Lab Prints too dark for the 2nd time! Newbie, please help!

Discussion in 'Digital Darkroom' started by kristyc, Jan 27, 2008.

  1. Hello, I am a newbie trying to get my photos to match their printing output but
    they keep coming out too dark (or dim, they don't "pop," are dull- take your
    pick). So please hang in there with me...

    Here's the problem:
    I am working with a very well respective print lab (suggested by the members of
    PN). It is obvious that their quality of prints is outstanding by the first set
    of test proofs I received from them, but they appeared too dark. After
    discussing this with them, I calibrated my monitor and received a 2nd set of
    proofs...and guess what? The proofs still look too dark (and now we're going
    for a third set as we speak, hence this post). What am I doing wrong here? I'm
    slighly frustrated and in need of the expertise in this forum!

    The details:
    My camera is set to Adobe RGB; I am working in PSCS3 with Adobe RGB 1998; the
    lab will print to Adobe RGB 1998; I have downloaded/installed all of the lab's
    ICC profiles for soft proofing; and I've borrowed a friend's eye one display
    for monitor calibration.

    Now, I'm recalibrating my monitor for a second time with the following
    suggested settings:

    White point 6500K
    Gamma 2.2
    Luminance 130cd/m2

    My questions are these:

    1) When I view my photo image in PS, the color looks fairly saturated (enough
    to pop but not overdone according to the histogram), but by no means dark. Yet
    when I view the proof from the lab's soft-proofing ICC profile, the image
    instantly lightens. (That's right it LIGHTENS, but when I get the test proofs
    back they are always DARKER than what my monitor shows, even when viewed in
    Adobe 1998!) Why is this?

    After the 2nd set of test proofs came back, the lab suggested that I research
    what my luminance level was due to the fact that I was using a flat panel
    monitor (a Dell 1908FP). They suggested that "some people have difficulties
    with brightness levels when using a flat screen monitor and the output of their
    prints," and so, I should reduce the gamma level from 2.2 to 1.8.

    So I did that, and recalibrated my monitor again. I found out that the
    luminance level was at 140cd/m2, so I adjusted it back to 132 (which was the
    closest I could get it to 130) and set the gamma to 1.8. I again looked at my
    photo, and viewed the soft proofing profile for the type of paper/printer combo
    I want them to use. The image looks even brighter than before.

    2)How am I supposed to "guess" at what the level the brightness is going to be
    (on the print) when I'm editing my files? I thought the whole reason why we
    calibrate our monitors was to be able to "see" what we are getting and get as
    close as possible on the printer output when all is said and done.

    3)Then the brightness and contrast levels went from 51 & 54 to 45 & 96
    (respectively) just by adjusting the luminance and the gamma levels(to 132 and
    1.8). Does this sound right? Like I said, I'm new to this part of photography
    and I'm trying to get all this down, but at the moment I'm terribly confused
    about their relationship to each other in order to produce a decent print!

    4)And finally, whenever I recalibrate my monitor using the eye-one, does it
    calibrate the monitor off the previous calibration setting or should I be
    recalibrating off a default setting?

    I'd appreciate any feedback. I have read through the forum at some different
    threads but I haven't seen any pertaining to this. So any help would be

    Thanks, Kristy :)
  2. I'd keep the gamma at 2.2 and try to knock the monitor brightness down as far as the calibration software will let you. I use an Eye-One as well and used it on a LCD at work.

    I'd also send them a 21 step greyscale patch (do a search) and have them print it. Hold it up next to your monitor to see how it differs (and if you only plan to print there, readjust your monitor to match it).
  3. You have to calibrate brightness for your room lighting, with gamma set to 2.2, as Roger said.
  4. Before anything else, double-check and make absolutely sure you are soft-proofing using the profile for the paper you're ordering the prints on!

    I like the idea above, to send them a grey scale and have them reproduce it. They're the ones who sent you their profiles; if you're monitor is correctly calibrated and you soft proof with their profile, it should match.

    However, one other thought: I also agree that you should see if you can reduce your screen brightness, but bring it down to maybe 100-110 rather than 130 (I assume you're using an LCD display). When I was using one awhile back, I did that and got successful matches on my printer once I did so, using the paper manufacturer's profiles for the paper and my printer.

