The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Printer vs Photoshop color controls?

plinden

New member
Hello, I would appreciate any thoughts, suggestions, or explanations you might be able to give regarding something I recently encountered when printing from Photoshop.

I thought I knew what I was doing but now I’m not so sure. All along I’ve been able to make high quality prints but with some considerable trial and error. I was following what I thought was appropriate protocol using Photoshop. [I use high quality cameras/lenses, do my image editing on a 5K iMac and print using an Epson 7900 printer and I’m generally satisfied with my icc profiles.]

My Method

I use the Adobe RGB setting in my cameras and shoot raw of course. Then, in the raw conversion process I open a file using the ProPhoto RGB color space in Photoshop, where I do my editing. I employ Photoshop’s soft proofing feature, using the appropriate paper color profile (and specifying perceptual rendering intent). When it comes time to make a print I let Photoshop do the color management, specifying the paper profile, and perceptual rendering.

Usually the first print comes out flat/low contrast and desaturated relative to what I see in the soft proof, but through trial and error I add fudge-factor layers to get the look I want from the final print. While tedious at times I was resigned; I thought this was the way to do it.

Their Method

Recently my college teacher daughter explained that in her photography class (that uses a more consumer grade Epson 800 printer), she and a tech guy co-teacher weren’t happy with this “traditional” (Photoshop) protocol and have resorted to a work-around:

After opening a file in Photoshop they convert the image profile to the desired paper profile, using relative rendering intent, and then they use the printer’s (Epson’s) color controls through the Photoshop dialog box, again specifying relative rendering intent. In their eyes the results looked more accurate and neutral.

This didn’t make sense to me so I began experimenting, and darned if a few of my test prints didn’t look MUCH better using their method, and they were a better match to the soft proof screen!

So I tried this approach using a GretagMacbeth Color Checker as a test image, as well as another common color reference image. The results are consistent with my first impressions, the work-around gives better looking results!

Any Thoughts?

Has anyone else experienced this? Do you have any ideas why this is happening?

I’m most troubled by the idea of it, I’m happy to use this approach in my work, but what should a college level photography professor advise her students?

Thank you.

Peter
 

Jack

Sr. Administrator
Staff member
My Method

I use the Adobe RGB setting in my cameras and shoot raw of course. Then, in the raw conversion process I open a file using the ProPhoto RGB color space in Photoshop, where I do my editing.
Firstly, the camera Adobe RGB setting only affects the in-cam jpeg. Since you say you shoot in Adobe RGB, are we to assume you also convert your raws to Adobe RGB? If so, and you then edit in ProPhoto RGB, your first step has to be converting the image to ProPhoto RGB! If you don't do that, you will get flat results. In PS Open image; then goto Edit>Convert to profile>choose ProPhoto. Of course it's much easier just to output the raw to the same color space as you work in*... Regardless, once the image and working space are the same, edit and print per your described method (which is sound) and all should work fine.

*You could (should IMHO) output a ProPhoto RGB or camera space RGB image during raw conversion, and edit in that same space so that now your image matches your working space for editing and output. Alternatively, you could keep your converted file in Adobe RGB, and set your PS working space to Adobe RGB and you'll be fine too. But on par, editing in ProPhoto or the camera space is probably slightly better for maintaining color fidelity throughout the workflow...
 
Last edited:

glenerrolrd

Workshop Member
You may have to search to find the blog post ....but..there was a post on Luminous Landscape where the author found that color managed by the printer produced better prints . This as you mentioned is inconsistent with almost everything else that has been suggested regarding print work flow settings . The testing was extensive and the authors argument was compelling .
 
In your daughter's method, three things are changing – (3) the variable you mentioned – whether PS or printer is in control; (2) the unusual practice of converting the file itself to a paper profile, (3) the more common practice of using Relative instead of Perceptual intent.

It would be a good idea to isolate the variables in your experiments, to see if one or another alone, or some combination, is what gives you more acceptable results.

And be sure to let us know!

Kirk

(My own guess is that RC instead of Per is giving you a little more contrast. And my experience is that ProPhoto isn't the boon it's supposed to be. It can push some colors out of gamut when you print in a more restricted range.)
 
