Luminous Landscape Forum
Raw & Post Processing, Printing => Printing: Printers, Papers and Inks => Topic started by: FrankG on April 05, 2022, 09:31:21 am
-
This question is regarding print/display luminance matching. I understand that the soft proof can show out of gamut col.
Everything I've read says to un-check the Simulate Paper Col box. Why not? See the extracts quoted below the post.
I have a calibrated iMac display (xrite i1- D65, 2.2, 90) and a col corrected viewing booth (pantone). I have calibrated to a low-ish luminance value (90) to compensate for the screen being too bright resulting in slightly too dark prints (I use a printer/paper/ink profile in the Ps print dialog).
Generally, I also raise my midtones by about 10 points with a curve prior to printing & this seems to compensate (maybe its for over inking effect?) and it gives me satisfactory results.
I recently sent a bunch of images to a pro lab for c-type printing (Fuji Matte Maxima- a warmish paper) and used their custom icc profile for soft proofing the paper.
The prints are a little too dark.
However, if I go back to the file on my display, with the print alongside, and look at the soft proof again (Perceptual & Bl Pt checked as per lab instructions), and if I toggle Simulate Paper Col, then the display matches the print.
So, my questions are - is checking the Simulate Paper Col box never to be done, why not? Or could/should it be done on a paper boy paper basis as per my above experience?
I obviously want to hit the mark and not waste paper, ink, or lab costs + wait for a lab print to be done before finding out that it's slightly too dark.
What do you advise?
Simulate Paper col on or off
Perhaps lower Luminance value further to 80.
Thanks
Here's what I found to explain what it does -
From Red River - "So the white you will see is the white point of your monitor, not the print. When turned on, the simulation uses the paper's white point. In short, the system tries to replicate the actual white shade of the paper you are using."
From Cambridge in Col - "Simulate Paper Color. This setting converts the image's on-screen white to match the color temperature of white on paper (which is equivalent to an absolute colorimetric conversion). In addition, this also compresses the dynamic range of the on-screen image so that it matches the narrower range in the print (similar to the above "black ink" setting), but it does so by also decreasing the on-screen intensity of white. When paper color is checked, the black ink setting is therefore usually unavailable (since the on-screen dynamic range is already being compressed)."
-
A soft proof cannot show out of gamut colors (colors that are out of display gamut). If they fall outside display gamut but exist in the output color space, you can't see them.
If everything you've read says un-check the Simulate Paper Col box, you're reading the wrong stuff**! :D
The DR of the display is very often far exceeding the DR of the print. The Simulate options provide a more 'accurate' (visually matching) soft proof because it takes paper white and black ink into account. It may look flatter and uglier (Jeff Schewe famously calls this option the "Make my image look like crap button") but reality can suck sometimes.
When you wish to compare the print next to the display for a match, you want that check box ON:
(http://digitaldog.net/files/Print_to_Screen_Matching.jpg)
As for your 'pro lab', if they do not allow you to actually convert to the output color space, based on the rendering intent you select and using the profile you provide, they are not as professional (in terms of color management) than you think they are and if the prints are 'too dark' (compared to the display), part of the issue is they do not allow you to actually use the profile as intended.
But this is all summed up here, so just watch:
Why are my prints too dark?
Why doesn’t my display match my prints?
A video update to a written piece on subject from 2013
In this 24 minute video, I'll cover:
Are your prints really too dark?
Display calibration and WYSIWYG
Proper print viewing conditions
Trouble shooting to get a match
Avoiding kludges that don't solve the problem
High resolution: http://digitaldog.net/files/Why_are_my_prints_too_dark.mp4
Low resolution: https://youtu.be/iS6sjZmxjY4
**Cambridge in Col (ignore this site, filled with misinformation). A comment here from a person who knows what he's talking about had me in stitches when I read this, so true:
https://forum.luminous-landscape.com/index.php?topic=120665.msg1004366#msg1004366
I wonder how many people reading that link realize the image they incorrectly tout as the "Profile Connection Space" is actually the human gamut CIEuv chromaticity graph? Might as well have a cat picture.
-
Thank you for this. I watched the audio/video & will need to watch again (several times) to absorb it all.
The first thing to establish is how to measure my viewing booth light to ensure I am viewing the prints under sufficient/correct illumination.
My viewing booth does not have a dimmer as yours does. It has 3 switches which activate 3 different col temperature bulbs - I think that one is 5000K and another is 2800K and the 3rd has them all on with a combo of 3800K.
I also have a desk lamp with a 9W 800 Lumens 5000k led bulb.