    You might also ask for the manufacturer's profile for the paper you're printing on, and ask the lab to try printing using that profile.

    If all else fails, try sending the lab files that have been converted over to sRGB -- again, use their profile for soft proofing.

    All that aside, I think maybe there's a lot to be said for photographers owning and using their own printers!

    But, in any event, I feel for ya and wish you the best of luck!
  5. Many labs don't have ICC profiles provided to clients or even downloadable including some pro labs.

    If you have a calibrated monitor - fantastic.

    But you may still be guessing about the printer of the lab.

    Ask the lab and show them the prints and ask if they have a ICC and if not they may tell you to say "this" and they adjust them for you or maybe you can send prints at diff brightness to the lab and get them all back and see which one you like best and them sorta mindmap that one so in the future just raise the brightness.

    Alternatively use a lab with ICC files or get own printer.
  6. She already has a profile from her lab which she uses for soft proof. If her colors are spot on, I would suggest that she have one more try with her lab, and then find a good 'fudge factor', dial it in, and live happily ever after.
  7. Thanks you guys!

    Ok, Roger, I did see an online automated geyscale patch test at drycreek and I remember that on my LCD monitor I couldn't tell a difference in the first four geyscale steps and couldn't see a really significant difference until about step 8-9. So, I will look back into that more and see if the lab can print it.

    Jeffrey, this eye-one is just the eye-one display (which I think has been discontinued) so I don't think this model calibrates for the ambient light in the room. Which brings me to another quick question: This may sound silly, but if I am looking at my monitor which is adjusted for 6500K, should my (ambient) work light (lightbulb) be the same color temp as well for more optimal viewing for photo editing? (I think the current bulb is a standard 5000k bulb,but it's somewhat dim so that my monitor light will standout in the room)...and from looking at the photo differences onscreen, somehow the gamma 2.2 version "looks" better to me.

    Michael, trust me, I have double and triple checked on the soft-proofing profile for the particular paper it's being printed on, LOL! But I think I'll recalibrate, using the 2.2 gamma that Roger and Jeff suggested and then pull down the LCD monitor luminance to what you suggested, (I just didn't realize you'd have to pull it down so far) and have them print a greyscale.

    And Ray, Les is correct. This is a pro lab and they have provided me with their ICC profiles for soft-proofing(many choices of paper and printer combos...which is cool! Variety is the spice of life!LOL).

    And after this next set of proofs, and the color stays spot on, (it was just this brightness factor), then I may ultimately have to get tweaked in with that "fudge factor" Les suggested because I really do like their service and the quality of prints from this lab. They have been awesome to me so far and if I can get this issue straightened out, they'll have my business.

    Thanks to all you guys who wrote back so promptly, I'll give it a go and let this thread know how it turns out. Have an awesome day! Kristy :)
  8. Kristy, I'm not familiar with the Eye-One Display. I use an outdated Monaco Optix XR Pro which generally calibrates the monitor's brightness and fine tunes it to my room's lighting condition. The general brightness calibration is way brighter than when it's fine-tuned for my room's lighting, like 75% down to 50%. If it didn't calibrate for the room's lighting, what I was seeing on the screen would be way brighter than what I would get from my printer, or even if I sent my prints to a lab, the prints would come back darker than what I was seeing.

    I believe, as far as brightness calibration is concerned, the amount of ambient light is more crucial than the temperature of the light.
  9. I use an Eye-One display myself. The ambient light feature of newer devices probably doesn't do much of anything- or at least that's what Andrew Rodney posted elsewhere.

    Keep glare off the screen so you can see the full contrast, and other than that you should be fine.

    You can download 21 step grayscale files here:
  10. Back again with an update!

    I looked into the 21 step greyscale patch test and the most recent link (thank you Roger), and I recalibrated (using the eye-one display) to a 6500K white point; a gamma level of 2.2; and a luminance level of 110 (per Michael's suggestion). I also saw an earlier thread from the forum by Elliot N. which suggested this (instead of just using a single jpg picture to show all of the 21 steps together), which I also tried out:

    Elliot N. suggested: "A 21 step grayscale is too coarse to judge the quality of your calibration.


    Create a full screen black document in Photoshop.

    Make a square inch selection in the middle. Hide the selection outline.

