Luminous Landscape Forum
Raw & Post Processing, Printing => Colour Management => Topic started by: Jessica Holden on July 07, 2010, 03:18:13 am
-
Newbie question here, but I have been spinning my wheels on this for a week and an completely overwhelmed at this point...
I have always successfully calibrated my (very old) monitor with my i1D2. But I got a new computer and a new monitor (an HP ZR24w with a ATI Radeon HD5770 video card), and I can't get the new monitor to calibrate correctly. (I used targets 6500, 2.2, 110, and at the end of calibration they were spot on, and the red green and blue lines looked smooth and together, but the OSD settings in the red green and blue channels were red 88, green 50, and blue 66.) I edited a few photos on it and they looked GREAT on-screen, but when I got the test prints back from WHCC they were VERY cool.
So I ordered a new calibrator since I'd read the pucks can go bad and under-read the red channel, but the new one arrived today, and OF COURSE that was too simple an answer, because it did all the same things my old i1D2 did.
So where it stands right now, I have my monitor back on the default profile (in my Color Management window, for the monitor, I have it set back to the "default HP ZR24w LCD Monitor" profile, called HP_zr24w.icm), but the OSD is at the settings I set last time I calibrated (I used Target 6500, Brightness 110, and Gamma 2.2, and ended up with brightness 29, contrast 99, and RGB at 72/30/47). But that was still too red, though, so I gave in and just manually eyeballed the RGB channels in the OSD with my photos open in CS5, and when I move the red slider down to 56, green at 30, and blue to 52, and it's pretty darned spot on to those photos. So that's where my OSD settings are now.
I am completey overwhelmed and realize I don't even understand what a profile IS and how it's important, if I can adjust the monitor by hand and have it look good, when calibrating makes it look bad.
I have until Thursday morning to send my monitor back, but is that the right thing to do? Or is the fact that it "looks" right now per my prints, should I leave it? It's a beautiful monitor, now that the colors match.
Every time I take a step, I realize I am in WAY over my head. I would love to hire a consultant to help me, but don't even know what to look up in a phonebook to do that, PLUS I am very nearly out of time if I DO need to return the monitor. So in my tons of research I came upon you guys several times, and decided to post and see if any of you brilliant folks have any suggestions?
I really want to do this right!!!!
-
There is some discussion whether the existing colori meters/software can cope with the extended gamuts of monitors like the HP ZR24W.
Whether that is what you face or something else I do not know but sending the monitor back may not be right solution.
I switched to my spectrometer for calibration/profiling of my wide gamut monitor
Google with: colorimeter correction matrix for wide gamut monitors
Edit: I checked before writing the above but made a mistake: the HP ZR24W isn't a wide gamut model so not the right answer to the issue.
met vriendelijke groeten, Ernst Dinkla
Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/ (http://groups.yahoo.com/group/Wide_Inkjet_Printers/)
-
I have always successfully calibrated my (very old) monitor with my i1D2. But I got a new computer and a new monitor (an HP ZR24w with a ATI Radeon HD5770 video card), and I can't get the new monitor to calibrate correctly. (I used targets 6500, 2.2, 110, and at the end of calibration they were spot on, and the red green and blue lines looked smooth and together, but the OSD settings in the red green and blue channels were red 88, green 50, and blue 66.) I edited a few photos on it and they looked GREAT on-screen, but when I got the test prints back from WHCC they were VERY cool.
New display, that doesn’t mean the old calibration target aim points should be the same. And did WHCC provide a profile for soft proofing and allow you to use it to convert to the output space? If you have a soft proof profile, and the print to screen match are off, you may need to alter the calibration of the display to produce a closer visual match. How are you viewing the prints next to the display?
-
Jessica,
if you have a moment, go back on this sub forum to several threads recently discussing 10 bit output displays.
The discussion has hovered around whether the aftermarket calibration solutions commonly available are suited for displays like yours or my lp2475w.
They're both wide gamut displays, and the colorimeter you have seems to be the common one.
The displays apparently go more luminescent or something, than that puck is designed to read.