-
The first thing to establish is how to measure my viewing booth light to ensure I am viewing the prints under sufficient/correct illumination.
My viewing booth does not have a dimmer as yours does. It has 3 switches which activate 3 different col temperature bulbs - I think that one is 5000K and another is 2800K and the 3rd has them all on with a combo of 3800K.
I also have a desk lamp with a 9W 800 Lumens 5000k led bulb.
You don't necessarily have to measure the viewing system and if you could, it might not be useful. The calibration of the display targets to the display soft proof is trial and error. You do want to stick with one CCT (color temp) probably CCT 5000K which is a large range of possible colors so YMMV. That's the color of white, not the intensity and again, you'll have to experiment with calibration settings to get the match as shown in my screen capture earlier.
-
This line that you wrote is really, in a nutshell, the answer to my original question. For that 'Thank you'.
When I view the prints from the lab, under suitable viewing conditions, they match my display only when the Simulate Paper Col box is checked -
"...The Simulate options provide a more 'accurate' (visually matching) soft proof because it takes paper white and black ink into account. It may look flatter and uglier (Jeff Schewe famously calls this option the "Make my image look like crap button") but reality can suck sometimes.
When you wish to compare the print next to the display for a match, you want that check box ON:"
With that box ON, I can make necessary adjustments to the image & more accurately preview the print (perhaps make a duplicate & have 2 windows open side by side, one to toggle the soft proofing on/off & the other with adjustment layers).
Would you say that having the "Make my image look like crap button" on is true across the board for all or most papers (as long as the profile is good).
-
Would you say that having the "Make my image look like crap button" on is true across the board for all or most papers (as long as the profile is good).
Yes because again, you're viewing a reduced contact ratio and when you see the soft proof update, it looks 'worse' than before. The trick is to position your cursor over the option (or in PS make a saved Customize Proof Setup) and toggle this ON without viewing it update. Let your eyes adjust away from viewing this for a second or two, then view the new preview. Also, view in Full-Screen view because the GUI in PS (whites) doesn't undergo this white simulation and your eye adapts to the whitest white. So you want to compare the soft proof to the print in this kind of viewing condition: Simulation ON, no other GUI elements.
-
Thanks.
And could you say something about the separation of the White and Black simulation - 2 check boxes (Simulate Paper Col & Simulate Bl ink).
Why, in what kind of scenario, would you not want both On at the same time?
Perhaps so that you can make an adjustment for each individually/separately?
-
I don't see the need for separation but it exists. For soft proofing to a print, you want both.
-
Thank you
-
How would you typically go about bringing back the contrast that is lost with the Simulate button ON?
A Curves or Levels Adj layer, or?
And, while the Soft Proof dialog is open, with the check/uncheck the preview button, you can't access the Layers palette, so what is the best workflow to make the necessary adjustments?
-
How would you typically go about bringing back the contrast that is lost with the Simulate button ON?
Mostly, you can't. Just as you can't use Vibrance or Saturation to boost out of gamut colors; they are out of gamut.
The display is an intermediary device you use to view your images as you prep them for print and when you soft proof, that's the reality. When you post to the web, while you can't control how others see your images, you can soft proof to a display (yours) and that's what you get.
You can make minor adjustments to some areas of the image under soft proofing to provide some changes based on the soft proof indeed. Here is where Lightroom shines as you can build a virtual copy with those edits. Or use Photoshop and adjustment layers for that print.
I don't see how turning on and off the soft proof help you considering the above. But if you want a 'before and after' preview, fine. But again, the print media has some limitations (as do displays) and there are areas where you can't fit round holes in square pegs.
-
"You can make minor adjustments to some areas of the image under soft proofing to provide some changes based on the soft proof....use Photoshop and adjustment layers for that print. "
Yes, this is what I'm talking about. How to make these minor adjustments while the soft proof reference is open/available.
To recap, I was looking at print/s that were 'too dark' compared to my 'ideal' displayimage in Ps.
Then I used the softproof profile & checked the Simulate box, now the print was a close match to the display image. Which tells me that the soft proof is good indication of how the image prints on that printer/ink/paper. It didn't/won't print like my ideal image because I didn't account for the paper white/black points.
Going forward, before outputting to print I would want to make adjustments to the file so that the soft proof image will look closer to how I would want the print to look when printed. Taking into account the unavoidable difference between the 'reality' of the backlit screen vs reflected light paper. And the subsequent loss of contrast etc. If I preview how it's going to turn out printed, the I can do something about it.
