Ad
Ad
Ad
Pages: [1] 2 3 »   Bottom of Page
Print
Author Topic: prints dont match....  (Read 8231 times)
kerriann85
Newbie
*
Offline Offline

Posts: 20


« on: March 23, 2011, 08:41:19 PM »
ReplyReply

My monitor is supposed to be calibrated.
Here is a screen capture of the windows7 folder (not the previewer) next to the same file open in photoshop cs5.  To me the windows image looks very red. Sunburned if you will.
That is the way my image came back from the pro lab.  Both when I sent it without lab correction, then when I sent it with lab color correction. (for about double the price)

 It's also the way the print came back from Walmart when I requested them to not use their auto correction.
The photoshop print is closer, (must less red) to the way the print came back from Walmart when I let them print it the way they normally print (with correction).

I'm almost frustrated to tears.  I have a whole set of staff headshot prints from the lab and to me they all look like they are sunburned.  They don't look that way in photoshop,
I've read the "why are my prints too dark"
If it helps any, the values on her right cheek bone read: c-0, m-19, y-16, k-0  and R254, G212, B200

I'm just about frustrated to tears.  Can someone tell me something helpful?
Logged
Iliah
Sr. Member
****
Offline Offline

Posts: 415


« Reply #1 on: March 23, 2011, 09:13:28 PM »
ReplyReply

Some questions:
1. What colour space is the image?
2. Is the colour space embedded into the image?
3. What colour space the lab expects?
4. How your monitor profile is set in the system and in Photoshop?
Logged
Luca Ragogna
Full Member
***
Offline Offline

Posts: 163



WWW
« Reply #2 on: March 23, 2011, 09:29:27 PM »
ReplyReply

If the uncorrected print and corrected print from a pro lab look the same I would say that they're charging you for nothing.

Any chance you could post the actual JPEG? Those RGB numbers seem wack, those aren't the numbers I get when I check your screen grab and I tried to match those numbers with a quick curves adjustment and the shot ends up REALLY red and blown out. I'm leaning towards the conclusion that your monitor isn't really calibrated. What are you using to calibrate?

Also, make sure you're sending sRGB Jpegs to the labs, they do a crap job of converting files.
Logged
Czornyj
Sr. Member
****
Offline Offline

Posts: 1436



WWW
« Reply #3 on: March 24, 2011, 03:46:53 AM »
ReplyReply

You have a wide gamut monitor, with a color space volume similar to AdobeRGB. The picture is rendered to sRGB editing color space, that is much smaller than the color space of your display. The display is calibrated and profiled, so Photoshop reads the display profile that is describing the color space of the display, and manages the colors. The picture viewer doesn't seem to manage the colors of the picture, so that's why they're over-saturated. You need an image browser with color management module to get proper sRGB image colors on wide gamut display.

If prints from the lab are still less colorful than in Photoshop, you should try to:
-calibrate the display to a lower luminance level
-increase the display to a higher blackpoint brightness
-illuminate the print with more daylight
-softproof the image with the ICC profile of the lab+paper, and paper color simulation applied
« Last Edit: March 24, 2011, 07:17:56 AM by Czornyj » Logged

digitaldog
Sr. Member
****
Online Online

Posts: 9222



WWW
« Reply #4 on: March 24, 2011, 09:07:27 AM »
ReplyReply

My monitor is supposed to be calibrated.

Calibrated how is the $64K question. And can you soft proof the images using the lab’s provided ICC profile?
See: http://www.luminous-landscape.com/tutorials/why_are_my_prints_too_dark.shtml
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
Iliah
Sr. Member
****
Offline Offline

Posts: 415


« Reply #5 on: March 24, 2011, 09:47:06 AM »
ReplyReply

> Calibrated how is the $64K question.

