Lab prints too dark

[aaronp]

Hey all,

I have a series of photos that I am trying to get printed for an upcoming show

and this is the first time I?ve really tried to print color photos. I have

tried the two most ?professional? labs in town (Jackson, MS) and also mpix.com

but with all three places the prints come out nearly a stop darker than what I

see on my monitor. One of the local labs and Mpix use a lightjet, the local

lab prints on Fuji Crystal Archive and Mpix prints on Kodak Pro Portra Endura;

the other local lab prints using some sort of projection onto a Kodak photo

paper, I?m not sure what the exact print process is. So it doesn?t seem

related to a specific printer/paper. I have had all three places print

without any corrections, so it seems the problem is on my end.

 

I am really confused and frustrated and don?t know what to do short of adding

a curve layer to imitate the print and then another curve on top of that to

adjust to the way I want it to look (basically an ?inverse? of the first

curve), and then applying only the second curve when I get it printed. But

that seems like such an archaic and backwards way in this age of color

management.

 

Has anyone else had trouble with this, or have any idea where the problem

lies? Any help would be hugely appreciated.

 

I use PS 7, am working in sRGB, and have a Dell Ultrasharp 19? monitor

calibrated with the Gretag-Macbeth Eye-One.

 

Thanks,

 

Aaron

[frans_waterlander]

Chances are that your monitor's brightness is way too high. This is even more likely if it is an LCD. I'm not sure if the Dell Ultrasharp 19" is an LCD, but probably so. If your monitor is too bright, you are inclined to reduce the brightness in your editing program, resulting in prints that are too dark. Your monitor brightness should match the brightness of your digital darkroom lighting, so it is hard to give one particular number, but in general your monitor should be calibrated to a brightness of something between roughly 90 and say 120 cd/m^2 (also called nits).

[aaronp]

The Ultrasharp is an LCD, but at the time I bought it, it seemed to be about the best quality for photo editing for the money. I assumed that calibrating my monitor would take into account the brightness, but maybe I'm wrong. How would I measure the "nits" of brightness?

[doug_axford1]

Every good lab will give you a digital file with matching lab print. Simply calibrate your monitor with this image to match their print and all images should now be very close to perfect. Keep it Simple.

[david_henderson]

Certainly with my Dell Ultrasharp 19" it isn't possible to set the contrast, and the Brightness control is reset pretty much by eye when calibrating with a Monaco XR. I happen to have a Kodak Professional Colour Management checkup kit which camw bundled with my Monaco kit. It seems as though I set the brightness pretty well since I get a good match between my screen and the test prints. But it would have been possible for me to set the brightness too high within the calibration routine.

[frans_waterlander]

Aaron,

 

I'm not familiar with the Gretag-Macbeth Eye-One calibration program but it should, I would hope, let you set the brightness level.

[anthony johns]

Try using this...(attachment)

Find the skinny 11.5 pluge bar(the one underneath the black bar underneath the red.) It is on the right side. Set your moniter untill this just visible. Your contrast is up to high if the pure white square blooms and glows alot.

[aaronp]

Thanks everyone for your responses. I ran the Eye-One again, this time on the advanced option instead of the quick option and it does have a brightness level setting. My monitor was set too bright, but that still didn't completely fix the problem. What I see on my monitor is now a little closer to my prints, but still between a half stop to a stop off, even when I adjusted the brightness down to around 90 nits. There is so much more detail in the shadow areas that just gets lost in the print.

 

Anthony, with my brightness set to ~90nits all three pluge bars are discernable. Should I go lower than 90nits?

 

I also downloaded the gamma chart from http://www.normankoren.com/makingfineprints1A.html and my threshold black level is up closer to 1.8 (I have a Dell desktop) and the gamma seems to be around 2.7. Does this make sense to anyone?

[frans_waterlander]

Your gamma should not be higher than 2.2. What gamma did you choose during calibration?

[aaronp]

I chose 2.2, I know, that's what throws me off.

[anthony johns]

Aaron, all three pluge bars should not be very visible. If they are, your set to bright. This is what pros use to calibrate tv studio moniters and cameras. Since your at it, what color is your yellow bar? Does it have any green or orange tint to it? When it is pure yellow, your sc phase(hue) is set correctly.

[aaronp]

Anthony, I'm not in front of my monitor right now (it's at home, i'm at my real job) so I can't say what color the yellow bar is. My brightness scale is turned down all the way, and I have started to turn the individual colors down just to get down to the 90 nits. It seems crazy to me that my monitor would be inherently that much too bright. When I get home, i'll try to see what it takes to get down to where only the 11.5 pluge is visible. Something still isn't adding up to me though, is it possible my monitor is messed up?

[anthony johns]

Ya, Something must be set wrong in windows then. When it comes to graphics cards and adobe and software and things like that, I'm no expert.

[RaymondC]

Yeah I am a newbie in color mgmt but I would assume brightness and stuff are taken care with the calibrator. The one I use allow you to see the before and after effects.

 

If you got a printer try printing at home and see how close they match up. Home printers come with ICC files that are pretty good.

 

Does the lab that you use provide custom ICC to be downloaded? Maybe try a different lab which allow this. Then see for yourself how they match up. It may be with your current lab you need to calibrate their printer and get a ICC.

[arc1]

Ambient lighting conditions can also change the way an image looks on your screen.

Preferrably, view images in a dark room without direct sunlight (close curtains, etc). Also,

try converting your document to Adobe 1998 (edit>color settings) color profile and soft

proof your image (View>proof colors). In your monitor profile, is your white point set to

6500k?

If you took your files to a professional lab, they should give you the option of getting a

proof print made. I would go back to one of the labs you used and ask them to make a

test print for you (a stop or 1/2 stop or whatever darker) and any other color corrections

you want made from your original print.