Last edited:

Rand47

Active member
Since you shoot in Adobe RGB, and then edit in ProPhoto RGB, your first step has to be converting the image to ProPhoto RGB! If you don't do that, you will get flat results. In PS Open image; then goto Edit>Convert to profile>choose ProPhoto. NOW edit and print per your method. Should work fine.

Alternatively, you could output a ProPhoto RGB image during raw conversion, and now your image matches your working space for editing and output. Or, you could keep your converted file in Adobe RGB, and set your PS working space to Adobe RGB and you'll be fine. But on par, editing in ProPhoto is probably slightly better for maintaining color fidelity throughout the workflow...
This is just wrong. The “in camera” AdobeRGB setting affects only the jpeg image. RAW data doesn’t have a colorspace. If the OP’s working colorspace in Photoshop is ProPhotoRGB and he is importing raw files into Photoshop that should be “it” and will not be the cause of any of the issues he’s describing. The only other possibility for AdobeRGB is if he’s going in via Camera Raw, in which case it should also be set to ProPhotoRGB (but could be something else).

His daughter’s work around sounds like a real Kluge to me.

I’d suggest softproofing in Lightroom, so much easier than Photoshop and the LR print utility with its ability to make presets for print/paper settings is worth learning the Print Module, even if that were the only thing you use Lightroom for.

I shoot raw, import to Lightroom (default color space is a flavor of ProPhotoRGB) do basic adjustments in LR and round-trip to Photoshop if I need additional work that cannot be done in LR. Then soft proof and print in Lightroom to my selected paper and whichever rendering intent looks best for that image (very image / paper dependent and defaulting to Preceptual is a suboptimal way to work, IMO).

My screen to print match is usually spot on, the first time. I sometimes make further iterations based on evaluating the print under proper lighting, but almost never because what came out of the printer isn’t what I saw on the screen.

I would be willing to bet that there’s something else amiss in the OP’s work flow that is causing his problems with screen to print match.

Questions for the OP:

Do you calibrate your monitor with a hardware/software solution such as X-rite i1Display Pro ?

Have you printed a standard print evaluation image w/o changing anything in that file? (Printing Insights #48) And then adjusted your monitor luminance to match the luminance of the print while viewing it under proper print evaluation conditions? Gti Light booth, Solux color proof lighting, or something similar?

Are you editing in consistent ambient lighting conditions?

Another possible problem might arise if you are using the gamut warning and trying to adjust for “out of gamut” colors in your soft proof. The gamut warning is, as a practical matter, useless. It only indicates that something is out of gamut, but is incapable of telling you anything useful about “how much” out of gamut the colors are. Some of my students try to elimainte any out of gamut indications by manipulating the soft proof and in doing this almost always end up with a flat and desaturated print.

Rand
 
Last edited:

archiM44

Member
My prints from an iMac and an Epson SC P800 match the screen pactically perfectly, color as well as brightness on many different papers.
I print from Lightroom, but it make no difference whether I print from Capture One Pro or Photoshop.
Using X-Rite i1PRO and i1Profiler Software I have made printer profiles for all my papers and have my iMac screen profiled to a luminance of 70. The standard luminance is much too high.
Working space in Photoshop as well as Capture One is Prophoto, Lightroom uses something close to Prophoto but it makes no difference whether I print using the original adjusted RAW file or a SRGB jpg file. As long as I turn off color adjustment in the printer software and use the correct paper profile
 

Rand47

Active member
"As long as I turn off color adjustment in the printer software and use the correct paper profile . . . "

Yup... if you don't you'll have dueling profiles and that gets really ugly! :)

Rand
 

archiM44

Member
In my post I neglected to add that it is important to specify the paper type, both when profiling and printing (i.e. premium glossy, luster, matte etc.)
 

