Setting Custom White Point to get a screen-print match ?

[MartinG]

Hello everyone,

I have a NEC PA241W monitor and have been playing with the calibration since november. At 6500K, it always had an annoying yellowish tin, mostly in the whites.

I reached perfect match with the prints by using the Custom White Point function in the Nec SpectraviewII software, then re-calibrated using the measures obtained. It seems a bit far from the standard 6500K target (7673K / cie x=0.3002, y=0.3024) but everything is spot-on.

This way, the white and colors on screen also matches all my other calibrated monitors, and both printed and on screen pictures look identical. My guess is that I it is the way to reach the right colors.

Is that the right way to do this ? Is it a monitor defect or it didn't finish it's "break-in" period of 100hours (around 80hrs so far)? Should I leave it at 6500K so other people who will bring their images would have the same target at home or it does not matter. I am a bit confused by this. My MacbookPro is spot on using 6500K with my X-Rite i1displayPro and I didn't have to tweak the white point like this.

Thanks for your inputs,

Martin

[digitaldog]

Is that the right way to do this ?

IF you got the match you want, the answer is a resounding YES. That's why we have those nice controls in the software. And why, when people ask about products that only offer WB on one axis, or the inability to tweak as you did, I often say the software is crippled.

[MartinG]

Great! Thanks DigitalDog.

I do have a great match now. Everything matches perfectly and I am very picky when it comes to color accuracy. When I look at the printed output under daylight and compare with the screen, both luminance and colors are perfectly spot-on.

It is possible that this particular NEC display needs to be calibrated at a 7673K white point everytime I calibrate or it could "settle" and be calibrated at different temperatures over time ?

If one didn't have the SpectraView software, he could have the yellow cast forever and have to deal with it ?

This is the part what makes me wonder the most.

Thanks again for the information!

Martin

[digitaldog]

It is possible that this particular NEC display needs to be calibrated at a 7673K white point everytime I calibrate or it could "settle" and be calibrated at different temperatures over time ?

I too have a very custom CCT value to match my booth and I've always kept it that way and always get the match I expect.

There is no single correct value other than the one that produces a match! Considering we all use different viewing conditions, makes sense why. As long as that viewing condition and the display remain stable, you should be done. Just recalibrate on a regular basis with the same targets. You may want to build more than one (assuming you are using different papers who's white point there could vary) or just differing contrast ratio's (Matt vs. Glossy). The SpectraView will let you build as many as you want and switch on the fly.

[MartinG]

Excellent, thanks so much. Now I can work on other stuff and be confident that the 60'' prints coming out from the 11880 will match the screen and have the right colors.

I will indeed make other calibration targets in SVII for some other papers that I carry. It is a very useful and efficient software.

Cheers,

Martin

[l_d_allan]

... when it comes to color accuracy. When I look at the printed output under daylight and compare with the screen, both luminance and colors are perfectly spot-on.

Something I've wondered ... is viewing prints with light from a north-facing window at mid-day on a clear day considered to be "good light" for this? Granted, this is not at all in close proximity to the monitor so I could compare the print with the monitor.

I suppose with a complex print, my memory of what the monitor looked like might be undependable. My speculation is that the north light would be suitable for seeing if flesh tones look acceptable.

It also would seem "better than nothing" for comparing a print like the Printer-Evaluation-Image from OutBackPhoto test-print with a known-to-be-decent print made with OEM inks and OEM paper.  In the ball-park? Or off?

My wife is an artist and has a simple, entry-level OttLite that seems to work ok for this hobby'ist without print viewing booth.

[Schewe]

Something I've wondered ... is viewing prints with light from a north-facing window at mid-day on a clear day considered to be "good light" for this? Granted, this is not at all in close proximity to the monitor so I could compare the print with the monitor.

North light may be a good light to paint under but for print eval, it's pretty darn cool. Northern Skylight is about D75 (7500º K). So, it'll be a potential problem...the OTTLite is prolly better (but it depends on which bulb is in it).

[l_d_allan]

Good point about north light being cool (> 6000). Thanks.

[Ernst Dinkla]

Excellent, thanks so much. Now I can work on other stuff and be confident that the 60'' prints coming out from the 11880 will match the screen and have the right colors.

I will indeed make other calibration targets in SVII for some other papers that I carry. It is a very useful and efficient software.

Cheers,

Martin

If it is a sign for outdoor use you are on the right track. If not then you have to find a wall in your house, gallery, museum, customer's premises that has the same D75 lighting to hang your print. The painter that likes to work with D75 faces the same choice. The natural northern light was not selected for its color temperature but for being the most consistent during the day when artificial lighting was not more than an oil lamp or candles. Most display conditions are below 5000K. Where the lighting has to be reduced for longevity (musea) it can be near 3000K and the two choices need one another considering the Kruithof curve.

I bet that at D75 even the iPhone display matches. It would not wonder me if the inkjet paper used has a nice load of OBAs too.


--
Met vriendelijke groet, Ernst

http://www.pigment-print.com/spectralplots/spectrumviz_1.htm
December 2012, 500+ inkjet media white spectral plots.