There will always be differences between your monitor and a lab monitor. Calibration only

works within YOUR workflow. Sending a file to another printer, using different papers and

different printers, the photograph will be interpreted differently. You could also download

the profile for the printer and paper the lab uses to see more accurately what your print

wil look like when outputted.

That's all. I know it's a lot. There are so many variables when sending your images out.

Hope some of this helps.

[anthony johns]

Aaron, I just got an idea. Send the color bars to the lab to print and see what comes out!

[frans_waterlander]

Whether or not you should calibrate your monitor to 6500K totally depends on the color temperature of your digital darkroom lighting. Ideally the color temperature of your monitor and digital darkroom lighting should be the same for the best possible monitor-to-print matching. Your digital darkroom lighting should furthermore be of the highest quality possible which in my honest opinion means you need low-voltage tungsten halogen lights like those from SoLux.

[Tim_Lookingbill]

Anthony,

 

Where did you get the color bar target? I've been looking all over

for that on the web.Thanks for posing it.

 

Just some questions about that target. Why are all the color

purity's diminished in luminance to 193 RGB combo's? Why isn't

it at 255 like most color purity targets?

 

Another thing, what is sc phase and how do you adjust it on a

CRT? The yellow bar on my CRT is very dim as if a gray filter is

placed over it. In fact the entire target looks dim. Does this have

something to do with compensating TV broadcast's pre-amped

gamma? I imagine this target is so dim because my TV is so

much brighter than my CRT. I'm just guessing, though.

[aaronp]

Annie,

 

Does it make a difference if I soft proof in sRGB instead of Adobe 1998? I do have Mpix.com's ICC profile for the Kodak paper they use, but soft proofing with it primarily seems to change the way the colors are represented on my monitor and and not the overall density/brightness, though on a couple of images with warmer colors soft proofing seems to brighten what I see, which gets me even further away from the print. The local "professional" labs don't have any sort of profile I can use (one even said that they didn't think their lightjet was able to be profiled!) which is one reason I decided to try a trusted online company. I don't think that the differnce between my monitor and the lab monitor is an issue, because I asked for the prints to be made with no corrections, so if everyone is calibrated and profiled, then it seems like in my simple mind that it should be wysiwyg (what you see is what you get).

 

Frans, my white point is set to 6500K. I have no idea what my room lighting temp is, or even how to measure that. Would the color temp change the overall brighness of appearance, or just the color shift of the print? I'm fine with the colors I get, I just want them brighter.

 

Anthony, that is a good idea. I have a pretty good idea of what it would look like, but I might do it anyway.

 

Thanks again all for your time and effort.

[frans_waterlander]

Aaron,

 

If you are happy with the colors, don't change your monitor color temperature setting, but you may want to play with it anyway to see if things improve.

 

You can measure your lighting color temperature with an expensive meter, around $1000 or rent one for less than $50 for a couple of days. Standard incandescent light bulbs are around 2000-2500K and special photo flood incandescents are about 3000-3200K. Fluorescents are all over the place but rarely go beyond 5000K and contain many horrendous spikes at different colors. As you can see, all of those have a color temperature that is a far cry from you monitor's 6500K. A critical eye will clearly see the difference where the print seems to reddish as compared to the image on the monitor. If you want to learn more about digital darkroom lighting you may want to read an article that I wrote on this subject. You can find it at: www.solux.net/ies_files/Digital%20Darkroom%20Lighting.pdf

[Tim_Lookingbill]

Aaron,

 

I just downloaded the Mpix E-surface profile off their website and

assigned to an sRGB color target I use to test drift printing to my

local Noritsu minilab on glossy Kodak Royal Digital Paper.

 

The Mpix profile has a noticeably reduced gamut compared to

my Noritsu and AveragePhotoPrinterRGB. Also its density is

much lighter and nowhere close to 2.2 gamma. The farther away

the printer is from sRGB the more color errors will result whether

you convert to this space or edit to compensate through Soft

Proofing.

 

This is what a profile is designed to do when you convert to it,

however you can't make a silk purse from a pig's ear and no

editing or converting will do the same in regards to printer

gamut.

 

If this Mpix E-surface is the same profile you're using as

mentioned in your original post, I would find you another printer.

[aaronp]

Frans, thanks for the info on the color temperature. I didn't realize there was such a difference between room lighting and monitor temperature. If (hopefuly when) I get a room devoted as my darkroom I will definitely try to incorporate that.

 

Tim, yes, I used the Mpix E-surface profile, however they print it on a matt surface instead of a glossy. I don't know how big a difference the gamut is between the two. With the profile gamut being less than sRGB, I guess it makes sense that some of my images seem to lighten when soft proofing since I have my intent set on perceptual. When you say its density is lighter, that means that the print should be lighter, correct? If so, that gets me further away from my desired output as my prints are already too dark. Do you have any recommendations for an online printer that would give better results?

 

Thanks again, guys.

[aaronp]

ps - Tim, have you been dam sliding over in Ingram lately?

[aaronp]

pps - just so you won't think I'm a stalker, I noticed on your bio you were from Kerrville; I've got some family that lives around there, and I visted them in the summer once.

[Tim_Lookingbill]

Dam sliding...mmh sounds fun.

 

I'm afraid I haven't tried that, yet. I see others over at the Ingram

dam doing just that but wasn't sure I could brave scraping up my

legs and hands.

 

Found a great watering hole in Hunt next to the La Hunta

recreation camp. Crystal clear water where you can see the fish.

 

To bad Kerrville's section of the Guadalupe is so murky green

from algae bloom due to cattle fecal run-off from ranches up

stream.