Contrast Not Accurate On Print

[LeonD]

I use almost exclusively RAW files with Lightroom 5.7, an iMac with a profiled screen, an Epson 3000 and Epson Velvet paper (usually 13x19).

The color I see on the monitor is very close to the color I get on the printed page.  No complaints there.  Where I am running into problems is with the contrast.  While the image on the screen looks fine, the print almost always has too much contrast.

I could purposely have the screen image soft with the hopes of getting a correct print but that doesn't seem right.  Any ideas how I can get the contrast on the monitor to reproduce accurately on the print?

[mbaginy]

Do you soft proof in LR?  I do regularly and then adjust to satisfy what the paper & printer can offer.  Contrast is one of the attributes shown in soft proofing and is adjustable to the gamut permitted by the factors involved.

[digitaldog]

The color I see on the monitor is very close to the color I get on the printed page.  No complaints there.  Where I am running into problems is with the contrast. 

The two sentenced here seem to conflict somewhat  . Color and contrasts matching are both attributes of proper calibration. You do need to soft proof and you do need to calibrate for that kind of 'preview':
http://www.luminous-landscape.com/tutorials/why_are_my_prints_too_dark.shtml

[Slobodan Blagojevic]

Just a guess: if you turn "simulate paper color and black point"* on, the soft-proof screen image might looked more washed out (i.e., less contrasty) than the print. By the same token, the print might look more contrasty.

*EDIT: I mispoke, I meant to say "simulate paper color and black ink"

[digitaldog]

Just a guess: if you turn "simulate paper color and black point" on, the soft-proof screen image might looked more washed out (i.e., less contrasty) than the print.

Which is exactly the opposite of what proper soft proofing (with proper display calibration with a good profile) should provide.

[LeonD]

Thanks for all the suggestions.

I am using ColorMunki Display to profile my monitor.


The two sentenced here seem to conflict somewhat.  Color and contrasts matching are both attributes of proper calibration.

And this is what's leading to my confusion.  The color appears fine; the contrast does not.


I do use Soft Proofing and Simulate Paper and Ink using the profile for Epson Velvet paper/Epson R3000 printer.  I bring up the side by side view with the original and proof copy.  Using matt paper, the proof copy always looks washed out.  It's at this stage where I'll usually either use the tone curve, clarity or contrast to get the proof copy to look like the original.

Thinking out loud here, could this step be overdoing the contrast?  It doesn't make sense as I thought the point of this step was to make the proof copy look similar to the original.

[Fernando García]

Thinking out loud here, could this step be overdoing the contrast?  It doesn't make sense as I thought the point of this step was to make the proof copy look similar to the original.

My first experiences with soft-proofing led to prints with too much contrast and clarity. I find that a restrained correction works best for me, meaning that I try to approximate the proof copy to the orginal. If I try to make them look exactly the same, side by side on screen, I tend to overcorrect the proof copy. A little experience helped to educate my eyes on what to expect from soft-proofing. Soft-proofing is a great help but screen and print are different media.

Assuming your calibration and profile are correct, could this be your problem?

Anyway, this is my experience, hope this helps.

[howardm]

I'm not seeing any place the OP indicates what his calibration target values are for temp, luminance and black point (or contrast). 

I personally set mine (for gloss/lustre) for something like 90cd/m2 w/ 0.3 black which results in about 300:1

[LeonD]

That's because I didn't indicate what they are.  I don't remember.  I usually pick the default.

What should they be for printing on matt paper?

[howardm]

there are no standard 'defaults'.  Factors such as the room and it's lighting and the lighting in the place you actually view the prints come into play as does the device and it's software capabilities.

it's important to keep a log or track of the settings you're using so you can zero in on the correct values.

A luminance of 90-100, black of 0.3 or 0.4 and a temp of maybe D65 (although I use 5500).  Matte paper is pretty low contrast to start with so a contrast ratio of 200-250:1 gets you in the ballpark.  Remember, contrast ratio = luminance/blackpoint

[digitaldog]

And this is what's leading to my confusion.  The color appears fine; the contrast does not.

