Ad
Ad
Ad
Pages: [1]   Bottom of Page
Print
Author Topic: Calibrating Wide Gamut Monitors  (Read 13917 times)
Etienne Cassar
Jr. Member
**
Offline Offline

Posts: 95


« on: September 15, 2008, 01:09:48 AM »
ReplyReply

I just purchased a NEC Spectraview 2690 monitor (EU version) and I am trying to calibrate it using the Gretagmacbeth i1 Display 2 and the Spectraview Harware and software calibration.  I am however not convinced with the results that I am getting.  The colours on the monitor still don't match my prints (I don't print myself, but I send my prints over to a lab).  I also tried to set up a grid in photoshop with exact colour patches that are on the Gretagmacbeth colour checker chart, but again the colours are off.  Somewhere on the internet I remember reading something from Andrew Rodney that there is a problem when calibrating wide gamut displays with present colorimeters, and that he himself was getting a white balance result that was 500K off.  Any ideas of how can I get around this?
I tried to alter the White point from Programmable mode to sRGB after doing a calibration and the colours look much more accurate.  But I am still not convinced that this is the way to go about it.
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 7126


WWW
« Reply #1 on: September 15, 2008, 06:51:54 AM »
ReplyReply

Quote
I just purchased a NEC Spectraview 2690 monitor (EU version) and I am trying to calibrate it using the Gretagmacbeth i1 Display 2 and the Spectraview Harware and software calibration.  I am however not convinced with the results that I am getting.  The colours on the monitor still don't match my prints (I don't print myself, but I send my prints over to a lab).  I also tried to set up a grid in photoshop with exact colour patches that are on the Gretagmacbeth colour checker chart, but again the colours are off.  Somewhere on the internet I remember reading something from Andrew Rodney that there is a problem when calibrating wide gamut displays with present colorimeters, and that he himself was getting a white balance result that was 500K off.  Any ideas of how can I get around this?
I tried to alter the White point from Programmable mode to sRGB after doing a calibration and the colours look much more accurate.  But I am still not convinced that this is the way to go about it.
[a href=\"index.php?act=findpost&pid=221505\"][{POST_SNAPBACK}][/a]

If you are sending the files to a lab, the first approach I would recommend is to get the lab's printer profile so that you can adjust your image under soft proof using that printer profile. While you have evidence that the display profile may not be giving a totally accurate result, it is possible that the larger error may be a workflow issue whereby your colour and luminosity adjustments don't take the printer profile into account.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
digitaldog
Sr. Member
****
Offline Offline

Posts: 9316



WWW
« Reply #2 on: September 15, 2008, 08:53:12 AM »
ReplyReply

Have to figure out if the issue is the display or the print. Without a print profile, and having proper lighting to evaluate the print near the display, there's no hope in getting a match OR figuring out what's not matching.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
Etienne Cassar
Jr. Member
**
Offline Offline

Posts: 95


« Reply #3 on: September 16, 2008, 01:14:49 AM »
ReplyReply

I am not using a pro lab for developing my prints, but they did supply me with a printer profile, one for the glossy prints and another for the matt prints.  I am trying to compare my prints with a soft proof image in photoshop using these profiles.  They also supplied what they call a cablibration print which you can use to match with the same photograph you download from their website.
I don't have a viewing booth that is true, but I can't afford now for now.  Any ideas of how I can get better viewing the cheaper way?
Another query is on the Spectraview software.  It has an option to do hardware calibration only and another to do both software and hardware calibration.  What is the difference between the two.  When doing just a harware calibration the resulting graph is a perfect straight line with all R G B channels right on top of each other and the prints seems to match my softproof images much more closely.  With both hard and software calibration I get quite a deviation between the R G and B channels of the graph.  Which one should I use?
Thanks for you help.

