Consistently dark prints

I am printing with an Epson 3800 in Windows 7 64 bit and PSCS5 64 bit. I have the Epson print dialogue setting in PSCS5 at no color adjustment and my PSCS5 print window has Photoshop Manages Color. The profile is Adobe RGB and I use Relative Colorimetric as my rendering intent. Black Point compensation is checked. My prints are dark even when I add a brightness and contrast adjustment layer. I know my prints won't ever match the backlit luminosity of my display, but the prints display very little of the luminous quality at all. Is there some other setting in Windows 7 I should be adjusting?

On a side note, I have made a new Print Profile using Spyder3Print. I've made two profiles two days in a row now using both version 4.0 and 4.2.1 and those profiles makes things even worse. At least my old profile from before the upgrade to version 4.0 comes a bit closer.
Aside from the general frustration, I have to make 50 or so proof prints to deliver to a printer for a calendar by the end of February.

Thanks for any suggestions.

The most common reason for this problem is having the monitor brightness set too high. If the monitor has been calibrated and the brightness is set to 100-110 cd/m2 your prints should be quite close to the monitor. I don't have Windows 7 yet but I'm sure others will add their thoughts about any settings changes you may need to make. Good luck!

Thanks Alan. I appreciate your input. Unfortunately, that's not the problem. My display is set at @91 cd/m2. My software, Spyder again, found my ambient lighting to be low and therefore recommended 90. I just ran another test just to be sure and that's ok.

Steve, feel free to email me your monitor and printer ICC profiles and I can take a look at them with some tools such as Gamutvision and see what may be going on in terms of the color management pipeline. Search for "Gamutvision" in the archives and you'll see quite a number of in-depth debugging threads where we've explored all kinds of color management issues. The more "interesting" situations usually have multiple issues going on at the same time, which can make resolving things a bit complicated.

Alan is correct that the most common reason for dark prints is overly bright monitors. But we've also seen mistaken color management settings, bad monitor profiles, bad printer profiles, double profiling, operating system problems (usually on Mac), and combinations of images & print media that were simply very challenging to work with. Just to name a few. Searching the archives, I note you had some speed bumps with monitor calibration & profiling back in late 2009, after getting the Spyder3Studio kit going on your new system. From reading that thread it's not clear to me how those issues were finally resolved.

A couple of other questions. What monitor are you using, and what kind of video connection to the computer? (Back in 2009 it looked like a combination of an older ACD and a newer Dell.) Also, what paper are you printing on? And do you have a link to an image online that you're specifically having these printing challenges with? It's useful to do some analysis of profiles using one of the actual related image cases.

Also consider the effects of ambient light on your perception of the monitor and print. I know you're aware of this from your recent post about evaluating prints, Ott lights, etc. but it may be a factor. A monitor obviously puts out its own light; but depending on what monitor you have as your primary, a 90 cd/m2 luminance target may be too low. Typically only high-end monitors designed for color critical work can be accurately calibrated to that low a level. Whereas on the other side, a print can only reflect whatever ambient light is available. If that light is dim, the print will look dim as well regardless of what the monitor shows.

Hi Royce.
Thanks! I'll definitely take you up on that.
I'll try to answer a couple of the questions as best I can, but whatever understanding I appear to have is definitely rudimentary at best. I know where all the cables go. After that...
The issue from back in 2009 is the same as now. Until recently I made very few prints and, from some reading I did, was under the impression that even with the best of profiles some continued tweaking after the first print wasn't unusual, so I've been tweaking along since then.
I'm using a Dell 2408WFP that's about two+ years old. The video card is ATI Radeon HD 4600 Series. The display is connected with a DVI cable and a USB. I do also have an ACD, but that is used only as a palette display in PS. I am printing on Epson Premium Luster.
Here's one of the images and here's another. But it's the case with all images I try to print.
As far as the ambient light issue. The 91 Cd/m2 came from the software's suggestion based on it's reading of the ambient light with the Ott Light off. However, I use the Ott to view the print alongside the display so the room ambient light is a reduced factor....I think.
I'll try to capture the profiles and email them along with these image links.
I really appreciate you taking the time. This is way beyond me. Maybe it will turn out to be something simple, but finding simple things can be pretty complicated sometimes.

These questions are often solved with just one little wrong thing in the convoluted Epson driver. posting screen shots of every single driver screen will help us debug the problem.

If everything is right, do you soft proof, sometimes that will help get things right prior to sending to the printer? Also, the monitor will always look brighter since it is transmitted light rather than reflected light. Do you have a proper print viewing set up with 5000 Kelvin bulbs?

Holy Crap! 5000 bulbs? That's a lot of bulbs. Just kidding.

Yes, I do soft proof and there is a slight difference but not nearly as great a difference as with the print. No I don't have a 5000 Kelvin setup. Just my Ott Light. I'd love to have a viewing booth, but it's not in the budget.

The concept of a viewing booth is great, except most photos will not be displayed under those iidyllic conditions. If you are printing just for yourself, view the prints under the same lighting condition as to where you want to hang them. Daylight versus incandescent makes a big difference on how the print will look.

