prints.

If your calibration device sets it start with a brightness of 110 cd/m^2. If the device you purchase doesn't let you set the cd/m^2 and if you are using a Dell IPS monitor start with the brightness set to around 15% (yes, it will look pretty dim ).

Soft proofing works very well, however I wouldn't get overly fussy unless:
- you have an ISO compliant print viewer
- you have a good understanding of the lighting that the print will be displayed under
- you calibrate your display to the recommended contrast ratio for the paper you're printing to

... and even then it takes some technique/judgement when you toggle Proof Colors on and off.

Bassun wrote in post #15189245
I think I need to first get some ink sprayed and see how they look compared to my monitor.

EZPrints uses wet chemistry photographic paper not inkjet. IIRC they use Kodak Endura.

I have a calibrated NEC PA241W monitor with the luminance set such that the brightness of the image on the monitor closely resembles the print for my viewing conditions. When I calibrate my monitor to a contrast ratio of approximately 250:1 there is only a slight loss of contrast when Proof Colors is toggled on. After making a slight contrast adjustment using Curves the prints from EZPrints match as close as I could ever expect.

If my monitor wasn't calibrated (including the proper target contrast) and if I didn't dial down brightness correctly, there would huge shifts in contrast/brightness when Proof Colors is toggled and the print colors would be hit or miss.

Bob_A wrote in post #15189463
I wouldn't bother with soft proofing without a calibrated monitor. Once you do that and get things close get some test prints made then adjust your monitor brightness based on those prints.

If your calibration device sets it start with a brightness of 110 cd/m^2. If the device you purchase doesn't let you set the cd/m^2 and if you are using a Dell IPS monitor start with the brightness set to around 15% (yes, it will look pretty dim ).

Soft proofing works very well, however I wouldn't get overly fussy unless:
- you have an ISO compliant print viewer
- you have a good understanding of the lighting that the print will be displayed under
- you calibrate your display to the recommended contrast ratio for the paper you're printing to

... and even then it takes some technique/judgement when you toggle Proof Colors on and off.

Bob -- I think you probably hit the nail on the head right here. I am getting worried about things that I don't really need to worry about just now. Almost putting the cart in front of the horse. For now, I think I will relax a bit, get some prints made and see just how far off I am. I suspect from what I've seen so far they will be a little darker, as expected, and a little less punchy. In reality, that will probably be just fine. When I get to the point where I am not happy with the results, or as special needs arise, then I will look into this again very seriously. For now, I'll just do the soft proof to make sure nothing looks way different and learn through experience what to expect from the print house.

Thanks to EVERYONE for such great info. I will definitely reference this thread in the future and have certainly learned a bit more over the last few days. It's great knowing there are so many knowledgeable people in this forum who are willing to step up and help out someone who isn't at their level of experience yet. I genuinely appreciate it.