Etienne
« Last Edit: September 16, 2008, 01:15:06 AM by ecassar » Logged
Czornyj
Sr. Member
****
Offline Offline

Posts: 1446



WWW
« Reply #4 on: September 16, 2008, 04:09:34 AM »
ReplyReply

Quote
I am not using a pro lab for developing my prints, but they did supply me with a printer profile, one for the glossy prints and another for the matt prints.  I am trying to compare my prints with a soft proof image in photoshop using these profiles.  They also supplied what they call a cablibration print which you can use to match with the same photograph you download from their website.
I don't have a viewing booth that is true, but I can't afford now for now.  Any ideas of how I can get better viewing the cheaper way?
Another query is on the Spectraview software.  It has an option to do hardware calibration only and another to do both software and hardware calibration.  What is the difference between the two.  When doing just a harware calibration the resulting graph is a perfect straight line with all R G B channels right on top of each other and the prints seems to match my softproof images much more closely.  With both hard and software calibration I get quite a deviation between the R G and B channels of the graph.  Which one should I use?
Thanks for you help.

Etienne
[a href=\"index.php?act=findpost&pid=221684\"][{POST_SNAPBACK}][/a]

You can buy Solux 4700K halogen bulbs - they are inexpansive, and they have a perfect spectral characteristic (even better than a viewing booth)

Always use hardware type calibration, it calibrates precise, internal LUT of the panel, and that's why you get those perfect straight curves.
Logged

Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 7126


WWW
« Reply #5 on: September 16, 2008, 08:21:46 AM »
ReplyReply

Quote
I am not using a pro lab for developing my prints, but they did supply me with a printer profile, one for the glossy prints and another for the matt prints.  I am trying to compare my prints with a soft proof image in photoshop using these profiles.  They also supplied what they call a cablibration print which you can use to match with the same photograph you download from their website.
I don't have a viewing booth that is true, but I can't afford now for now.  Any ideas of how I can get better viewing the cheaper way?

Etienne
[a href=\"index.php?act=findpost&pid=221684\"][{POST_SNAPBACK}][/a]

Etienne, good that you have the lab profiles. You say you viewed the files on display under soft-proof and compared them with the prints you got back from the same files.  What major differences are you seeing? (The same ones you mentioned in your original post?) Are the differences smaller or larger for matte versus glossy prints. Are the differences mainly related to colour rendition or to luminosity (brightness and contrast)?  I agree that a halogen lamp with a 4700K Solux bulb is the best inexpensive way to view the prints, but you need to take care about the distance between the bulb and the print and the angle of reflection so that the print is not obscured by reflection, nor too close (bright) or too far away (dark).

Another important factor here is that there is never really a perfect match between a display view and a print view of an image, because of the fundamentally different quality of the lighting. The one is direct transmitted light and the other reflected. Therefore you will always have more contrast, somewhat deeper blacks and apparently richer saturation from the display compared with what you see via reflected light off the paper. This difference is larger for matte papers than for non-matte papers. One gets accustomed to this with experience and makes allowance for it in the adjustments. Using soft-proof, first I have my display luminosity set at a low 110 cd/M^2, and others have recommended even lower - 90~100 for example. This helps tone-down the display to somewhat mimic the reflectance of papers. It's an approximation. Then I watch my Curves carefully to make sure I have enough contrast and brightness under softproof so the print doesn't come out flat and dull when it should not be so. While I once used matte paper, I am now using Ilford Gold Fibre Silk, a kind of luster paper with very high DMax, good neutrality and easier to match between display and print because of the high DMax and good reflectance.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
Etienne Cassar
Jr. Member
**
Offline Offline

Posts: 95


« Reply #6 on: September 16, 2008, 02:39:45 PM »
ReplyReply

Quote
Etienne, good that you have the lab profiles. You say you viewed the files on display under soft-proof and compared them with the prints you got back from the same files.  What major differences are you seeing? (The same ones you mentioned in your original post?) Are the differences smaller or larger for matte versus glossy prints. Are the differences mainly related to colour rendition or to luminosity (brightness and contrast)?  I agree that a halogen lamp with a 4700K Solux bulb is the best inexpensive way to view the prints, but you need to take care about the distance between the bulb and the print and the angle of reflection so that the print is not obscured by reflection, nor too close (bright) or too far away (dark).