    Hit the tab key to hide the palettes.

    Create a curves adjustment layer.

    Select the shadow point and use the cursor keys to adjust from 0 to 1 to 2 etc.

    Do this in a darkened room.

    You should be able to (just) make out your selection at 1, and then see it get progressively lighter as you click on the up arrow key.

    Do the same with a white document, reducing the brightness of your selection from 255, to 254 to 253 etc."

    My question about this whole greystep test is- Am I supposed to be able to see a gradual color change (in lightness or darkness- maybe density is a better word to use here)upon adminstering this greyscale patch test?

    And if so, does this mean (if I can make out all the 21 steps-which I can) that the monitor is now sufficiently calibrated? Is this correct thinking?

    This stuff can all be so confusing to the newcomer--but I'm definitely trying to hang in there and a certainly APPRECIATE you guys hanging in there with me!

    Thanks again!
    Kristy :)
  11. >>>I use an Eye-One display myself. The ambient light feature of newer devices probably doesn't do much of anything- or at least that's what Andrew Rodney posted elsewhere.<<<

    In my previous post I explained what calibrating room lighting does: it truly calibrates the monitor's brightness so that the print's brightness matches the monitor's brightness. If you don't, or can't do this step, it's just a crapshoot, or else you resort to trying to match the monitor to the print, and then you won't have a calibrated and profiled monitor.
  12. Thank you Jeff,I totally get what you're saying.

    The original eye-one display (that I borrowed) doesn't calibrate for the ambient light in the room, and it is now discontinued. But I believe the eye-one display 2 does calibrate for ambient light which should be a big help and I am looking into getting one very soon; because you are right, the monitor image was "brighter" than my lab prints which appeared much darker when they came back.
  13. Here's another update:

    After trying and documenting various tests at different (forum)suggested level settings, here are my current monitor settings now using the eye-one display:

    White point 6500K

    Gamma 2.2

    Luminance level 110 cd/m2

    (*The monitor was originally at 287; then brought to 130-still too bright; then brought down to 120 and finally to 110cd/m2 on a Dell 1908FP LCD monitor)

    Brightness 39 (which was originally factory set to 75)

    Contrast 100 (originally set at 50)

    The "visual" result is much closer now than it was before when viewed under the soft-proofing ICC profile from the lab (but who knows until the print really comes back, right?).

    However, I notice that the lab is using a "perceptual" rendering intent in their ICC profile and I've been reading how "relative colorimetric" may produce a better looking print (Is this printer or ICC profile specific? Or image specific? Sorry if my "Digital Darkroom language" is off, I'm still trying to understand and learn all of this).

    So, I set my soft-proof view to a relative colorimetric rendering intent and now it looks so much closer to the archive print. In fact, the "absolute colorimetric" rendering intent looks even better on this particular image I'm working with.

    Which one do I pick? Or, am I stuck in a sense because I'm looking at how the image is supposed to be printed under THEIR ICC fixed (with perceptual)profile? Is it possible to have the lab print the image in "colorimetric" or "absolute" rendering? Or is that something they will change on their end upon producing the final print, keeping in mind that I am only looking at soft proof view at the moment?

    Help! Newbie confusion is setting in!

    I will be sending in a request for another set of test proofs in the next day or so under the current monitor settings (above) and I will keep this thread updated whether or not the prints match or if they come out darker (again).

    If anyone has further thoughts, I am open to any and all...

    Thanks again to all who contribute here thus far, I really do appreciate it!

    Kristy :)
  14. >>>Many photographic images could benefit from a relative colorimetric rendering, rather than a perceptual rendering. Remember, color-management systems know nothing about the content of the image itself; they only know about the gamut of the color space the image inhabits. When you use perceptual rendering, the color management system applies the same gamut compression to all images, even when the image contains no visually or aesthetically significant out-of-gamut colors. For instance, with pastel images, perceptual rendering will apply unnecessary gamut compression, while relative colorimetric rendering may produce a result that's more faithful to the original. - Bruce Fraser<<<

    Kristy, maybe the above helps. Since "relative colormetric" seems to produce the image you want, try sending the image files with that intent, and see what happens. I'm not sure how all this is handled on the lab's end. Maybe ask the lab?
  15. "Thank you Jeff,I totally get what you're saying.