-
This thread, learning to use simulate paper col in soft proofing, has been a game changer for me. I opened my file as I'd like to see it with final edits, duplicated it & viewed the 2 side by side in full screen mode. Opened the soft proof & selected the profile & checked the boxes. Then started to adjust it with levels & curves until it matched as close as I could get (unfortunately I have to close the soft proof box before each adjustment & then reopen it). Printed it and voila, the print matches the image.
Thank you for that.
-
...The DR of the display is very often far exceeding the DR of the print. The Simulate options provide a more 'accurate' (visually matching) soft proof because it takes paper white and black ink into account. It may look flatter and uglier (Jeff Schewe famously calls this option the "Make my image look like crap button") but reality can suck sometimes.
...But this is all summed up here, so just watch:
Why are my prints too dark?
Why doesn’t my display match my prints?
A video update to a written piece on subject from 2013
In this 24 minute video, I'll cover:
Are your prints really too dark?
Display calibration and WYSIWYG
Proper print viewing conditions
Trouble shooting to get a match
Avoiding kludges that don't solve the problem
High resolution: http://digitaldog.net/files/Why_are_my_prints_too_dark.mp4
Low resolution: https://youtu.be/iS6sjZmxjY4
Andrew, your video tutorial offers the best advice I've seen yet on the whole "screen to print matching" methodology that printmakers need to master in order to get softproofing with the simulate paper color checkbox "on" to work as advertised. I totally agree with you that all too many monitor calibration videos give calibration setpoint recipes but skip over the additional and very significant steps which need to be considered for the print viewing environment in order to make screen to print matching work as well as it should.
That said, I don't follow your logic of adjusting the dynamic range of your monitor (when it's possible) to more closely match the dynamic range of the chosen print media. My understanding is that this methodology is only useful when one does not intend to use the "simulate paper color" setting because simulate color paper is in fact already performing that compensation. A commercial photo lab operation might use a reduced monitor contrast range, for example, when running high volumes of RC gloss/luster type photo media so that staff does not have to contend with all nuances of invoking ICC profiles and softproofing yet still see a reasonably closer visual "match" to print output.
AFAIK, the vast majority if not all of today's display profiling apps map RGB 000 to L= 0 and RGB 255 to L=100, whether the monitor's dynamic range can achieve that or not, in effect a relcol with bpc type of scaling algorithm baked into the display profile. For a monitor with monitor white scaled to L*= 100, the monitor white to monitor black cd/m2 measured values must have a contrast range of approximately 1000:1 to generate an actual L*min of less than 1.0 for maximum black. That outcome also assumes no veiling flare reflection from the screen surface. My NEC Spectraview natively reaches about 685:1 during calibration, thus a black L* min of approximately 1.4. That's quite good but definitely not perfect. And to achieve even that, veiling flare has to be carefully controlled because even the best anti-reflection coatings still reflect about 1% of the light reaching the display panel surface. Thus, illumination on your monitor screen surface needs to be kept very low (i.e., less than 50 lux) otherwise your observable monitor black will climb higher than the calibration predicts.
To summarize, when the CMMs translate the print media L*min value to the soft proofed black visual appearance, the calculation is essentially based on the display profile mapped to the full 0-100 scale, again, whether the monitor/viewing environment has achieved that range or not. When the monitor calibration comes up short such that the viewable black is in reality significantly higher than L=0, then the displayed soft proof dark tones also end up a little on the lighter and lower contrast side of perfection. With today's high dynamic range flat panel displays, the typical display contrast error isn't all that big any more, but back in the CRT days with typical contrast range of 300:1 or so (achieved L*min approximately 3.0), I actually edited the paper profile's black point downward about 3-4 L* units to compensate for the display profile's L* min error. Otherwise, the simulate paper color checkbox aka, the "make it look like crap" mode made the soft proof look even lower contrast than was deserved. The problem was traceable to the display profile's assumption of a perfect L* contrast range. Modifying the ICC paper profile's inverse transform LUT was the only way to correct this error back then because profile editing software didn't exist to edit the display profile directly (and still doesn't to my knowledge). I do recall a display profiling app made by Integrated Color Corp (now out of business) that could perform absolute scaling of the measured display LAB data as an option rather than recol w/bpc , but it was definitely an exception to the norm.
kind regards,
Mark
http://www.aardenburg-imaging.com
-
That said, I don't follow your logic of adjusting the dynamic range of your monitor (when it's possible) to more closely match the dynamic range of the chosen print media.
