Calibrating new monitor... something's not right

[Mike Guilbault]

I know this isn't about a printer, but I'm setting up a new monitor and my prints aren't matching what I see.

I purchased a new NEC PA272W-BK monitor. After setting it up it was significantly magenta. NEC agreed and sent me a replacement. Although not nearly as bad, when I try to calibrate the monitor (with SpectraViewII & i1 Display Pro - D65, Intensity 80, Gamma 2.20, Contrast 500:1), it's still coming up more magenta. I have my old Eizo set up beside it and it's definitely warmer in tone and the prints look identical to the Eizo. They are no where near what I see on the NEC. I tried setting the WB on the NEC to D55 which warmed it up a bit, but still more magenta than the print.

I'm not sure what else to try. Any ideas?

[Mark Lindquist]

I'm not sure what else to try. Any ideas?

Are you running the monitor power through an APS Mike?

If yes, try running it straight to the wall.

If no, try running it through the APS.

Sometimes power fluctuations and/or different sources make a difference.

Mark

[Mike Guilbault]

Will an extension matter?  Right now it's running to my APS.

[Mark Lindquist]

It could matter Mike.  Only thing you can do is experiment.
Also, what about your video card?  Do you have the latest drivers installed?

Is your video card the right one and most powerful for your application?

I'd start at the wall and work backwards.

Plug the monitor into the wall straight - see if it makes any difference.
Then go from there - eliminate the extension cord, etc.

Never know what's causing the problem.

Mark

[Chris_Brown]

Did the monitor come with the i1 puck? (Some models did, some didn't).

And your brightness seems low. Is the "80" from a preset or your own preference?

Lastly, what platform/OS and video card is it running on?

[Mark D Segal]

I know this isn't about a printer, but I'm setting up a new monitor and my prints aren't matching what I see.

I purchased a new NEC PA272W-BK monitor. After setting it up it was significantly magenta. NEC agreed and sent me a replacement. Although not nearly as bad, when I try to calibrate the monitor (with SpectraViewII & i1 Display Pro - D65, Intensity 80, Gamma 2.20, Contrast 500:1), it's still coming up more magenta. I have my old Eizo set up beside it and it's definitely warmer in tone and the prints look identical to the Eizo. They are no where near what I see on the NEC. I tried setting the WB on the NEC to D55 which warmed it up a bit, but still more magenta than the print.

I'm not sure what else to try. Any ideas?

Mike, perhaps you know all this - but just in case: (1) the difference between calibrating and profiling. Those settings you mention are the calibration, and the white point in particular of D55 I would suggest reducing to D50. Once the display is calibrated to those parameters it needs to be profiled and you need to make sure that profile is the one loaded in your operating system as the default profile for that display. Is this what you are doing? Once you have done this, do not change any of the calibration settings externally, only recalibrate  and then reprofile with the new calibration through the software. Just in case you are having a software issue, you could try downloading a 14 day trial of BasicColor 5 display to see whether that solves the problem. If it doesn't, this could indicate a hardware issue as long as you are using Spectraview correctly.

[Mike Guilbault]

I'll have to try the different outlet tomorrow. Got fed up and went home. 

Ok... the i1 is the Display Pro - did not come with the NEC. I already had the i1 Display Pro so just purchased the SpectraViewII software separately. I'm running it on a MacPro (Early 2008- I know, time to upgrade) using an NVIDIA GeForce 8800 GT 512MB card.
And yes, I've calibrated and then saved the settings to a profile which is the one selected for the monitor by the system. When I changed to D55 it seemed to help a little, so I'll try D50. What I don't get is that the Eizo is using D65, with an Intensity of 90 and Gamma of 2.20. I would have thought that at the same settings, the NEC would look pretty close to the same - but no.  Why would I have to set the NEC at D50 to match the Eizo at D65?

Here's an iPhone shot of my two screens - the colour isn't quite accurate so it's hard to tell, but you can definitely see the NEC on the right is stronger on the magenta side than the Eizo on the left. Both calibrated and profiled - Eizo using i1 DisplayPro software and NEC with SpectraViewII and the i1DP sensor. And my prints come out pretty much like on the Eizo - hard to see here, but a little on the warm side. So if I adjust according to the NEC, they're going to come out way too warm.

[Mark D Segal]

You have a point - the same calibration settings should produce very similar profiles and colour appearance on both displays, so it shouldn't be necessary to use such differentt white point settings. That's the whole idea. Did you calibrate and profile the Eizo with the same colorimeter (i1DP)? Does the Eizo have similar wide-gamut as the NEC? I'm wondering whether there is a mismatch between the gamut of the NEC display and the capability of the colorimeter producing an outlying result; if not, all this leads me to think there may still be a hardware issue with the replacement NEC. Wondering whether they produced a bad batch.

[Mike Guilbault]

YEs, the were both calibrated with the same sensor, but using the i1 Display Pro software for the Eizo and the SpectraViewII for the NEC. I don't believe the Eizo has as wide a gamut as the NEC (99% Adobe RGB)... I believe the Eizo is about 8 years olds as well (CE210W). Everything I've read says that the i1 Display Pro is compatible with the monitor and SpectraViewII software, so that shouldn't be the issue.  I did notice that the "new" replacement NEC monitor had a "Refurbished" sticker on the packaging. I'll check the serial numbers to see how close they are.

[Mark D Segal]

Maybe it wasn't refurbished too well, but only NEC knows what was wrong with it to start with - maybe the same issue, maybe not. Starting to sound like a hardware problem. Maybe you need to have a probing discussion with them.