Stuart Richardson

Active member
Hi Peter,

I run a Hahnemühle Certified Studio, so I spend a lot of time printing, and I have run the full gamut (pun intended?) of tests regarding color management. I use a wide gamut Eizo monitor and have made custom profiles for my papers as well as used expert made profiles from companies in the US and used Hahnemühle's own profiles. Currently I use the P9000 printer, but I used the 9900 in the past. Believe it or not, I have tried the Adobe RGB method your daughter employs, and in many cases it gives excellent results. In particular, I have found that it gives a better black level and shadow gradation in matte black prints on Hahnemühle Photo Rag as compared to my custom profile and Hahnemühle's profile. My guess is that there is that the profiling process introduces some degree of error or somehow slightly elevates the black level compared to the depth that the printer itself is capable of producing. In general, if you convert your file in Photoshop to Adobe RGB or leave it that way, and then print with Photoshop color management off and set the printer to Epson color and then Adobe RGB, you can take advantage of this extra black level and color depth. That said, I do not believe that the saturated colors are always as accurate as they are with a profile. For important jobs I generally run test prints in both methods and see which gives the better results. If absolute color accuracy is more important than black depth and vibrance (which I find rarely to be the case in working for the artists I work for), then I would stick with profiles, though it does sound like something is going wrong for you. I am able to achieve very good matches to the screen with both methods, so it sounds like something might be slightly off for you. At the end of the day, the most important thing is to get the results you want with as little ink and paper waste as possible. If that means using Epson color controls, then by all means, go for it. The printer color management has come a long way from what it used to be in the early days. I actually came to the Adobe RGB method in printing a job for a client who had an art piece with a digitally rendered whirlpool...it was a smooth gradation of gray into a void, but every time I printed it there was a hard clipping in the gradation. So instead of a smooth transition from light grey to black, as the color got darker there was a noticeable step from gray to black. No matter what profile I used I could not get rid of the harsh transition. Finally, I tried printing it without a profile and just treating it as Adobe RGB and sure enough the smooth transition on screen was rendered on the print. After that I no longer blindly questioned that profiles were always the best solution for every print. They are a translation of the data, and sometimes things get lost in translation.
 
You may have to search to find the blog post ....but..there was a post on Luminous Landscape where the author found that color managed by the printer produced better prints . This as you mentioned is inconsistent with almost everything else that has been suggested regarding print work flow settings . The testing was extensive and the authors argument was compelling .
Roger,
Thanks for quoting this article. It was written by Edmund Ronald in July of 2013.
I corresponded with Edmund and thanked him for the article as it made a difference with a difficult image I had been trying to print.
Unfortunately, what was once free to read when Michael Reichman was alive and ran the website will now cost you under Kevin Raber's ownership. Everything published prior to Kevin's ownership is now offered at cost, which is a bummer, since Kevin is now monetizing contributor's posts and articles that were done out of a desire to contribute to the community.
 

Rand47

Active member
Hi Peter,

... In particular, I have found that it gives a better black level and shadow gradation in matte black prints on Hahnemühle Photo Rag as compared to my custom profile and Hahnemühle's profile. . . .
Some of Epson's newer matte papers have excellent Dmax. Epson Legacy Fibre as an example. Especially when used with the HDX inkset.

As a practical matter any dark areas with detail w/ an L value of less than about 7 seem to block up on most matte papers.

Rand
 

plinden

New member
Thank you all for your helpful comments. This information enabled me to decide what I might say to a class such as my daughters'. I've passed on your inputs. I'm curious how she and the co-teacher decide to handle this. (I see it as a "teachable moment")

Peter Linden
 

Shashin

Well-known member
Ultimately, it is all about results. The path may be different. The only criteria that is important is a color managed monitor and good view conditions. Beyond that, what gets you there is the best approach. In one case in my lab, we had two generations of the same printer series from Epson. Having set them up, on first blush on an image with a fairly large palette of colors like a test target they appear identical. One student however said they were different, given he was color blind, I did not put a lot into it. However, a little later we were getting complaints about the colors on one of the printers. After a bit of work, one printer was printing blues deeper than the other. When an image was blue dominant, the difference was really obvious, but with a fairly balanced image, it was hard to see. There is more to color that meets the eye...
 