I get a better visual match.
https://www.xrite.com/service-support/contrast_ratio_in_i1profiler
-
I get a better visual match.
https://www.xrite.com/service-support/contrast_ratio_in_i1profiler
Looks like the Xrite to Calibrate transition messed up that link. I get to a page that looks correct, but the link on that page Is no longer to your article.
cheers,
Mark
-
Looks like the Xrite to Calibrate transition messed up that link. I get to a page that looks correct, but the link on that page Is no longer to your article.
cheers,
Mark
Here's what it (used) to say:
When we calibrate a display, we specify what is called target calibration aim points. We inform the software that we want a certain White Point or Luminance (specified in cd/m2) of backlight intensity. These aim points are selected to produce a visual match between the display and the print viewed next to it. Therefore, the correct values are those that produce a visual match between these two items.
The new i1Profiler software provides another useful and important target calibration aimpoint: Contrast Ratio. To understand what this aim point does, and why it's so useful, consider what the contrast ratio describes. It's the range between the brightest white and darkest black. Contrast ratio is described using a scale such as 300:1. In this example, the brightest white is 300 times brighter than the darkest black. The higher the value, the greater the range between black and white. Here's the math for figuring out the contrast ratio of a display: Divide the white luminance by the black point. If you calibrated the luminance to 150cd/m2 luminance and a black point of .25, that would result in a 600:1 contrast ratio between the brightest white and the darkest black of this display.
Modern LCD displays are capable of producing very high contrast ratios. Manufacturers of such displays like to provide these high values as they are attractive for those playing video games or viewing video content. Yet these high values are not useful for those processing images and desiring a good soft proof between a display and a print. On a very glossy paper stock, using a very black ‘photo ink‘, this print may provide a 300:1 contrast ratio. Yet your display may have a contrast ratio of 800:1, 1000:1 or higher. If you soft proof in Photoshop and use the Simulate Paper Color and Ink Black checkboxes found in the Customize Proof Setup (Fig 1), Photoshop uses the ICC printer profile to adjust the print contrast ratio onto the display. The results are often disappointing because you see this very high contrast ratio display become dim as the proper contrast ratio for print viewing is simulated. It is far better to calibrate the display's contrast ratio rather than adjusting the ratio solely by using the paper and ink simulations in Photoshop. When using just Photoshop to do this simulation, only the image, not the rest of the user interface is adjusted which is far from ideal.
This is where i1profiler’s new contrast ratio target calibration aim point comes into play. If you know you will be viewing a glossy print and you wish to have a better match of the screen to the print, start with a target calibration of around 300:1 and compare the image being soft proofed on the calibrated display with the print properly illuminated next to that display. You may need to raise or lower that setting until you get a better match. The same is true for adjusting the white point and luminance! For matching to matt papers, a 200:1 contrast ratio may be a good starting point. It's not possible to provide an exact value! It takes trial and error to produce the best settings because everyone’s print viewing conditions, inks, papers, and displays are different. The idea is to control not only the target calibration aim points for white point and luminance for a print match but the contrast ratio too. When soft proofing in Photoshop, with the paper profile loaded and the two simulate checkboxes on, you should see a very good screen to print match. The key is the appropriate calibration target values, and setting the contrast ratio has a profound effect.
And previously by many years (2003), an article with the aid of Karl Lang for Sony Artisan:
http://digitaldog.net/files/BlackisBack.pdf
-
It is far better to calibrate the display's contrast ratio rather than adjusting the ratio solely by using the paper and ink simulations in Photoshop. When using just Photoshop to do this simulation, only the image, not the rest of the user interface is adjusted which is far from ideal.
This doesn't make sense to me (not saying it is wrong though). If the simulate paper and ink does the job, why would it matter what happens with the rest of the interface? Perhaps a more thorough explanation of why this is the case would be helpful.
-
This doesn't make sense to me (not saying it is wrong though). If the simulate paper and ink does the job, why would it matter what happens with the rest of the interface?
That part of the GUI doesn't undergo white simulation, so your eye adapts to that white, not the white in the image. It is why the "make my image look like crap" button makes your image look like crap just after you click on it ;D (if you view it). See my suggestion about not doing so.
Now you don't have to work in full screen mode to view just the image.
-
That part of the GUI doesn't undergo white simulation, so your eye adapts to that white, not the white in the image.
Very helpful, thanks. Do you suggest always working with lower contrast (not just when printing the image)? The NEC Spectraview software I use also offers the ability to make a display profile with a lower contrast ratio.
-
Do you suggest always working with lower contrast (not just when printing the image)?
I don't really see any reason to do so, the idea is to better match the DR of the display to the printing which varies (sometimes a lot).