So I understand filters or software are solutions to the misread the puck makes.
My solution is a coke can against a green piece of velvet.
Or the berry brambles and fruits out my winder.
The one thing is the uniformity across the display.
Pure white screen, that's the test for me.
I had green dragons fighting in a dry pink fog on my first one.
-
Jessica,
The discussion has hovered around whether the aftermarket calibration solutions commonly available are suited for displays like yours or my lp2475w.
Based on what Jessica is saying, I doubt its an issue with the Colorimeter. Even Colorimeters with sRGB expectant filters can work on an wide gamut display. Adjustments to the target calibration aim points may need to be made. There is a disconnect between what she see’s and what the lab is providing. It could be the display but it could be the lab, or a bit of both. If Jessica has a desktop printer, we could rule this out assuming she’s got a decent output profile for this printer.
-
I was under the impression when I bought this display that it was NOT a wide-gamut display? I tried to NOT purchase that as I thought it might be overkill for my needs (as a portrait photographer who might one day go pro). Here are the specs:
Color Gamut Area vs. NTSC: 83% (CIE 1976), 72% (CIE 1931)
Color Gamut Coverage of sRGB: 97% (CIE 1931)
I read that it's not 10-bit either--here are the specs:
HP ZR24w specs (http://h18004.www1.hp.com/products/quickspecs/13557_na/13557_na.html)
I am such a novice!
I do have a desktop printer (just an injjet), but I have not yet gotten it set up with icc profiles or anything--I was trying to straighten this out first.
My most immediate issue is whether I should return this monitor today, but my fear is that if there IS nothing wrong with it, I might even be exchaning a good one for a bad one. To my eyems, the monitor LOOKS good, as long as I manually adjust the red and blue channels.
My screen DOES look good (uniformly white) when pure white, and when I adjust the red green and blue, the colors appear correctly. But if I can't calibrate my monitor, am I going backwards?
Thanks so much for all your help already--this is a wonderful forum and I am glad I stumbled across it!
-
I do have a desktop printer (just an injjet), but I have not yet gotten it set up with icc profiles or anything--I was trying to straighten this out first.
What’s the printer?
My most immediate issue is whether I should return this monitor today, but my fear is that if there IS nothing wrong with it, I might even be exchaning a good one for a bad one. To my eyem, the monitor LOOKS good, as long as I manually adjust the red and blue channels.
Stop worrying about the curves for the time being.
See if you can setup the printer, hopefully you have a few sheets of decent paper and a matching profile. Output the image you sent to your lab. Are they even remotely close? Is one closer to the display?
If you do decide to return the display, I think you could do better with one of the NEC SpectraView line where you simply tell it want target calibration you want, hit one button and you’re done. No OSD to play with thankfully, the host software and high bit panel all do the work.
-
OK--I just printed a photo from my inkjet (Canon iP4700), and it looks pretty close to the WHCC prints (and to my monitor, now that I have manually adjusted the red and blue channels). Maybe slightly more yellow and slightly less pink.
I used "Printer manages colors" on CS5, and document profile sRGB. The display as adjusted is closer to the WHCC prints, but the home prints are not bad.
ETA--I am leaning toward keeping this monitor--aside from being thus far unable to calibrate it with my i1D2, it LOOKS fantastic. But am I shooting myself in the foot here? I am accustomed to being able to say with pride, "I'm calibrated."
-
OK--I just printed a photo from my inkjet (Canon iP4700), and it looks pretty close to the WHCC prints (and to my monitor, now that I have manually adjusted the red and blue channels). Maybe slightly more yellow and slightly less pink.
I used "Printer manages colors" on CS5, and document profile sRGB. The display as adjusted is closer to the WHCC prints, but the home prints are not bad.