Another important factor here is that there is never really a perfect match between a display view and a print view of an image, because of the fundamentally different quality of the lighting. The one is direct transmitted light and the other reflected. Therefore you will always have more contrast, somewhat deeper blacks and apparently richer saturation from the display compared with what you see via reflected light off the paper. This difference is larger for matte papers than for non-matte papers. One gets accustomed to this with experience and makes allowance for it in the adjustments. Using soft-proof, first I have my display luminosity set at a low 110 cd/M^2, and others have recommended even lower - 90~100 for example. This helps tone-down the display to somewhat mimic the reflectance of papers. It's an approximation. Then I watch my Curves carefully to make sure I have enough contrast and brightness under softproof so the print doesn't come out flat and dull when it should not be so. While I once used matte paper, I am now using Ilford Gold Fibre Silk, a kind of luster paper with very high DMax, good neutrality and easier to match between display and print because of the high DMax and good reflectance.
[a href=\"index.php?act=findpost&pid=221743\"][{POST_SNAPBACK}][/a]

Thanks for your input.  I have recalibrated my monitor using hardware calibration alone and the result is much better now.  There is still some difference however with the yellows.  But I have realised that seeing the same photograph under different lighting conditions is resulting in different colours, so probably the differences I am seeing have to do with uncontrolled lighting conditions.  
The difference is greater with matt prints. They are lacking in saturation when compared to the images on the monitor and appear quite flat.  This is even more obvious with portraits.  The skin tones on the prints appear quite pale while those on the screen are much more natural.
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 7126


WWW
« Reply #7 on: September 16, 2008, 03:49:13 PM »
ReplyReply

Etienne, OK, what you are getting is expected. To compensate for the flat appearance of the matte prints, when you do the soft-proofing you need to make sure that you have checked the option "Simulate Paper White" in the Proof Set-up dialogue box. This has been referred to as the "make my picture look like s..t" option, but it's good, because it gives a better screen impression of how the print will look on paper. Then with this soft-proof condition active, you should adjust your contrast and saturation to compensate for the matte effect and print. You will see a noticeable improvement if you are not now doing that.

Mark
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
Etienne Cassar
Jr. Member
**
Offline Offline

Posts: 95


« Reply #8 on: September 18, 2008, 06:54:23 AM »
ReplyReply

What is the best White point, Luminance levels and Gamma to use when profiling the monitor?  The best result that I am getting seems to be when I choose to measure the White balance and luminance levels in my room using the i1 with the ambient light measurement head on. The white balance is found to be about 6539K and the white luminance is 122.6cd/m2.  I set Black luminance to min Native and Gamma to L*.  Is it right to do it this way or shall I use the D50 white point and another white luminance level?
Etienne
« Last Edit: September 18, 2008, 06:55:00 AM by ecassar » Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 7126


WWW
« Reply #9 on: September 18, 2008, 07:37:54 AM »
ReplyReply

So much of this really depends on your working environment and your printing media that you would be the best judge of what to use. I use 6500K for white point, 110 cd/m^2 for luminance, and L*, (and I view the prints under D50 illumination) so this is not too far from what you do, but my combination may not be optimal for you. That said, you may find a D50 white point somewhat yellowish on the display and you may get a closer luminosity match to your prints reducing the luminance below 122. Best is to experiment a bit and use the settings which provide you the most reliable outcomes.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
Etienne Cassar
Jr. Member
**
Offline Offline