-
That part of the GUI doesn't undergo white simulation, so your eye adapts to that white, not the white in the image. It is why the "make my image look like crap" button makes your image look like crap just after you click on it ;D (if you view it). See my suggestion about not doing so.
Now you don't have to work in full screen mode to view just the image.
Except that it's easy in PS to set the canvas color PS uses to surround the image when you zoom out such that it matches the paper color LAB value or perhaps be slightly darker gray, at which point the human visual system will not get fooled by any small bits of monitor white in various menu headers or tool bars. Those small monitor white areas in your field of view will look somewhat "brighter than paper white" not unlike specular highlights in a real world scene, but the soft proof's paper white appearance will still appear to be paper white. In fact, the human visual system's adaptation for whitepoint is dominated hugely by one's peripheral field of vision which typically takes in much more than just your monitor unless you like to sit really close to your monitor! If that weren't the case, humans would constantly be sensing continual color balance shifts as they focus directly on different colored objects in the scene. This adaptation via peripheral vision is also why many folks prefer a D65 calibrated monitor over a D50 monitor calibration. The D50 setting often looks inherently too yellow because they are adapting to the color temperature of the daylight hitting neutral surfaces in the room as registered by their peripheral vision rather than the monitor whitepoint color as seen in their central field of vision. Light the walls with 5000K and the D50 monitor loses it's yellow color cast!
With regard to Karl Lang's article about the Sony Artisan, his examples were all about the benefits of matching monitor black point to a paper media black point without any further softproofing methodology. Perhaps I missed something, but the cited article did not consider how the contrast reduction acts in conjunction with "simulate color paper" or "simulate ink black softproofing features" in PS.
I do see how monitor contrast reduction can be a viable approach for folks who don't want to go all the way with a rigorous screen to print matching approach facilitated by softproofing and simulate paper color or black ink modes. I also see how one can use a contrast corrected monitor in conjunction with a soft proof mode when it stops short of invoking these simulate paper color or black ink options. However, invoking both features simultaneously (i.e., the reduced monitor contrast along with either the simulate paper color or black ink feature) are going to combine together to produce an even lighter black value rendered to the monitor. That's a given based on how display profiles are implemented today. Thus, it doesn't make sense to me that one can routinely achieve better screen to print matching by combining both methods as opposed to creating an accurate printer profile for your chosen ink and media and letting the soft proof with simulate color paper mode correct both whitepoint and blackpoint appearance on screen.
cheers,
Mark
http://www.aardenburg-imaging.com
-
Except that it's easy in PS to set the canvas color PS uses to surround the image when you zoom out such that it matches the paper color LAB value or perhaps be slightly darker gray, at which point the human visual system will not get fooled by any small bits of monitor white in various menu headers or tool bars.
Nope, that's not enough. You can got to go into full screen mode, hit the tab key to remove ALL panels and dialogs and menu items and end up with just the image alone, no tools or anything else (which again, doesn't undergo the simulation) to remove all the incorrect white of the display. Doable. But now you can't edit the image: open up say Curves because you want to apply them: Boom, incorrect white is back.
Full-screen mode, no GUI whatsoever doable and fine if all you want to do is view the image soft proof while you compare the print in viewing booth. Forget any kind of editing, however.
Adjust the contrast ratio of the display calibration, Boom! Everything displayed is now where you want white. And Black!
With regard to Karl Lang's article about the Sony Artisan, his examples were all about the benefits of matching monitor black point to a paper media black point without any further softproofing methodology.
It isn't Karl's article although it is his design (way back when) to control contrast ratio and the results of the control over black. And White. As illustrated in the article images. In no way is this and soft proofing mutually exclusive by design.
I do see how monitor contrast reduction can be a viable approach for folks who don't want to go all the way with a rigorous screen to print matching approach facilitated by softproofing and simulate paper color or black ink modes.
Actually the opposite. It is why Karl's Artisan and i1P and NEC and Eizo, among others provide such controls. While so many others do not sadly.
Thus, it doesn't make sense to me that one can routinely achieve better screen to print matching by combining both methods as opposed to creating an accurate printer profile for your chosen ink and media and letting the soft proof with simulate color paper mode correct both whitepoint and blackpoint appearance on screen.
The contrast ratio can be 'controlled' by the profile, the display or a bit of both but one takes care of anything displayed and the other doesn't. All that matters is it is controlled to provide a match and it does. But when you do it with the display, you control the white of everything on that display. And the "make my image look like crap" issue goes away. And you can easily edit the image, with all the tools, without having to worry that the white of those tools affects the preview (soft proof).