Well, if delta EV between (L,a,b) (100,0,0) and (50,0,0) in total darkness is about 2 to 2.5 while (L,a,b) (100,0,0) is measured 8 to 10 EV than brightness is OK. Colour temperature and grey balance can be verified shooting grey patches of varying L displayed on the screen in Photoshop, out of focus, 1/15 of a second or slower, and checking colour temperature and uniformity of white balance for shots of different L in a raw converter. Only central portion of the patches should be used.

IMHO the issue at hand is more about colour management consistency.

L(cd/m^2)=2^(EV-3)
EV=log2(L)+3

For ISO =100:
EV = 2*log2(f) - log2(s); f is aperture number, s - exposure time in seconds.
For ISO 100, f/8, s=1/8 sec:
EV=2*log2(Cool - log2(1/8) = 2*3 - (-3) = 9

Colorimeters and spectrophotometers are not always great at measuring light flux, while exposure meters are designed to do just that.
Logged
kerriann85
Newbie
*
Offline Offline

Posts: 20


« Reply #6 on: March 24, 2011, 07:15:21 PM »
ReplyReply

Yes, I was totally amazed when the lab corrected prints came back. 
Here is the original .jpg
How do all of you people feel about the color?  To me she's a nice "peachy" Caucasian.  Not sunburned.
Logged
Iliah
Sr. Member
****
Offline Offline

Posts: 415


« Reply #7 on: March 24, 2011, 07:24:34 PM »
ReplyReply

The colour is hypersaturated and on magenta side. So it is with eyedropper in Photoshop, so it looks on my monitor.
Logged
kerriann85
Newbie
*
Offline Offline

Posts: 20


« Reply #8 on: March 24, 2011, 07:28:58 PM »
ReplyReply

Some questions:
1. What colour space is the image?
2. Is the colour space embedded into the image?
3. What colour space the lab expects?

sRGB as far as I know....   I'm going to do some more digging around

4. How your monitor profile is set in the system and in Photoshop?

that's what I'm not totally sure about.  I'm going to try some screen captures to see if things look like they should to you guys.
Logged
kerriann85
Newbie
*
Offline Offline

Posts: 20


« Reply #9 on: March 24, 2011, 07:32:27 PM »
ReplyReply

The colour is hypersaturated and on magenta side. So it is with eyedropper in Photoshop, so it looks on my monitor.

okay, that sounds like the description of the print.  And the description of the picture on the left in the first
post above.  What I see if Photoshop looks fine, like the image on the right in the first post.

So.... if the picture in photoshop doesn't look hypersaturated and magenta but comes back from the
lab that way, that's what I'm hoping to fix!!
Logged
kerriann85
Newbie
*
Offline Offline

Posts: 20


« Reply #10 on: March 24, 2011, 07:43:55 PM »
ReplyReply

Maybe someone can see something here that looks wrong in Photoshop?
Ill go get my calibration settings next.
Logged
kerriann85
Newbie
*
Offline Offline

Posts: 20


« Reply #11 on: March 24, 2011, 07:47:48 PM »
ReplyReply

And here is how my calibration settings look.   

Logged
kerriann85
Newbie
*
Offline Offline

Posts: 20


« Reply #12 on: March 24, 2011, 08:05:33 PM »
ReplyReply

You have a wide gamut monitor, with a color space volume similar to AdobeRGB. The picture is rendered to sRGB editing color space, that is much smaller than the color space of your display. The display is calibrated and profiled, so Photoshop reads the display profile that is describing the color space of the display, and manages the colors. The picture viewer doesn't seem to manage the colors of the picture, so that's why they're over-saturated. You need an image browser with color management module to get proper sRGB image colors on wide gamut display.