Its a good (better) sign. Ideally you would use Photoshop Manages Color and select a profile. And you would first setup a soft proof using that profile, pick the rendering intent you prefer. Then turn on the Simulate check boxes in the soft proof (it makes your image look like crap but you are now seeing the contrast ratio correctly). Hit tab key, F key until you are in full screen mode (no palettes, no menus, just image). Now compare the display to the print (the viewing conditions of which you haven’t defined and are important!). You need to adjust white point and luminance of the display calibration to result in a visual match (the numbers themselves are rather meaningless):
(http://digitaldog.net/files/Print_to_Screen_Matching.jpg)
-
Its a good (better) sign. Ideally you would use Photoshop Manages Color and select a profile. And you would first setup a soft proof using that profile, pick the rendering intent you prefer. Then turn on the Simulate check boxes in the soft proof (it makes your image look like crap but you are now seeing the contrast ratio correctly). Hit tab key, F key until you are in full screen mode (no palettes, no menus, just image). Now compare the display to the print (the viewing conditions of which you haven’t defined and are important!). You need to adjust white point and luminance of the display calibration to result in a visual match (the numbers themselves are rather meaningless):
Oh my gosh, I meant to say that--I DID use Photoshop measures colors, and I selected the sRGB profile.
Off to follow the rest of your instructions...
-
OK, so I am able to do that in soft proof (adjust what I am seeing in the soft proof on-screen to match my print using my OSD).
Is that "good enough" to consider myself "calibrated"? And I should keep my monitor?
My viewing conditions are a well-lit room. Natural light from a big north-facing window and a smaller west-facing window. My desk faces the north window but I have the monitor aimed with it's back to the edge of the window, mostly against wall. I do 90% of my work during the daylight.
-
Oh my gosh, I meant to say that--I DID use Photoshop measures colors, and I selected the sRGB profile.
It should be Photoshop Manages Color. Then you select the paper profile NOT sRGB from the popup menu (Printer Profile).
-
Is that "good enough" to consider myself "calibrated"? And I should keep my monitor?
Do you have a good visual match?
My viewing conditions are a well-lit room. Natural light from a big north-facing window and a smaller west-facing window. My desk faces the north window but I have the monitor aimed with it's back to the edge of the window, mostly against wall. I do 90% of my work during the daylight.
Not ideal. The daylight changes all day (and when the weather changes). You need something controllable like the illustration above, or at the very least, a half decent desk lamp that doesn’t spill onto the display. Something like this:
https://www.solux.net/cgi-bin/tlistore/clampon.html (https://www.solux.net/cgi-bin/tlistore/clampon.html)
-
HP ZR24w specs (http://h18004.www1.hp.com/products/quickspecs/13557_na/13557_na.html)
Sorry, although I checked before writing the reply it I made a mistake there, it isn't a wide gamut one.
Ernst Dinkla
-
Its a good (better) sign. Ideally you would use Photoshop Manages Color and select a profile. And you would first setup a soft proof using that profile, pick the rendering intent you prefer. Then turn on the Simulate check boxes in the soft proof (it makes your image look like crap but you are now seeing the contrast ratio correctly). Hit tab key, F key until you are in full screen mode (no palettes, no menus, just image). Now compare the display to the print (the viewing conditions of which you haven’t defined and are important!). You need to adjust white point and luminance of the display calibration to result in a visual match (the numbers themselves are rather meaningless):
I am working on this now. I am using Photoshop Manages Color, and then there are a few dozen profiles listed, many of which are listed specifically for my printer. I assume I should be choosing one for my printer; should I select the one that I find most pleasing to the eye, and then soft proof to that one?
And my second question is that during soft proofing I only seem to be able to select EITHER the Simulate Paper Color check box OR the Simulate Black Ink box; which should I choose? (And you're right, they look like crap--LOL!)
Then (I want to be sure I understand completely), I print using that same chosen profile, and soft proof that print against the full-sized image on my screen (as large as I can get the photo on my monitor, with no extraneous windows/menus in the way). I then adjust my screen to match that print using my RGB controls and the brightness controls on my OSD.
Right? I can assume I am "calibrated", then, as well as I can be with my current equipment?
-
Jessica, you select the profile that is named after or best suited to the paper you are using. Print with that, and also select that when you are soft proofing in Photoshop.
-
I assume I should be choosing one for my printer; should I select the one that I find most pleasing to the eye, and then soft proof to that one?