If you are printing for sale, most people don't have a viewing booth they would be using. Instead, they will most likely hang their photos in locations ( living areas ) which are usually either lit using daylight during the day, or are artificially lit using incadescent lighting. In either case, they are not viewed under 5000 Kelvin bulbs.

My only point on the 5000K was that the only way to have a prayer of the print looking like the monitor is if you look at the print under 5000K lighting conditions. If you look at a print under incandescent lighting, it can't even remotely look like the image on the monitor.

I use the syder3 studio with the 4.2.1and I am extremely happy with the results but there is apparently an issue if you use the measure ambient light set up so I never use it as apparently it can cause more problems than it solves. I am amazed how close my prints are to my monitor. I do use an NEC monitor though.

Thanks guys.
The reason I got the Ott light was it's reportedly close resemblance to natural light. I try to remember to turn off the ambient lighting (spiral fluorescent) before viewing.
Trev...I'll try skipping the ambient light step later on today and see how much of a difference it makes. It would be great if that solved the problem.

When you soft proof, do you check the "simulate paper white" box? I found with that box checked, my prints are always too dark. Uncheck the box and things are much closer. Default may be with the box checked, IIRC.

I might be wrong but I don't think the "simulate paper white" box has anything to do with the actual printing, it IMO is a waste of time as to me even in soft-proofing your images before printing it doesn't give an accurate "proof" of how the print will look. Just a washed out, low contrast looking example of your print.

Steve and I have been cranking through some stuff offline. So far, the biggest issue is that the Spyder-generated printer/paper profiles were quite troublesome. Probably there are some other things going on as well, but reverting back to using the stock Epson 3800 profile for the luster paper in question as a sanity check has already put things back into a better zone.

I'll have more info when I've got time to assemble various screenshots and commentary...

Royce-I was interested in this particular comment about OS problems with Mac (which I use-OS 10.4.11 on a Dual 2 Ghz Power PC G5). I have an Epson 4000 and recently bought a 3880. I use ..(I know...cheap and not too great) Huey for my HPZR24w monitor. Have the Epson paper profiles for both printers and some Red River profiles. I work from Photoshop 4 in Adobe RGB.
Colors are quite accurate for my 4000, though I usually adjust slightly for an inherent magenta cast. I too have had problems with the 3880 - colors are dark...or should I say saturated? Can't fathom why there is such a difference, though I realize the workflow is very different. In the print dialogue for the 4000 I check "proof", and use "Proof setup" with the proper paper profile. I set "no color management". Usually print at 750 DPI as there seems to be very little difference between that and 1440 for most things.
For the 3880, I click "Document", "Photoshop manages Color", Printer Profile with Black point compensation, and disable "Printer manages color". (It should be automatically disabled, I think, but it isn't). I can't always choose 750 dpi- 1440 is OK. Nice rich color, but usually I end up desaturating the colors on my screen to get the proper color on the 3880. It's hit or miss.

So...maybe I can climb on this bandwagon? Any thoughts about Mac in particular?

Thanks!!!

P.S. Monitor is not "overly bright"....

[quote="Royce Howland"].......

Alan is correct that the most common reason for dark prints is overly bright monitors. But we've also seen mistaken color management settings, bad monitor profiles, bad printer profiles, double profiling, operating system problems (usually on Mac), and combinations of images & print media that were simply very challenging to work with.

sgingold wrote:The profile is Adobe RGB and I use Relative Colorimetric as my rendering intent.

I'm sure this was checked... but "what" profile is Adobe RGB? Not the Printer Profile in the Adobe print dialog, I hope? I don't have CS5, but in CS4, this is where I'd select my specific printer/paper profile, which is obvious not Adobe RGB....

Hi Rob. Thanks for jumping in. Naw, that's just my convoluted way of expressing myself in a very disorderly fashion. Adobe RGB is the profile in Photoshop before going to print rather than ProPhoto or sRGB. I do use printer profiles in the print dialogue.

That's a relief! I'll let Royce & co. offer their expert advice then...
Oh.... Matte or luster paper?

Luster paper.

Thanks to everyone who has contributed their thoughts to this. It's ongoing, but with Royce's help I'm getting closer to a solution.

Datacolor was/is of little help. I finally did connect with a tech who gave me more than stock answers, but he kept insisting my main problem was that I wasn't selecting "NO color adjustment" in the Epson dialogue, even after sending him a screenshot of the dialogue showing it to be selected. Upon sending him the xml file of the target readings he did point out that the pure patches of color were actually contaminated with other colors, so that did help me evaluate the quality of my targets with a loupe. But the Spyder profiles are still turning out much less than ideal.

I followed two of Royce's suggestions and one has already been an improvement....using some profiles from Bill Atkinson. The other, having Eric Chan make a profile, is still in the arrangement process, but I'm sure that will solve my issues.

So thanks again to all, and especially to Royce who has been putting up with my incessant questions.

EJ's point about what light you view your print in cannot be stressed enough. You might take your finished print outside in daylight, inside into the kitchen for flourescent usually, and then into incandescent light. I was shocked at the differences. I print specifically for the light of my display area.

If you are into photography contests, one sneaky trick is to find out what type of light the judges will be viewing the prints.