    The original eye-one display (that I borrowed) doesn't calibrate for the ambient light in the room, and it is now discontinued. But I believe the eye-one display 2 does calibrate for ambient light which should be a big help and I am looking into getting one very soon; because you are right, the monitor image was "brighter" than my lab prints which appeared much darker when they came back."

    You're welcome Kristy, yes, the Eye-One Display 2 does measure for ambient light, as you mentioned. See Below:

    >>>Eye-One Display, which I reviewed in 2004 included a pretty good colorimeter along with a slightly limited software application. In general, it lacked many of the elements users now demand from display calibration and profiling packages (i.e. speed, DDC compliance, and AMBIENT LIGHT MEASUREMENT). However, with the introduction of the Eye-One Display 2 colorimeter GretagMacbeth (GMB) have significantly increased the speed of the calibration process without compromising quality, and they've also managed to include the ability to measure both ambient light luminosity and colour temperature.<<<
  16. Cool! Jeff, You ROCK! I think we're in the home stretch!

    I sent an email to the lab last night in fact, asking what you suggested this morning and I am waiting for a response today!

    The above quote by Bruce Fraser makes sense to me (thanks for including it!)And it DOES seem to take a long time (especially on the Advanced setting) to calibrate the monitor with the Eye-One Display. So hopefully, when the new Eye-One Display 2 arrives, it will measure the ambient light and we'll arrive at a precisely calibrated monitor! (One can only hope, right?! Whew! What a ride! LOL)

    However, here's a question for ya...if the Eye-One Display 2 takes into account the ambient lighting in order to calibrate more efficiently, and I set the white point of the monitor to a color temperature of 6500K, do I also need to be working under a 6500K bulb (desklamp) to have a more accurately calibrated monitor? If so, will that allow me to produce more visually correct, edited images when viewed under the soft-proof preview? (Wow, that was a mouthful!)I think the one I'm using now is a standard 5000K bulb with a low wattage.

    What do you think?

    Thanks again for hanging in there with me Jeff (and others)! Kristy :)
  17. Kristy, thanks, YOU ROCK! :) And I'm glad to see some progress being made with these issues.

    Okay, don't take this as gospel because I've relied on my common sense, but I surmised that the light in the room (ambient light), if it isn't 6500k, will cause the colors on your monitor to shift, but brightness is still brightness. If you just have regular tungsten bulbs, I would try to keep the ambient light as low as possible, or even off if possible, just because of color shifts. What I've done is turned off my tungsten lights and I use an Ott light (6500K approx.) next to my monitor, and slightly behind it, for viewing prints and comparing them to my monitor (regular daylight works too), plus my monitor has a hood which helps. Also, just a heads up, if you change your room lighting, you need to recalibrate your monitor (of course using a calibrator like the Eye-One Display 2). Obtuse enough for you? ;)
  18. Kristy, just to clarify, I keep my Ott light (actually "Ott Lite" see: turned-on when I calibrate my monitor, and after it's calibrated I make sure I don't move it, to keep lighting conditions consistent.

    This isn't state-of-the-art ambient lighting, but it seems to work great. You can drop a ton a cash trying to get your lighting perfect, but the Ott Lite seems to be a perfectly adequate solution for all but the most critical applications.
  19. Thanks Jeff, I'm still waiting to hear back from the lab, so I'll keep the thread updated. But in the meantime, I'll take a look at the Ott-Lite. Take care, Kristy :)
  20. Kristy, Great idea! Keep us updated.

    You should end up with a calibrated and profiled monitor that will be able to "communicate" with any lab, as well as a personal printer. Also, you'll be able to compare your prints to your monitor and have a darn good match.
  21. The Bruce Fraser quote doesn't apply any more as the Eye One Display uses the SAME software as the Display 2- upgrade to the latest version on the Gretag website.

    I calibrate the monitor and photo edit under the same lighting conditions (a CFL bulb overhead)- works fine. I don't think color temp. of room lights does much to your perception of the monitor as the monitor is backlit and produces its own light. The more important question is what your prints look like under your room lighting- I display prints under similar lighting as I use in my workspace.

    Don't worry about the ambient light "features" newer calibrators have. As Andrew Rodney said elsewhere they're gimmicky at this point.