Nope. You select the profile based on the paper. The profile names should be specific (such as Epson 3880 Luster etc).
And my second question is that during soft proofing I only seem to be able to select EITHER the Simulate Paper Color check box OR the Simulate Black Ink box; which should I choose? (And you're right, they look like crap--LOL!)
Both should be on.
Then (I want to be sure I understand completely), I print using that same chosen profile, and soft proof that print against the full-sized image on my screen (as large as I can get the photo on my monitor, with no extraneous windows/menus in the way). I then adjust my screen to match that print using my RGB controls and the brightness controls on my OSD.
You examine to see if the current calibration target values produce a match to the soft proof. If the print looks much darker than the display, you need to lower the cd/m2 target. If the image on screen looks too cool, you lower the white point value.
-
You examine to see if the current calibration target values produce a match to the soft proof. If the print looks much darker than the display, you need to lower the cd/m2 target. If the image on screen looks too cool, you lower the white point value.
Would this be based on adjusting in the OSD for the monitor, or by trying to get the i1D2 to calibrate to that soft proof?
-
Would this be based on adjusting in the OSD for the monitor, or by trying to get the i1D2 to calibrate to that soft proof?
By adjusting the target calibration aim points in the software you use to produce the calibration and profile.
-
I was under the impression when I bought this display that it was NOT a wide-gamut display? I tried to NOT purchase that as I thought it might be overkill for my needs (as a portrait photographer who might one day go pro). Here are the specs:
Color Gamut Area vs. NTSC: 83% (CIE 1976), 72% (CIE 1931)
Color Gamut Coverage of sRGB: 97% (CIE 1931)
Yeah, it is not a wide gamut. It gets confusing since some makers tag a "w" at the end to mean wide format (16:10 or 16:9) and some to mean wide gamut.
What format image to did you send out to print? Some printing places only accept sRGB and if you send AdobeRGB or ProPhotoRGB you will get nasty results back.
As others have already said you want to set, on the printing page, Photoshop Manages Colors, select the profile that matches whatever paper you are currently using, also very important is to go to the advanced tab in printer setup and make sure to select ICM management and then pick the "Off" option (otherwise either the OS or printer will manage on top of what photoshop does and you get a real mess). In the epson driver setup you also want to select the paper type there and other high quality options such as PhotoRPM or PhotoBest or something like that.
-
So where it stands right now, I have my monitor back on the default profile (in my Color Management window, for the monitor, I have it set back to the "default HP ZR24w LCD Monitor" profile, called HP_zr24w.icm), but the OSD is at the settings I set last time I calibrated (I used Target 6500, Brightness 110, and Gamma 2.2, and ended up with brightness 29, contrast 99, and RGB at 72/30/47). But that was still too red, though, so I gave in and just manually eyeballed the RGB channels in the OSD with my photos open in CS5, and when I move the red slider down to 56, green at 30, and blue to 52, and it's pretty darned spot on to those photos. So that's where my OSD settings are now.
Jessica, besides all the other discussion going on... I'm appalled that you had to move the RGB controls so far from their default 255-255-255 settings. Have you tried leaving them at their native default settings and allowing the software to do the work of adjusting the white balance?
-
Jessica, besides all the other discussion going on... I'm appalled that you had to move the RGB controls so far from their default 255-255-255 settings.
ETA - see post 25 for an update to my thoughts with regards to the rgb values, I'm not entirely on the ball tonight
Why? 255, 255, 255 represents the "native" white point. The native white point of the screen and the white point of your print more often than not have absolutely nothing to do with one another. In addition, the numbers that colorimters return may as well be arbitrary values. They are not lab grade devices and the variability from instrument to instrument can easily be 1800ºK. This presents a problem because when you enter a value for the color temperature you're going to have issues if you expect your screen to actually be that color temp (even though the puck says it is) when all is done. Colorimeters are good for relative profiling... if the colors are out of whack, it brings them back in. If you want an accurate screen to print match and you're using a proper full spectrum bulb and an ICC profile from your lab for soft proofing, then you're going to need to adjust the color temp of your screen to match up with the light source/paper white combo. The color temp of the light source you use will have an impact on what color temp your screen should be at.