It works. It's been done for a very long time. At least since the early 2000s when Karl designed the Artisan. For Black and White. For the entire ratio of the display. Other's followed Karl as I indicated. For good reasons.
-
Nope, that's not enough. You can got to go into full screen mode, hit the tab key to remove ALL panels and dialogs and menu items and end up with just the image alone, no tools or anything else (which again, doesn't undergo the simulation) to remove all the incorrect white of the display. Doable. But now you can't edit the image: open up say Curves because you want to apply them: Boom, incorrect white is back.
Full-screen mode, no GUI whatsoever doable and fine if all you want to do is view the image soft proof while you compare the print in viewing booth. Forget any kind of editing, however.
Adjust the contrast ratio of the display calibration, Boom! Everything displayed is now where you want white. And Black! It isn't Karl's article although it is his design (way back when) to control contrast ratio and the results of the control over black. And White. As illustrated in the article images. In no way is this and soft proofing mutually exclusive by design.
Actually the opposite. It is why Karl's Artisan and i1P and NEC and Eizo, among others provide such controls. While so many others do not sadly. The contrast ratio can be 'controlled' by the profile, the display or a bit of both but one takes care of anything displayed and the other doesn't. All that matters is it is controlled to provide a match and it does. But when you do it with the display, you control the white of everything on that display. And the "make my image look like crap" issue goes away. And you can easily edit the image, with all the tools, without having to worry that the white of those tools affects the preview (soft proof).
It works. It's been done for a very long time. At least since the early 2000s when Karl designed the Artisan. For Black and White. For the entire ratio of the display. Other's followed Karl as I indicated. For good reasons.
Thanks for that.
Or could it be that the procedural steps you use, and the steps I use achieve similar visual accuracy by virtue of the fact that the true wildcard in screen to print matching not accounted for by CIE color science is the human perceptual phenomenon called lateral adaptation? I pay particular attention to this effect when I'm making image edits in my softproofing workflow. Wasn't mentioned in your tutorial, and it's not accounted for by monitor contrast range adjustments or by ICC profile LUTS or the CMM. Must be accounted for by the printmaker's full understanding of the phenomenon.
kind regards,
Mark
http://www.aardenburg-imging.com
-
Thanks for that.
Any time. ;)
Or could it be that the procedural steps you use, and the steps I use achieve similar visual accuracy by virtue of the fact that the true wildcard in screen to print matching not accounted for by CIE color science is the human perceptual phenomenon called lateral adaptation?
Try the procedural steps I outline, let us know.
"It is better to debate a question without settling it than to settle a question without debating it". -Joseph Joube
-
Any time. ;)
Try the procedural steps I outline, let us know.
"It is better to debate a question without settling it than to settle a question without debating it". -Joseph Joube
I did so yesterday, and confirmed that, in my opinion, matching the contrast ratio of the monitor to the media white/media black contrast range does not lead in my viewing environment(s) to fundamentally better softproofing (better being defined by me as a soft proofing experience that enables superior image edits for print in less time and material losses and which result in superior image quality in a typical real world range of viewing environments.
Moreover, constantly resetting monitor contrast to match print output is impractical for printmakers like myself who frequently change media, inks, and even use different printer models on a regular basis. Additionally, the impact of double compensating for paper black, first by adjusting the monitor to the paper contrast, then by invoking the simulate paper color function in PS which, wait for it, also dramatically compensates the image contrast albeit not the non color managed element of real estate on the monitor. And this double compensation error gets worse and worse for media choices as their L* minimum printed black point goes up, e.g., for fine art matte papers and canvas media where L*min > 15 is very common, i.e., typically < 50:1 contrast range; negating for me, any realizable advantage gained by lowering the visual contrast of non color managed real estate on the monitor. As noted in an earlier post in this thread, I find I can mitigate the influence of the non color managed real estate in view by setting menu color/tones responsibly, and no,I don't feel a need to resort to full screen image deployment in PS to achieve an excellent soft proofing experience, but that is indeed a personal "how one's brain ignores imperfections or not" kind of issue ;D
And as I've noted earlier in our debate/discussion I do agree that professional display calibration software should indeed allow the enduser to specify the monitor white point and monitor black point over a generous contrast range e.g., as low as 50:1 to the "native"highest response of the monitor (which nowadays can easily be 700:1 or greater). If, for example, a photo lab or print studio owns multiple monitors that aren't all factory/dealer matched and may indeed be made by different manufacturers, then wise color management practice would be to calibrate all the monitors to a commonly achievable contrast ratio. I can also see where some printmakers will prefer Digitaldog's recommendations and perhaps even tune their setup to soft proof with reasonable expectations without even turning the "simulate paper color" feature on. But from a color science perspective compensating for print contrast range by using both monitor contrast reduction and then applying a simulate paper color algorithm on top of that is not technically defensible. It's kind of like double profiling. Not really sure why it works for you Digitaldog, but if that's your experience, I'm not inclined to believe I can get you to change your mind. So, let's just agree to disagree. ;)
-
Moreover, constantly resetting monitor contrast to match print output is impractical for printmakers like myself who frequently change media, inks, and even use different printer models on a regular basis.
