calibrating NEC 2690wuxi2

[ejnewman]

Hello all, I am having some problems getting my screen to calibrate, I am using the NEC MultiSync 2690WUXi2, Eye 1 Dispay, trial of BasICColor Display on OSX running a Macbook Pro.

I am seeing less magenta and slightly less saturation and contrast in my NEC display compared to my macbook screen when looking at an sRGB image in Photoshop (with sRGB tags).

I have also seen the same image on a variety of displays at my office (film post production house), where we have a selection of Apple Displays, NEC 2490s and Sony CRTs, all of which are calibrated and I have seen the same image on my labs calibrated Apple Display too. What I have noticed is that all of these displays show the image with more magenta, more saturation and contrast than my 2690.

I wanted to illustrate this for you, I only have access to my laptop right now, but it has been recently calibrated, as has my NEC. I photographed both at the same time, and then comped them back ontop of one another to show the visual differences. I know a laptop screen is not the best example, but it is good enough to at least show the trend.

My question is: is there possibly something wrong with my 2690? Or do I simply need to invest in the wide gamut colorimeter + Spectraview software?

Thanks
Elliot Newman

[Mark D Segal]

Hello all, I am having some problems getting my screen to calibrate, I am using the NEC MultiSync 2690WUXi2, Eye 1 Dispay, trial of BasICColor Display on OSX running a Macbook Pro.

I am seeing less magenta and slightly less saturation and contrast in my NEC display compared to my macbook screen when looking at an sRGB image in Photoshop (with sRGB tags).

I have also seen the same image on a variety of displays at my office (film post production house), where we have a selection of Apple Displays, NEC 2490s and Sony CRTs, all of which are calibrated and I have seen the same image on my labs calibrated Apple Display too. What I have noticed is that all of these displays show the image with more magenta, more saturation and contrast than my 2690.

I wanted to illustrate this for you, I only have access to my laptop right now, but it has been recently calibrated, as has my NEC. I photographed both at the same time, and then comped them back ontop of one another to show the visual differences. I know a laptop screen is not the best example, but it is good enough to at least show the trend.

My question is: is there possibly something wrong with my 2690? Or do I simply need to invest in the wide gamut colorimeter + Spectraview software?

Thanks
Elliot Newman

Eliot, a laptop display is a notoriously difficult barometer for anything. All you need to do is change the angle of vision a bit and you can overwhelm any differences you are detecting. It could be that what you are seeing on the 2690 is more correct than what you are seeing on other older, smaller gamut displays. One good test would be to see how your prints compare with the display image. If the matching is good, perhaps there is no problem. But further checking may be anyhow worthwhile. So, I would ask whether all the calibration parameters are the same between the various displays on which you are viewing the image. Next I would ask whether the file numbers of the image are all within display gamut - say sRGB, as you are comparing displays that have differing gamut capability. Then I would turn to the hardware aspect of the 2690. It is advisable to use a wide-gamut capable colorimeter for optimal profiling results of a wide gamut display.  Finally, there is an application called Patchtool produced by BabelColor which has a display check tool. It generates patches independently of the verification routine in the profiling software, so it is an independent check on how well your display profile reproduces the file numbers in their reference file for the selected test.

[ejnewman]

Thanks Mark. Yes get what you are saying about the laptop screen, but my point is that my laptop is in the same ball park as other screens I have tested on such as the standard gamut NEC 2490 and Apple Cinema Displays - I was just using it here as an example to illustrate the differences, with all the caveats that naturally come with it.

There are two reasons why I don't think it is to do with the differences in a wide-gamut display:

1. The difference is not restricted purely to colour intensity, it's affectively more magenta on all the other standard-gamut displays I compare against.
2. If anything, the standard-gamut displays are producing more vibrant, rich colours than my wide-gamut NEC - not the other way around which one would expect if it is a wide-gamut difference.

I just "feel" like the screen is cooler/more cyan/green than it should be, even though I am specifying D65 in the calibration process - I use calibrated apple displays everyday at work so my eye is pretty well trained in seeing neutral, and I just don't feel like I am seeing that with my 2690.

The labs print is also coming out with more magenta - and I am seeing a very close match there with what I am seeing on every other display other than my 2690, so it really looks like the finger is pointing at my 2690 as the odd one out.

So, taking your advice, I should invest in a wide-gamut colorimeter, so that is the next logical step. My question is: how do I find one in the UK? Do I have to import?

Thanks.
Elliot

[Mark D Segal]

Try calibrating to D50 and see whether the problem goes away.