    Finally- don't use absolute colormetric rendering. Perceptual or relative should be fine but render out of gamut colors differently. I wouldn't worry too much about it.
  22. So you resort to fudging your brightness setting to match the print's brightness no longer having a monitor calibrated and profiled to a universal standard. Heck, on Mike Johnston's Photoborg site, he gives instructions for calibrating without a color calibrator, which is pretty much the same thing if you're not taking advantage of a calibrator's advanced features. Just throw away your calibrator and match prints! You won't be flying straight, but at least you'll be flying, I guess.

    Fraser's quote was about rendering intents, nothing to do with calibrator software.
  23. ">>>Eye-One Display, which I reviewed in 2004 included a pretty good colorimeter along with a slightly limited software application. In general, it lacked many of the elements users now demand from display calibration and profiling packages (i.e. speed, DDC compliance, and AMBIENT LIGHT MEASUREMENT). However, with the introduction of the Eye-One Display 2 colorimeter GretagMacbeth (GMB) have significantly increased the speed of the calibration process without compromising quality, and they've also managed to include the ability to measure both ambient light luminosity and colour temperature.<<<"

    I was referring to this quote which followed a Fraser quote but which was not attributed- I assumed that was from him but I guess was just cut and pasted from some site.

    I do recommend using a 21 greyscale patch for reference as in my experience Fuji Frontiers will print the same SRGB file somewhat differently than say a home Epson printer.

    The Frontier seems to darken the last few black steps and the Epson is more linear. So even if your monitor is calibrated to a known standard your prints *may not be consistent from device to device* and I find it helpful to know what to expect before sending them a print.

    A custom Frontier profile may take care of this- I haven't tried myself.
  24. The EyeOne colorimeter will measure ambient light luminance level and color temp, but it
    only tells you what that is within the software. It can't build adjustments within the
    resulting profile to make the display visually match exactly to your ambient lighting in
    regards to color temp due to the spectral differences between the display and the artificial

    It's mainly used to match luminance which you are better off adjusting the backlight on
    the display itself in reaching a target within the software as indicated by the ambient
    readout rather than letting the software do it through the video card. This is what Andrew
    meant by it being a gimmicky feature.
  25. You basically have two options with colorimeters-either use DDC in matching luminance
    AND color temp through the LCD's own internal multi-bit look up table or use native color
    temp and adjust the backlight using the display's front button for luminance which leaves
    all 8bits available on the video card to build the profile.

    The ambient light readout as measured will only give you a target to shoot for within the
    software, but you'ld want to use DDC for reaching those targets.
  26. Thanks Tim, interesting.
  27. Final Update:

    First of all, I want to thank EVERYONE who contributed to this thread! You guys are all AWESOME and you've all given me a tremendous amount of help to solve this problem. So...THANK YOU! :)

    Second, I want to say that once I ordered the new Eye-One Display 2, (which was another delay in resolving the issue at hand because the company had it on backorder),I found it somewhat easier to use than the original version. It did give me an option for measure the ambient light (for the surrounding work area) but it's reading was for informational purposes only and it's reading was not directly tied into the calibration process itself. Although it was helpful to know that my energy-saving "green" power compact bulbs were reading around 2400K and the lux was aroun 29 (recommended was 5000K and 32-80 lux I believe). So, I'll just change the viewing bulb at my work station to something more comparable. No biggie there. (And I did check out the Ott-Lite online but Michael's craft store also carries a huge array of styles of these types of lights too).

    Now, as far as my monitor-to-print matching goes...well, they are as close as I think they're going to get when viewing a reflective print vs. a brightly back-lit monitor. After so much trial and error, taking notes on the recalibration tests, my final data is this:

    White point 6500K

    Gamma 2.2 (Windows/PC)

    Luminance level 111 cd/m2 on a Dell 1908FP LCD monitor)

    Brightness 34

    Contrast 100

    RGB at 6500K (R=84;G=81;B=80)

    I wanted to report back to the thread because I've had so many people comment to me that they have appreciated all the info found here, so I wanted to make sure there was a conclusion, and a successful one at that.

    Again I want to thank everyone, all of you are true teachers and your efforts to be involved here were not wasted!

    Best regards to all, Kristy :)

Share This Page