NeilBurton

New member
I am also late to this thread but there may be another observation here. Apologies in advance for spelling colour with a u!

Prophoto and Adobe RGB have different gamma values and consequently different tone curves. Prophoto has a gamma of 1.8 while Adobe RGB has a gamma of 2.2. ItÂ’s conceivable you may be seeing these differences because of this.

Prophoto is IMHO problematic as a working colour space because of its extreme gamut volume and its gamma. ItÂ’s 1.8 gamma is inconsistent with the 2.2 gamma that we typically calibrate our monitors to. Prophoto’s gamut volume doesnÂ’t really correlate with colours in real life either - it is a HUGE colour space. Great for archiving strange colours but not so good for making big editing moves that lead to out of gamut values when translated back to a monitor or printer.

Our digital cameras are capable of capturing colour well outside of Adobe RGB. So Adobe RGB as a working colour space is not always ideal. Additionally many good printer profiles punch outside of the gamut of Adobe RGB as well.

An alternative working colour space should ideally have a gamma of 2.2 and a wide enough gamut to contain the colours that our digital cameras are capable of.

So a suggestion is to explore Joseph HolmesÂ’s DCAM colour spaces for use as your working space. DCAM 4 is big enough for anything my Phase One can capture and is my choice of colour space for most images. The gamma is 2.2 and the gamut is just right. If you need a saturation tweak, his chroma variants will increase saturation without the undesirable hue shifts that occur in RGB colour spaces.

DCAM 5 is too large in gamut. DCAM 3 is sometimes too small and can be clipped by colours from digital capture. I have explored this using ColorThink Pro software by Chromix.

ICC profiles can also be calculated with different tweaks for things such as gamut compression and saturation. So your ICC profiles can also be tuned to taste. I calculate 6 different tweaked ICC profiles for every target I measure on a Barbieri LFP S3 and calculate with Basiccolor Print software. Cross polarised light using the M3 measurement condition can also pay dividends for improving the dark tones into the blacks.

Hope this helps.
 

Godfrey

Well-known member
Hello, I would appreciate any thoughts, suggestions, or explanations you might be able to give regarding something I recently encountered when printing from Photoshop. {... snip ...}
Referring back to the original post: Missing from this discussion is a description of the complete workflow, from setup to output. I'll present my sum up of all the things I've used as the basis of my configuration and workflow that you can consider for your processes.

Camera configuration:

Color space setting in the camera ONLY affects in-camera JPEG files. When you're working with raw files, color space is applied at raw conversion time. So it really doesn't matter what color space you set in the camera. I always set it to sRGB, because that way in-camera JPEGs that are good are the most compatible with typical display setups.

Color spaces, gamma, and white point:

There are a bazillion different options here, but I choose to consider proper use of only three of them for compatibility.

sRGB is a color space defined in the middle 1990s by Microsoft et al to model the imaging gamut and capabilities of the 8-bit CRT displays in use then. It has become the de facto standard of the computer display world. Obviously, vast changes have happened in the intervening years, but this dinosaur with its small gamut has remained the standard.

AdobeRGB (1998) was invented by Adobe when the prepress world began the shift to digital image editing for web press machines. The gamut and dynamic range of CMYK printing engines is very different from that of computer displays, and this was one of Adobe's original business thrusts, so they modeled the CMYK printing press to obtain a better fit for editing files to be printed on these machines, reducing clipping. AdobeRGB (1998) can be used with both 8bit- and 16-bit-per-component files, and provides a useful expansion of editability without clipping.

ProPhoto RGB was designed about the turn of the century to allow editing the larger gamut and dynamic range possible with digital capture sensors and raw image capture. It is a 16bit-per-component color space intended to allow the maximum editability of digital captures without clipping. Note that it was NEVER intended to be a display or printer color model, it was designed with the notion of conversion and translation to smaller gamut color spaces required for such output devices in mind. The point was to make a color space that is far larger than any device or paper can image so that editing flexibility is maximized, and conversion to device color spaces can be contrived while minimizing clipping.