Posts: 95


« Reply #10 on: October 14, 2008, 04:04:24 AM »
ReplyReply

All of a sudden I am getting strange results when calibrating my 2690 Spectraview monitor using the i1 display 2 puck.  AT D65 and D50 the monitor is becoming obviously pinkish (there will be a pink hue all over the room), and the white luminance will no go up more than 102cd/m2.  When I reset the monitor to factory default the white luminance goes up to 136cd/m2.  I know that this monitor should be very bright, and it does look bright when I reset it, but the i1 puck says that the white luminance is 136cd/m2.  Could it be possbile the the puck went bonkers and I need to get a new one, or is it more probable that the monitor became faulty after a month of use.  I know that it will be best to contact the supplier of my monitor, but I can't get him to see it because I had to buy it from abroad since there is no supplier in my country.  Any tips of how I can verify if my i1 puck is faulty.
Thanks
Logged
Etienne Cassar
Jr. Member
**
Offline Offline

Posts: 95


« Reply #11 on: October 14, 2008, 02:42:27 PM »
ReplyReply

Quote from: ecassar
All of a sudden I am getting strange results when calibrating my 2690 Spectraview monitor using the i1 display 2 puck.  AT D65 and D50 the monitor is becoming obviously pinkish (there will be a pink hue all over the room), and the white luminance will no go up more than 102cd/m2.  When I reset the monitor to factory default the white luminance goes up to 136cd/m2.  I know that this monitor should be very bright, and it does look bright when I reset it, but the i1 puck says that the white luminance is 136cd/m2.  Could it be possbile the the puck went bonkers and I need to get a new one, or is it more probable that the monitor became faulty after a month of use.  I know that it will be best to contact the supplier of my monitor, but I can't get him to see it because I had to buy it from abroad since there is no supplier in my country.  Any tips of how I can verify if my i1 puck is faulty.
Thanks
I tried to calibrate my laptop monitor and another LCD monitor that I have with the same i1 colorimeter, and the white luminance values that I was getting from both of these was around 60cd/m2.  This is definitely incorrect so probably the i1 is at fault and I'll have to get a new one.  Now once we're at it should I get a new i1, or is there something better now on the market to calibrate my NEC 2690 spectraview?  Any suggestions?
Logged
Czornyj
Sr. Member
****
Offline Offline

Posts: 1446



WWW
« Reply #12 on: October 14, 2008, 02:51:07 PM »
ReplyReply

Quote from: ecassar
I tried to calibrate my laptop monitor and another LCD monitor that I have with the same i1 colorimeter, and the white luminance values that I was getting from both of these was around 60cd/m2.  This is definitely incorrect so probably the i1 is at fault and I'll have to get a new one.  Now once we're at it should I get a new i1, or is there something better now on the market to calibrate my NEC 2690 spectraview?  Any suggestions?

X-Rite i1pro and Colormunki spectrophotometers and Quato Silver Haze (excellent X-Rite DTP-94B colorimeter)
Logged

Etienne Cassar
Jr. Member
**
Offline Offline

Posts: 95


« Reply #13 on: October 17, 2008, 04:12:21 AM »
ReplyReply

Can anyone tell me what is the difference between an eye-one display 2 and an eye-one display LT colorimeters?  Will the latter have full functionality when use the spectraview software?
Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9316



WWW
« Reply #14 on: October 17, 2008, 04:22:01 AM »
ReplyReply

Quote from: ecassar
Can anyone tell me what is the difference between an eye-one display 2 and an eye-one display LT colorimeters?  Will the latter have full functionality when use the spectraview software?

Hardware is identical.
Logged

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

Posts: 1446



WWW
« Reply #15 on: October 17, 2008, 04:24:56 AM »
ReplyReply

Quote from: ecassar
Can anyone tell me what is the difference between an eye-one display 2 and an eye-one display LT colorimeters?  Will the latter have full functionality when use the spectraview software?

There's no difference. Display2 has a licence key that enables the use of advance mode in i1match software, you can even upgrade LT to Display2 by uploading the D2 key into the colorimeter's EPROM.