[Mike Guilbault]

May have to wait until after Christmas. Too many orders to finish up to fiddle with a new monitor that isn't 'right'. Looks like I'll go back to the Eizo for now until I have the time to investigate further. I will call NEC though to make sure the issue is logged into a support ticket.

Thanks for the input. I just couldn't figure out what I was doing wrong - and apparently I'm not.

[Mark D Segal]

Glad to hear you have lots of orders Mike - that's good! And at least you have one reliable display to help finish them. So there will be light at the end of the tunnel. What you have described so far from your end DOES sounds right; however a thorough discussion with NEC about how you are set-up and what you did to profile and calibrate it would be useful just to make sure there isn't some factor we haven't thought of interfering with correct performance. Failing that, they should send you a new display - again. This happens. I once had to return four office printers in a row for the identical defect one after another. Happy holidays.

[Czornyj]

Download NEC MultiProfiler and check if enabling or disabling "Metamerism correction" in Picture Mode reduces the effect. If still not there, use Edit... Quick Edit > White Point and use manual correction to get a match with the print.

[Royce Howland]

Mike, as always with digital colour issues, the challenge is figuring out what exactly is wrong, and from there what caused it. Then some targeted corrective actions can be taken. Anything else is more like shooting randomly, playing the odds and hoping to get a lucky hit. That can work, but I like to go for a more diagnostic approach.

The odds that 2 different PA272W monitors have the same problem to a greater or lesser degree, seems remote to me. It's not a zero chance, certainly it could happen. But you've got a few variables in your system, one of which is a significant gamut difference between your two monitors, and undoubtedly a degree of white point difference as well. Running 2 different calibration/profiling apps to baseline the monitors can't completely compensate for those differences. Something else going on in the digital colour pipeline easily could take an image that looks normal on your Eizo and render it highly magenta on the NEC, without it being the fault of the monitor itself. Before concluding that the NEC is wrong, and shipping it back for a 3rd unit, I'd look at some other things first.

If you have a bit of time during your production runs, may I suggest posting a ZIP file somewhere that contains the following: the ICC profiles for both your monitors, a low-res example image that clearly demonstrates the colour mismatch on the monitors, and a set of screenshots or text notes describing all of your settings in the SpectraView II software. I'd be happy to take a look based on what I know and what I can see using some tools, and see if anything is suggested from that.

By the way, I agree with the advice to download MultiProfiler and try the metamerism correction setting, and see what effect (if any) that has...

[scyth]

You have a point - the same calibration settings should produce very similar profiles and colour appearance on both displays, so it shouldn't be necessary to use such differentt white point settings. That's the whole idea.

only if you are using a decent spectrophotometer... with colorimeter, even with a decent like i1DPro and monitors as decent as Eizo/Nec it is possible that calibration matrices (that are exactly supplied because colorimeter does not measure the spectrum) supplied by Eizo or Nec for actual different monitors will not yield the same ( means visually not discernible ) result... actually a visibly different result

[Czornyj]

only if you are using a decent spectrophotometer... with colorimeter, even with a decent like i1DPro and monitors as decent as Eizo/Nec it is possible that calibration matrices (that are exactly supplied because colorimeter does not measure the spectrum) supplied by Eizo or Nec for actual different monitors will not yield the same ( means visually not discernible ) result... actually a visibly different result

Even calibration with the most sofisticated lab-grade spectroradiometers produce slight different results in white appearance of displays with different spectra of the backlight due to imperfections of CMF, and nature of retina's chromatic adaptation. Use metameric correction when available, and visual tweaking when it still doesn't produce a match in multi monitor setup.

[Zolty]

YEs, the were both calibrated with the same sensor, but using the i1 Display Pro software for the Eizo and the SpectraViewII for the NEC. I don't believe the Eizo has as wide a gamut as the NEC (99% Adobe RGB)... I believe the Eizo is about 8 years olds as well (CE210W). Everything I've read says that the i1 Display Pro is compatible with the monitor and SpectraViewII software, so that shouldn't be the issue.  I did notice that the "new" replacement NEC monitor had a "Refurbished" sticker on the packaging. I'll check the serial numbers to see how close they are.

I am not Mac user, but few questions coming to my mind:
-were you using same cable/connector to connect both copies of NEC? If so - maybe that is where the fault is (unlikely) - try to change/swap NEC/EIZO...
- you mentioned you used i1Pro soft to calibrate Eizo and NEC SV to calibrate NEC - any chance for some conflict between two softs installed (and running)? Could you try to calbrate/profile both monitors (EIZO/NEC) using same soft (i1Pro) and compare results?
- I ran out of ideas, but keep my fingers crossed

[Mike Guilbault]

I wrote NEC again this morning and they've sent out a replacement, again. This one is supposed to be "new". The rep said that once I receive this one, and if the problem persists, they will have their experts contact me to make sure I'm setting it up correctly I'm actually quite impressed with the level of service and support I'm getting so far. I just hope this next one does the trick. 

I noticed that the serial numbers of the two monitors I have now are only 77 numbers apart. Could have been a bad batch and why they're so quick to supply a replacement. Might be a known issue?  Anyway, once I get it, I'll try it out again, but I"ve got to get my orders for Christmas done and then I'll worry about the monitor. Not going anywhere this year, so I'll have some spare time to delve into this deeper.  I'll keep you posted!

[Mark D Segal]

Hi Mike; good for NEC - that's service and support for you and gives confidence regardless of the temporary issues. It is indeed interesting that they chose the path a second replacement before further exploring with you all the other factors that may be producing the outcomes you have been reporting. Normally one would expect the reverse, so does perhaps tell us something; but they are responding very well. I do hope this time it works properly, and please keep us posted - this is an interesting case.