Cheers, Joe
-
Why? 255, 255, 255 represents the "native" white point. The native white point of the screen and the white point of your print more often than not have absolutely nothing to do with one another. In addition, the numbers that colorimters return may as well be arbitrary values. They are not lab grade devices and the variability from instrument to instrument can easily be 1800ºK. This presents a problem because when you enter a value for the color temperature you're going to have issues if you expect your screen to actually be that color temp (even though the puck says it is) when all is done. Colorimeters are good for relative profiling... if the colors are out of whack, it brings them back in. If you want an accurate screen to print match and you're using a proper full spectrum bulb and an ICC profile from your lab for soft proofing, then you're going to need to adjust the color temp of your screen to match up with the light source/paper white combo. The color temp of the light source you use will have an impact on what color temp your screen should be at.
Cheers, Joe
a good colorimeter should be able to do an order of magnitude better than 1800K off
so i would definitely worry about being 2000K off
but it is true that you really can't get thing down to the .001 either
-
a good colorimeter should be able to do an order of magnitude better than 1800K off
so i would definitely worry about being 2000K off
but it is true that you really can't get thing down to the .001 either
I should probably quantify that a little more but let me assure you that these are real values from real colorimeters and they can be that far off from one another.
Spyder 3s from what I've observed tend to report a measured value as being cooler than it actually is. DTP94s by comparison report a measured value as being warmer than it actually is. Eye Ones are somewhere in the middle. If profile my monitor to say D65 with my i1 Pro and the measure the same white point with the Spyder 3 and the DTP94 I will get values of 7400ºK and 5600ºK respectively. The delta is +/- 900ºK in what these pucks are reporting back. They are not to be relied upon for accurate profiling of white point.
ETA - Now these are the differences I've noticed from one manufacturer to another. If you compare the same device to the same device you won't see quite that delta but I have seen 400ºK differences between two devices of the same make and model.
Cheers, Joe
-
Jessica, besides all the other discussion going on... I'm appalled that you had to move the RGB controls so far from their default 255-255-255 settings. Have you tried leaving them at their native default settings and allowing the software to do the work of adjusting the white balance?
Okay...
I gotta say though... yeah, those numbers do seem way off. Sorry... I'm a bit frazzled today. I have about 5 things exploding in my face at the moment so I'm lacking a bit of clarity. I suspect the luminance on the screen is too bright for the target value. Ideally one of those channels should be at 255 and one or two of them should be tweaked to get to the desired color temp. My apologies, you are correct. Something is off.
Cheers, Joe
-
I should probably quantify that a little more but let me assure you that these are real values from real colorimeters and they can be that far off from one another.
Spyder 3s from what I've observed tend to report a measured value as being cooler than it actually is. DTP94s by comparison report a measured value as being warmer than it actually is. Eye Ones are somewhere in the middle. If profile my monitor to say D65 with my i1 Pro and the measure the same white point with the Spyder 3 and the DTP94 I will get values of 7400ºK and 5600ºK respectively. The delta is +/- 900ºK in what these pucks are reporting back. They are not to be relied upon for accurate profiling of white point.
ETA - Now these are the differences I've noticed from one manufacturer to another. If you compare the same device to the same device you won't see quite that delta but I have seen 400ºK differences between two devices of the same make and model.
Cheers, Joe
Well I have heard that as many as a third of a spyder 3s can be like 15dE or more off.
I'm surprised the DTP94b and Eye1Pro were so far apart. I haven't heard of a huge 900K difference reported elsewhere.
Both DTP actually make my HDTV look a bit redder then when using an I1D2 which would mean they are reading cooler than the I1D2.
The two DTP94 that I tried only reported about 60-80K difference in color temp on an sRGB monitor (for some reason they were more like 200-300k off using icolor generic matrix on a wide gamut though).
I used to have a spyder2 that tended to make everything look red.
Did you try them on perhaps LED backlit panels?