The native gamma of the 1990s era CRT displays tends to be 2.2 due to the technology used, and the native white point of those same displays tends to be around 6300° again due to the technology used. These gamma and white point values were adopted by the sRGB standard as its bases when modeling those displays. Apple, big in the early prepress digital imaging market, found that the adjustability and precision of their color displays, when being used as imaging models for CMYK printing presses (which have a closer to 1.8 gamma), became expensive if they put all the hardware required to do the gamma adjustment for proper reference behavior into the monitor, and that that they could reduce costs and provide more flexibility by using 1.8 gamma as a standard and splitting the control and adjustment of that gamma between the adapter card and the in-display hardware. Thus was created the 2.2 vs 1.8 gamma standard split that has existed ever since.

This hints at the underlying notion of what display calibration and profiling is all about. In order for the editing and display process to be uniform across multiple systems for different people to be able to work on the same images and come up with consistent results, the display becomes the reference standard. So, outside of the actual editing process, the displays to be used must be calibrated to specific luminance, white point, and gamma targets so that what is seen and adjusted on the displays can match what a print output device can produce with high fidelity, the print output devices being the less adjustable of the two because of their link to inflexible physical properties of paper and ink. (Printers, ink sets, and papers all can be modeled too, of course, which is what paper profiles are all about. The proper configuration of the display and use of the proper paper profile is what allows high fidelity translation of what the person editing the image sees on the display to be outputted to a print.)

So the procedure to keep in mind when you develop an output/printing workflow is to

  1. Setup the display with proper calibration and profile so that the editor's eyes will see what the image is going to look like when printed to paper.
  2. Edit the image on that display applying the largest gamut/dynamic range color space available to minimize clipping errors in the editing process.
  3. Model or proof the edited output when translated/converted for its intended target device, be that a computer display or a print.
  4. Convert the final edited image to the appropriate bit depth and color spaces for that intended output.
  5. Distribute the output with the image properly tagged for its characteristics.

The detail specifics of the proposed workflow vary depending upon what equipment you have to work with and what the output is intended to be.

My own configuration and workflow:

  • Configure my displays to 110cDm^2 illumination, 1.8 gamma, 5600°K targets. I've found this set of targets makes it easier to model what my printer puts out with high fidelity.
  • Edit images in ProPhoto RGB color space at 16bit-per-component. I do most of my image editing with Lightroom, which automatically translates all imported images to a version of ProPhoto RGB with a gamma of 2.2 (named "Melissa", I think) for editing because it seems to be easier to work with given today's cooler default LCD display screens as a general rule. Works fine with my target settings anyway ... the display gamma translation is happening at the OS to graphics adapter level, way underneath the image editing level.
  • When outputting images for web display, I convert them to sRGB and 8bit JPEGs.
  • When outputting JPEG files for distribution to a client, I first ask the client what their editing staff prefers. If they don't know, I set the output to 8bit sRGB; otherwise, I set it to the specifics they request.
  • When printing, I print from Lightroom using 8- or 16-bit, depending upon the printer capabilities, and apply the correctly defined paper profile for my printer and ink set.

All of this workflow fluctuates some as new displays, new papers, new printers, new inks, etc come along. It's reliably produced very good results for me, and for my clients, since I started my photo business in 2006ish, and has continued to make fine prints that match what I see on the display today, long after I closed the photo business (2010).

G
 
Last edited:

dj may

Well-known member
I calibrate my monitor to 80cd/m and 6500.

I just did two sets of prints, one for a client and one for me (same images). The client's prints are on Ilford black and white paper, glossy. My prints are on Hahnemühle Photorag. I did not use paper profiles. I have had problems with consistency soft-proofing with profiles. I sent Adobe RGB, 16 bit, no sharpening tif files to the printing service (WhiteWall). The tif files were about 200mb each. With whatever calibration WhiteWall is using, the same files were printed on different printers and papers, with very similar results with respect to tonal range and brightness range.

The prints on different papers have different looks, however, they are equally good. The Ilford paper produced slightly better detail in the deep shadows.

It has taken multiple iterations and test prints to get to the predictable results that I now enjoy.

Jesse
 
Top