On the colorimeter - sorry - living in Toronto Canada I have no idea what the retail situation is in the UK, but there are other contributers here with lots of UK savvy.

[Thomas Krüger]

Check http://www.nativedigital.com/ in the UK for a colorimeter.

Spectraview Software: http://www.necdisplay.com/supportcenter/monitors/spectraview2/downloads/
Supported color sensors: http://www.necdisplay.com/supportcenter/monitors/spectraview2/compatibility/

[digitaldog]

I am seeing less magenta and slightly less saturation and contrast in my NEC display compared to my macbook screen when looking at an sRGB image in Photoshop (with sRGB tags).

The question you should be asking is why you are comparing the Macbook screen (a not very good display) to the NEC and assuming they would match.

[ejnewman]

DigitalDog - I am not expecting a laptop screen to match my NEC.

The point is:

Laptop, 3 x apple displays, 1 x 2490 NEC, 1 x Sony CRT = similar ballpark colours
NEC 2690 = totally out of ballpark colours.

[digitaldog]

DigitalDog - I am not expecting a laptop screen to match my NEC.
The point is:

Laptop, 3 x apple displays, 1 x 2490 NEC, 1 x Sony CRT = similar ballpark colours
NEC 2690 = totally out of ballpark colours.

Sounds like you do.

Its simple! You have to adjust the various calibration targets to produce a match. Are you saying that no matter what you set for WP, luminance, black point, you can’t get the NEC to match? And they are all the same basic gamut (cause if one’s a wide gamut, the lower gamut displays will never match out of gamut colors). And don’t assume the non wide gamut displays produce sRGB or its gamut.

[Mark D Segal]

DigitalDog - I am not expecting a laptop screen to match my NEC.

The point is:

Laptop, 3 x apple displays, 1 x 2490 NEC, 1 x Sony CRT = similar ballpark colours
NEC 2690 = totally out of ballpark colours.

I would suggest not comparing your 2690 image with what shows on other displays, because that's not important, and as Andrew implies, most likely not reliable, as you know in the case of the laptop. What is important is whether the image on the display matches the prints you get back. If you are sending the files to a lab for printing you have no control over what they produce (or very limited), but you do have control over how your display is calibrated and profiled to reliably predict, with softproofing ( can the lab give you a profile?), the outcomes. That's the acid test. And I suggest you download babelcolor and try the displaycheck tool to see whether it tells you whether or not the display is doing a decent job representing the file numbers.

[ejnewman]

DigitalDog:
Thanks for your comments, you must feel like your smashing your head against a brick wall with me, for that I appreciate your patience, as my simpleness frustrates me too!

Mark:
Yep, that is the main goal - to get as close to the final print as I can. I am happy with what my print lab displays are showing me as the print results match their displays, however they do not match mine. So in essence I am trying to match a print... I know that if I can get my display to look closer to their display then the prints will also match.

I guess the reason I am putting emphasis on comparing other displays and not just the final print output is because I am convinced something is wrong with at least one element of my calibration process, and I believe its the colorimeter. Put it this way: the imaging department at the post production facility I work for calibrate 3 different display types - Apple, NEC standard-gamut and Sony CRTs. As I look across the floor I see all these displays showing consistent results even when looking at the same image, there is no manual step of getting all the displays to line up together and we are able to integrate all three display types into our colour critical work because they match, at least to a reasonable enough level.

I then see the same image on the print labs display and it looks pretty dam close to all the displays at work. However my display is completely off, and it's off by quite some margin.

I have just re-calibrated all my displays using BasICColor, but this time I also calibrated a spare LCD display - its a Benq 24inch. The results are now all in a pretty good ballpark from one another, I am no longer seeing the huge magenta shift, maybe this is down to BasICColor doing a better job. So I believe the colorimeter is creating consistently incorrect results, regardless of the wide-gamut display.

To me this confirms that my Eye One Display colorimeter is what is causing the problem, it's over a year and a half old now, so it looks like the next logical step is to borrow the colorimeter from work and see if that gives me the results I am expecting.

Thanks all,
Elliot

[Mark D Segal]

..............

I have just re-calibrated all my displays using BasICColor, but this time I also calibrated a spare LCD display - its a Benq 24inch. The results are now all in a pretty good ballpark from one another, I am no longer seeing the huge magenta shift, maybe this is down to BasICColor doing a better job. So I believe the colorimeter is creating consistently incorrect results, regardless of the wide-gamut display.