(or wide gamut?)
anyway, interesting, i will have to look into it more
my NEC PA (suing NEC puck calibrated sRGB mode or factory sRGB mode) with either metamerism on or off doesn't really look the same in the grays as my DTP94 calibrated samsung monitor
-
I agree with Ernst. Try a Color Munki or other spectro.
David
-
Did you try them on perhaps LED backlit panels?
(or wide gamut?)
anyway, interesting
I've tried them on displays with white LED backlights, CCFL backlights, screens that are wide gamut, and screens that pretty much stick to the gamut of sRGB (and always after letting the monitors and the pucks warm up to a stable operating temp). Not only did the color temps vary but the reported luminance varied by as much as 15 cd/m^2 between instruments as well.
What I got out of it was, you can't rely upon a colorimeter to hit a quantifiable value. They're kind of like body fat monitors that are based on resistance. They aren't at all accurate at calculating body fat but used consistently, they do allow you to track relative changes. They do however allow you establish a reproducible standard relative to themselves and that is good enough for maintaining quality control. What you can't do is establish consistent control at remote sites with different pucks so... if you and I both had an NEC 2690 and we both calibrated/profiled to D65, gamma 2.2, @ 110 cd/m^2, we would end up with different results, especially if we had pucks from different manufacturers. Complicating matters more though is the fact that even different software will report back a different value for the SAME white point so there's lots of variables here.
If you want more consistency from instrument to instrument, and something that will get you closer to actual quantifiable values, in theory you would have to move to a spectrophotometer. I haven't as of yet had the opportunity to compare spectros to one another. I do have a Spectroscan/Spectrolino and an i1 Pro... one of these days I'll compare them to see how close they are. In an ideal world they'd render the same values but I know they do not do that as the Spectrolino is dramaticlly better in the shadows.
Cheers, Joe
-
I should probably quantify that a little more but let me assure you that these are real values from real colorimeters and they can be that far off from one another.
Spyder 3s from what I've observed tend to report a measured value as being cooler than it actually is. DTP94s by comparison report a measured value as being warmer than it actually is. Eye Ones are somewhere in the middle. If profile my monitor to say D65 with my i1 Pro and the measure the same white point with the Spyder 3 and the DTP94 I will get values of 7400ºK and 5600ºK respectively. The delta is +/- 900ºK in what these pucks are reporting back. They are not to be relied upon for accurate profiling of white point.
ETA - Now these are the differences I've noticed from one manufacturer to another. If you compare the same device to the same device you won't see quite that delta but I have seen 400ºK differences between two devices of the same make and model.
Cheers, Joe
I'm starting to think you may be right (sort of, I do seem to be noticing differences only I'm getting the DTP94b reading things as cooler than they actually may be and the other probes warmer, the exact opposite of what you wrote; EDIT: actually maybe not, it seems that it is perhaps making profiles end up with too much blue and not enough green, anyway whatever the exact details I think you are likely right about DTP94 perhaps not getting white point exactly on spec), at least about the differences in color temp various probes may read.
I just compared a DTP94b calibrated (but not profiled) sRGB CCFL LCD HDTV to a color checker chart (viewed during the day with windows open, but the sun not directly on it, so hopefully that is vaguely D65) and did the same for the set calibrated by the NEC custom puck and the same for NEC PA calibrated with NEC puck and using factory defaults.
The NEC (i1D2) calibrated HDTV looked way red, nothing more needs to be said.
The DTP94b calibrated Samsung CCFL LCD sRGB HDTV and Samsung sRGB monitor actually didn't seem to match quite as closely as the either the NEC PA with SVII+NEC puck or simply using the factory preset.
to be honest none of the displays really matched the checked when it was held up with each color side by side, even the NEC seemed noticeably off, but it did seem to be the closest overall and if the chart was held closeish but not too close it at least looked somewhat ballpark, the other displays needed the color checker held much farther away for things to begin to look ballpark.
Of course I really need to use a D65 lamp to be sure. (then again it seems that ones with a good spectral response cost rather a lot)
(of course you also mention that different software provides different values too and that was certainly an element in play here, very different software and calibration methods)