Takes me mere seconds on a SpectraView.
I can also see where some printmakers will prefer Digitaldog's recommendations and perhaps even tune their setup to soft proof with reasonable expectations without even turning the "simulate paper color" feature on.
That was never my recommendation!
So, let's just agree to disagree.
Use any workflow that works for you. The OP asked as did others, I explained mine and like you, they can go in any direction that they desire that works for them.
But the bigger point is, that control over calibration and DR of a display I've explained is common among higher-end products and solutions for a reason many of us find very, very useful.
-
UPDATE - Good news.
We have discussed at some length in this thread the potential impact of a non color managed PS or LR interface on one's ability to achieve good screen to print matching while soft proofing the image. Digitaldog's video discusses one approach to the problem, namely calibrating your monitor to a lower contrast ratio similar to what the paper and ink choice delivers so that no areas on the display exceed the contrast ratio of the paper.
I was curious to figure out why I have not felt a need to drop my monitor contrast as per Digital dog's recommendation in order to compensate for these non color managed areas of the PS interface. I used my Mac's digital color meter app to evaluate all the PS interface color values. It turns out that in recent versions of PS, Adobe has modernized its user interface appearance themes. There are four to choose from: very dark gray, dark gray, gray, and light gray such that for the most part they safely work within a narrower contrast range compatible with pretty much any chosen media, i.e., presenting a luminance contrast range no greater than approximately 50:1 over the vast majority of the PS interface. The dark gray and gray themes are good to go right out of the box. The very dark gray and lightest gray choices have some minor issues which I'll talk about below.
VERY DARK GRAY theme: If one chooses the very dark gray interface theme, then the preview area in the interface, i.e., the area surrounding the image when the image isn't magnified enough to fill the interface, is too dark, not RGB 0,0,0 but still too dark to be an optimal choice. Thus, when choosing this very dark gray theme the discerning user should simply go one step further to customize its preview area by right clicking on the Mac (probably similar shortcut key on a PC) with the mouse pointer hovering over this preview area and then choosing "select custom color". That brings up the color picker tool and using the tool's LAB numeric fields just select a value for L between 20 and approximately 90 with a* and b* left at zero. These values represent typical worst-case paper white and paper black outcomes. Or you can get fancy and match your paper white or paper black LAB values exactly if you like. In other words, simply customize the preview area of your chosen workspace interface so that it is compatible with even the lowest contrast matte papers you are likely to print on, or if you like, match your paper white or paper black LAB value exactly, and you're good to go with this VERY DARK GRAY theme.
LIGHT GRAY theme: Like the dark gray and gray interface appearance themes, the lightest gray PS interface theme is essentially good to go out of the box as well, but it does use L=100 for some minor menu bar areas, and that choice does not appear to be user addressable. These menu bars constitute less than 98% of your display area, so the human visual system will basically treat it like a specular highlight and not change its overall brightness adaption level. However, if you want to be obsessive compulsive about your soft proofing environment, just don't choose this lightest gray appearance theme. I've personally never used this lightest interface theme because I think it looks butt ugly, but somebody out there must like it :)
Bottom Line: While it can still be argued that in a perfect world the whole PS interface should be color managed using the chosen ICC printer profile when in soft proof mode, Adobe's modern interface appearance themes largely avoid brightness adaptation level problems by virtue of the fact that their neutral color range graphics design stays within a much lower (approximately 50:1) contrast range whether your display is calibrated to a much higher contrast range or not.
final note: I don't use LR and haven't checked its interface design, but I'd wager it follows the PS interface design closely. Hence, LR will likely also pose no brightness level adaptation problems when using its soft proofing features.
cheers,
Mark
http://www.aardenburg-imaging.com
-
final note: I don't use LR and haven't checked its interface design, but I'd wager it follows the PS interface design closely. Hence, LR will likely also pose no brightness level adaptation problems when using its soft proofing features.
http://www.aardenburg-imaging.com
Might want to check before we wager. ;D
In some respects, LR is superior and more modern in handling all this UI when soft proofing; you can actually select paper white for the image surround, unlike Photoshop, or like PS, a larger number of percentages of white to dark. But unlike PS, it doesn't do this everywhere; the menu text menubar is one example. But for that matter, PS doesn't alter the menu text either, although that's pretty tiny, and if you use key commands, not pertinent. And like PS, tab key provides full screen mode.