It's your display calibration! The URL I posted talks about this and more. We don't need to know the specific values to tell you they are incorrect.

[Slobodan Blagojevic]

I still believe that "simulate paper color and black ink," when checked, tends to overstates the effect, making the soft-proof duller, which we then tend to compensate by increasing the contrast. Or we need to be very careful when using that option not to have anything else on the screen (i.e., viewing the image in full screen mode) and not looking at the screen when turning it on (also avoiding switching it on and off in rapid succession). All this because our eyes/brains adapt to the lower contrast, but if we do not give it enough time, or if we keep going back and forth, the soft proof would appear too dull in comparison, thus making us increase the contrast too much.

[digitaldog]

I still believe that "simulate paper color and black ink," when checked, tends to overstates the effect, making the soft-proof duller, which we then tend to compensate by increasing the contrast.

Yeah, I'm sure you do. But that falls in the face of how color management, soft proofing, good profiles and proper display calibration is supposed to work and DOES work for many here and elsewhere. So your advise to the OP is that since you and he can't get an acceptable visual match, that's impossible? The Simulate check boxes are broken? Or you two need more help in getting the system to work as it should? The correct answer is #3 BTW. IF you get a better match without the simulation, something isn't working correctly!

[Slobodan Blagojevic]

...So your advise to the OP is that since you and he can't get an acceptable visual match, that's impossible? The Simulate check boxes are broken?..

That is not what I was saying, and if you read my post carefully, you'd know it. I said:

Or we need to be very careful when using that option not to have anything else on the screen (i.e., viewing the image in full screen mode) and not looking at the screen when turning it on (also avoiding switching it on and off in rapid succession). All this because our eyes/brains adapt to the lower contrast, but if we do not give it enough time, or if we keep going back and forth, the soft proof would appear too dull in comparison, thus making us increase the contrast too much.

[digitaldog]

You're right Slobodan, how the simulation is presented and the surround plays a role as we view the update. But LR does this right by gradually loading the soft proof and providing a lighter bkgnd and further, after adapting, the soft proof with simulations should always look more like the print than without (unless you've also nailed the contrast ratio of the display calibrations instead). IOW, with soft proof on, with ideal contrast for a visual match, you get the match! Which is what and why it was designed that way (view print and display together).

[howardm]

thats not clear (to me at least).  If the display is already cal'd to 200-300:1, will 'simulate paper white' (ESP. for matte paper) be doubly compensating?  It seems that way to me (I rarely use matte but when I do, hitting 'simulate paper white' really is too much and the UNsimulated is much more true to the test print (done w/o any softproof adjust. layers etc)

[digitaldog]

thats not clear (to me at least).  If the display is already cal'd to 200-300:1, will 'simulate paper white' (ESP. for matte paper) be doubly compensating? 

No. And again, the idea is softproof + simulations = match and have display calibration as part of that entire equation.

[howardm]

Hmm, not really the answer I was hoping for relative to what I see on the NEC vs. hardproof on Epson HPB.   As I said, I get much better agreement w/ simulate OFF (w/ matte).  At certain points I just want to shake my head at this whole thing.

[Slobodan Blagojevic]

Andrew, where does profiling get the info for paper color and ink? If from canned profiles, that come from printer and/or paper manufacturers, could it be that such a profile is not precise enough? After all, it is a simulation, not the exact thing. Do you have in mind custom made profiles, where you measure paper white very precisely?

[digitaldog]

Andrew, where does profiling get the info for paper color and ink? If from canned profiles, that come from printer and/or paper manufacturers, could it be that such a profile is not precise enough? After all, it is a simulation, not the exact thing. Do you have in mind custom made profiles, where you measure paper white very precisely?

It gets the info from the same place it get's all the info, from the measured data of the process used to build the profile. There's one or more measurements for paper white and ink black. And, the preview table is different from the output table so it's always possible that the two are out of sync. But good profiles don't or shouldn't suffer any of these issues or one where the profile isn't 'precise'.
Whatever isn't precise visually in that table is accounted for with the display calibration.