For Spectraview software it doesn't matter at all.
« Last Edit: October 17, 2008, 04:26:01 AM by Czornyj » Logged

Serge Cashman
Full Member
***
Offline Offline

Posts: 200


« Reply #16 on: October 29, 2008, 01:20:16 AM »
ReplyReply

Quote from: digitaldog
Hardware is identical.

The firmware (or EPROM or whatever they call it) is  different though. I think the key code that you can buy from Xrite changes the firmware somehow.

I own both and the last time I checked (more than a year ago) Coloreyes and Basiccolor didn't work with LT for instance.

They are not seen by your computer as the same thing. I thought someone would hack it at some point but then I just lost interest. Maybe it has changed since then. It really was different from the Spyders 2 which were all identical and the difference was all in the software.

However, if your software recognizes it and can use it then I would say that there should be no difference.
« Last Edit: October 29, 2008, 02:30:56 AM by Serge Cashman » Logged
Etienne Cassar
Jr. Member
**
Offline Offline

Posts: 95


« Reply #17 on: October 30, 2008, 06:26:50 PM »
ReplyReply

Quote from: Serge Cashman
The firmware (or EPROM or whatever they call it) is  different though. I think the key code that you can buy from Xrite changes the firmware somehow.

I own both and the last time I checked (more than a year ago) Coloreyes and Basiccolor didn't work with LT for instance.

They are not seen by your computer as the same thing. I thought someone would hack it at some point but then I just lost interest. Maybe it has changed since then. It really was different from the Spyders 2 which were all identical and the difference was all in the software.

However, if your software recognizes it and can use it then I would say that there should be no difference.
Thanks all for your feedback.  I have recieved my Eyeone LT and used it with the Spectraview software and it worked perfectly.  It was fully functional with the spectraview software just like to EyeOne Display 2.
Logged
walter.sk
Sr. Member
****
Offline Offline

Posts: 1334


« Reply #18 on: December 02, 2008, 06:14:37 PM »
ReplyReply

Well, I finally received my SpectraViewii software to use with my NEC 3090WQXi.  I have the i1 Display v2 colorimeter.  I have several questions about calibrating/profiling with this setup:

1)  Should I set the monitor preferences to use the Auto Luminance option, or would this defeat the purpose of profiling and calibrating?

2)  For Contrast Ratio, should I go for the Maximum Contrast Ratio or Best Grayscale Tracking?

3)  Is there any advantage to using the native gamma and white point rather than 2.2 and D65?

4)  The user's manual says "For wide gamut displays use the Factory Measurement of the primaries rather than Colorimeter except for Spectral Calibration Sensors.  Is the i1 Display 2 optimized for the wider gamut?  Do I lose anything by using the Factory Measurement option?

Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9316



WWW
« Reply #19 on: December 02, 2008, 07:29:05 PM »
ReplyReply

Quote from: walter.sk
Well, I finally received my SpectraViewii software to use with my NEC 3090WQXi.  I have the i1 Display v2 colorimeter.  I have several questions about calibrating/profiling with this setup:

1)  Should I set the monitor preferences to use the Auto Luminance option, or would this defeat the purpose of profiling and calibrating?

2)  For Contrast Ratio, should I go for the Maximum Contrast Ratio or Best Grayscale Tracking?

3)  Is there any advantage to using the native gamma and white point rather than 2.2 and D65?

4)  The user's manual says "For wide gamut displays use the Factory Measurement of the primaries rather than Colorimeter except for Spectral Calibration Sensors.  Is the i1 Display 2 optimized for the wider gamut?  Do I lose anything by using the Factory Measurement option?

1, yes, I'd set this to a desired luminance for print matching.
2. Use Best Gray tracking (who wants max contrast which greatly exceeds that of a print?
3. That colorimeter isn't optimized for wide gamut.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
Pages: [1]   Top of Page
Print
Jump to:  

Ad
Ad
Ad