To me this confirms that my Eye One Display colorimeter is what is causing the problem, it's over a year and a half old now, so it looks like the next logical step is to borrow the colorimeter from work and see if that gives me the results I am expecting.

Thanks all,
Elliot

Eliot, looks as if you may be zero-ing in on the problem area. But to clarify - BasicColor is a piece of software which does calculations based on the data it receives. It does not do a better or worse job from one exercise to the next, as you seem to imply above. What changes is the data, and that indeed is what comes from the colorimeter. It needs to be in calibration, needs to be properly warmed-up, and needs to be capable of reading the display gamut to give accurate enough results. I have heard that over time the dyes in the colorimeter filters can change even without using the device, but since yours is only 1.5 years old, this should not be an issue. Borrowing the one from work, as long as BasicColor supports it and it can read wide-gamut displays properly, may give you the added insight you need, depending on whether it is capable of reading your display gamut. Of course please be mindful of appropriate calibration settings, and do try to get a printer profile from your lab, enabling to softproof your images before you send them for printing. The appropriate comparison is the soft-proofed image with the print, using the correct printer profile for soft-proofing.

[Alan Goldhammer]

Finally, there is an application called Patchtool produced by BabelColor which has a display check tool. It generates patches independently of the verification routine in the profiling software, so it is an independent check on how well your display profile reproduces the file numbers in their reference file for the selected test.

The PatchTool website says that it only works with the i1 pro only.  I have the iOne Display 2 that comes with the SpectraView kit so I guess there is no sense in trying this software out.

[Mark D Segal]

The PatchTool website says that it only works with the i1 pro only.  I have the iOne Display 2 that comes with the SpectraView kit so I guess there is no sense in trying this software out.

According to the Patchtool manual, it supports Spyder 2, Spyder 3, i1 Display and Display 2, i1 Pro and DTP-94 (Monaco Optix XR). But it does support the NEC branded version of i1 Display 2. That is an issue for me too.

[Alan Goldhammer]

According to the Patchtool manual, it supports Spyder 2, Spyder 3, i1 Display and Display 2, i1 Pro and DTP-94 (Monaco Optix XR). But it does support the NEC branded version of i1 Display 2. That is an issue for me too.

I can always try the download for free before deciding to purchase.  Thanks Mark.

[digitaldog]

Finally, there is an application called Patchtool produced by BabelColor which has a display check tool. It generates patches independently of the verification routine in the profiling software, so it is an independent check on how well your display profile reproduces the file numbers in their reference file for the selected test.

Its a better tool because you can define the patches to check but you are still using the same instrument that calibrated the display to make the check. It is more useful than the check used by the profiling software because the colors they pick are often those that will easily produce low delta’s.

[Mark D Segal]

Its a better tool because you can define the patches to check but you are still using the same instrument that calibrated the display to make the check. It is more useful than the check used by the profiling software because the colors they pick are often those that will easily produce low delta’s.

Andrew, fair point, but suppose I have two different colorimeters or two copies of the same colorimeter - I could make two rounds of tests with the same set of patches, in which case I would be largely measuring instrument differences, which may shed light on the validity of either of the tests. At least, for those with doubles of equipment and Patchtool's independent colour lists, it would get one out of the circular loop issue.

[digitaldog]

Andrew, fair point, but suppose I have two different colorimeters or two copies of the same colorimeter - I could make two rounds of tests with the same set of patches, in which case I would be largely measuring instrument differences, which may shed light on the validity of either of the tests. At least, for those with doubles of equipment and Patchtool's independent colour lists, it would get one out of the circular loop issue.

It would tell you about the two units and their differences, but without a better reference grade device, which is right? Its useful to know they do not correlate and by how much, no question. It would also be useful to do this test say just after a device is certified by the manufacture and at progressively later dates.

[Mark D Segal]

It would tell you about the two units and their differences, but without a better reference grade device, which is right? Its useful to know they do not correlate and by how much, no question. It would also be useful to do this test say just after a device is certified by the manufacture and at progressively later dates.

True.

[DeeJay]

I occaisonally use a macbook Pro with an NEC2690 and found the same thing. The macbook pro screens are terrible. The reds and magentas in particular are well out. But I also found that once you unplugged the NEC and rebooted, the magentas and reds didn't seem so flarey on the Macbook Pro anymore. I don't know, I think it was just the way the Graphics card was working with the system and display together. Either way I trusted the NEC and it worked.

The point is though, forget referencing the macbook pro screen it's rubbish.