It turns out that in recent versions of PS, Adobe has modernized its user interface appearance themes.
This has been around since at least CC 2017.
-
Feedback & a question.
I just received c-type prints from the lab after soft proofing with their icc profile & using the above mentioned 'makes my image look like crap' button.
I made some contrast & other layer adjustments to bring the image back to being closer to how it was & as I'd ideally like it.
Worked out well.
However, one thing I did notice on a few images that contained warm tones (yellow / red) is that these colours are a little more saturated than expected. This could of course be a problem with their profile or my display calibration or my layer adjustments...
I contacted them and they recommend calibrating to D50 & viewing at D50. Whereas Ive been calibrating to D65 & my old Pantone viewing box is, I think/not 100% sure, 5000K (D50)
Q1: Would the difference between their recommended D50 & my D65 (1500K?) calibration be what accounts for the yellows being more saturated?
Q2:And if so, would it make more sense to re-calibrate to D50 (considering that that profile & paper/print type isn't the only one that I use), or to simply compensate by adding a Hue/Sat adjustment layer and taking down the yellow (&/or red) a little when using it (trial & error)?
-
However, one thing I did notice on a few images that contained warm tones (yellow / red) is that these colours are a little more saturated than expected. This could of course be a problem with their profile or my display calibration or my layer adjustments...
I contacted them and they recommend calibrating to D50 & viewing at D50. Whereas Ive been calibrating to D65 & my old Pantone viewing box is, I think/not 100% sure, 5000K (D50)
The print itself is more saturated than the display in yellows? These colors may be out of the display color gamut. If you post an example of the yellows (I need say 300x300 pixels, even a crop) and your display profile, I can compare the color gamuts in ColorThink Pro.
DO NOT futz with the calibration at this point or based on what they said just because some yellows are not a perfect match. Is the rest of the image when soft proofing close?
Q1: Would the difference between their recommended D50 & my D65 (1500K?) calibration be what accounts for the yellows being more saturated?
Q2:And if so, would it make more sense to re-calibrate to D50 (considering that that profile & paper/print type isn't the only one that I use), or to simply compensate by adding a Hue/Sat adjustment layer and taking down the yellow (&/or red) a little when using it (trial & error)?
There is only one device that actually produces D50 or D65, it is 93 million miles from your display. 1500K is a large range of possible colors too! CCT expressed in Kelvin isn't a defined color like D50 or D65 as seen below:
(http://digitaldog.net/files/LinesOfCCT.jpg)
On the line labeled e-f, any color can be called/considered 5000K.
You can recalibrate and see if the soft proof is closer or farther off but again if everything 'syncs' up visually but a range of yellows, you may likely be in a good place and just dealing with out of gamut colors. Altering the WP as suggested will alter all the colors which may not be what you want or need.
-
I went through the entire process and started a reply with an attached 300px crop + the display profile.
And then had another long and close look at the prints and the screen image and re-evaluated.
Now find that it's not as far off the mark as I had at first thought. The difference is more my misinterpretation of what exactly I delivered to the printer. I'm going to re-do the file/s and re-print them. I'll get back to the thread (it takes a couple weeks for the print job turnaround + the postal service to deliver). Sorry to waste your time now.
The chart with the e-f line shows the explanation of different colours all being at 5000K very well . Thanks.
-
Ever since I was taught here to check the 'simulate paper / ink' my results have improved dramatically. I do thenwork with adjustment layers to bring the image back to the pre-soft proof look (to match the original) as best I can within the limitations.
A small detail...in Edit>Col Settings >Advanced... what % do you desaturate your display ? The default is 20% (see screenshot). Or is it already taken into account by the display calibration software? Any other settings here that I should be changing?
For your interest only, no debate needed, there's a daily 'photo land' newsletter Petapixel that just this morning published another article on soft proofing that again omits the 'simulate paper & ink' button - https://petapixel.com/soft-proofing/
-
Your color settings seem fine* and indeed, keep the desaturate option OFF; it's a huge hurt me button you should never use.
The Petapixel omits an important step in soft proofing while mucking up some understanding of this topic; nothing new. :'(
*Edit: I personally prefer Ask over Preserve but that's not critical.
-
Thanks