If prints from the lab are still less colorful than in Photoshop, you should try to:
-calibrate the display to a lower luminance level
-increase the display to a higher blackpoint brightness
-illuminate the print with more daylight
-softproof the image with the ICC profile of the lab+paper, and paper color simulation applied


Well, the problem is just the oposite. The lab prints are too red and saturated. So Ineed a higher luminance lever and lower the blackpoint brightness?
I'm not really sure about softproofing.  Never done it (we don't do it at the photo studio I work for)  Do I just need to ask the lab for their ICC profile?
Logged
Iliah
Sr. Member
****
Offline Offline

Posts: 415


« Reply #13 on: March 24, 2011, 09:27:41 PM »
ReplyReply

It would be good if you can upload your monitor profile for evaluation. Also, please make sure that Photoshop uses the monitor profile.
Logged
kerriann85
Newbie
*
Offline Offline

Posts: 20


« Reply #14 on: March 25, 2011, 08:40:10 AM »
ReplyReply

It would be good if you can upload your monitor profile for evaluation. Also, please make sure that Photoshop uses the monitor profile.

Iliah, Thanks so much for trying to help me.  How do I find my monitor profile and how do I make sure that PS uses the monitor profile?
Logged
Iliah
Sr. Member
****
Offline Offline

Posts: 415


« Reply #15 on: March 25, 2011, 12:07:29 PM »
ReplyReply

You can start with "Color Settings" in Photoshop (Shift-Ctrl-K). In the RGB drop-down you should see the name of your monitor profile on the fifth line from the top. The line should read "Monitor RGB - <the_name_of_your_profile". Close the dialogue with "Cancel".
Logged
Alan Goldhammer
Sr. Member
****
Offline Offline

Posts: 1710


WWW
« Reply #16 on: March 25, 2011, 12:26:40 PM »
ReplyReply

Iliah, Thanks so much for trying to help me.  How do I find my monitor profile and how do I make sure that PS uses the monitor profile?
If you are on a Windows machine, you can access your profile through the Control Panel.  Open it up and click on  color management The dialogue box will have three tabs: Devices, All Profiles, and Advanced.  The Devices tab will show all the profiles associated with that device.  Since you are using Spectraview you should show a profile name that has your printer ID and it will also show you when you calibrated it and the rough settings.  I have a NEC P221 so mine reads:  P221W 92100475NA 2011-03-05 18-28 D65 2.20; yours will be different.  The checkbox 'Use my settings for this device' should be checked (I'm pretty sure that Spectraview automatically sets this up for you).  I don't understand Iliah's point; if you have properly calibrated your monitor, Photoshop will use that calibration.  What you can change is the colorspace to work in.  Since I do most of my work in Lightroom with Prophoto, I have PS set up to use the Prophoto space as well.

One observation on your calibration setting; you are using the default monitor contrast setting which is probably way to high as inkjet printers cannot render that much contrast in an image.  You may want to change this the next time you calibrate to something like 350:1.  This, however, should not be why you are seeing the color problems in the prints you get back.
Logged

Luca Ragogna
Full Member
***
Offline Offline

Posts: 163



WWW
« Reply #17 on: March 25, 2011, 12:40:35 PM »
ReplyReply

I spent 5 minutes colour correcting your file on my laptop. The monitor is calibrated and I find it's not that far off, but keep in mind it's still a laptop and I don't usually use it for colour critical work. I've attached a levels correction for that file. The highlights blow out a bit but I think the colour is much closer to where you want it (probably still not perfect). Try it out and see how it looks on your monitor.

Try changing your white point setting to 6500 K. 5000K tend to give everything a yellow cast.

Another thing to consider is that your calibrator might be wonky. It's not unheard of. Pantone replaced a ton of Hueys because the early ones were defective.
Logged
Iliah
Sr. Member
****
Offline Offline

Posts: 415


« Reply #18 on: March 25, 2011, 12:43:02 PM »
ReplyReply

Because it is a routine to check Photoshop settings. If those are OK no point in going to Control Panel. But if they are not - troubleshooting may be a bit different a quite a bit more complicated than you just described.

If the profile is in fact used by Photoshop a closer look into profile is in order. That does not involve Control Panel.
Logged
Iliah
Sr. Member
****
Offline Offline

Posts: 415


« Reply #19 on: March 25, 2011, 12:43:36 PM »
ReplyReply

It is about 13% green filter off.
Logged
Pages: [1] 2 3 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad