NEC PA242W Monitor Review
by Chris Heinonen on September 27, 2013 9:00 AM ESTUpdated 10/2/2013: Talking to NEC after this went live it was found that the review unit had a feature, Metamerism, enabled. This helps to match it to other displays, but also caused the pre-calibration numbers and non-SpectraView numbers, to be incorrect. I'm currently re-running the pre-calibration numbers on the NEC now and updating the pages as fast as possible. The end result is that the NEC performs far, far better out-of-the-box than initially thought.
Be prepared, as there is a lot of bench test data coming here. For my pre-calibration settings I used the sRGB color gamut, a color temperature of 6500K, and a gamma setting of 2.2.
Pre-Calibration (re-tested) |
Post-Calibration, 200 cd/m^2 |
Post-Calibration, 80 cd/m^2 |
|
White Level (cd/m^2) | 197.0 | 204.14 | 80.79 |
Black Level (cd/m^2) | 0.3457 | 0.366 | 0.1479 |
Contrast Ratio | 570:1 | 558:1 | 546:1 |
Gamma (Average) | 2.2633 | 2.1437 | 2.3734 |
Color Temperature | 6460K | 6426K | 6442K |
Grayscale dE2000 | 1.573 | 0.6504 | 0.6473 |
Color Checker dE2000 | 1.0198 | 0.6392 | 0.5331 |
Saturations dE2000 | 0.9682 | 0.6722 | 0.5675 |
Out of the box the NEC is practically perfect. The grayscale has a little bit of an error but one that should barely be visible if at all. The gamma has a small peak at 95% but no huge issues at all. The colors are reference quality and there is nothing to complain about with them. If this was a post-calibration result I would say it is amazing. The fact that it is a pre-calibration one makes it even that much more incredible.
Post calibration, aside from a dip in the gamma at 90-95%, everything else improves and becomes practically perfect. Yes, the overall error levels can be lower but you couldn't see it. None of the 96 samples in the large color checker chart even come close to a dE2000 of 2, much less the visible limit of 3. The average error of 0.63 is the lowest I've ever seen. It's perfect.
When targeting 80 cd/m^2 and the sRGB gamma target we see similar performance. The gamma curve isn’t perfect but really everything else is. The color checker chart hits an average dE2000 of 0.53 here which is even better, but not visible. What you see on the screen is what you are supposed to see.
With Metamerism turned off, the NEC measures perfectly. The only improvement I could see is in contrast ratio, but they might be letting that suffer to coax more reliable overall performance of the panel which is a trade-off that would be worth it for their target markets. Perhaps once OLED gets affordable we can see something better, but until then this is really, really good.
74 Comments
View All Comments
cheinonen - Monday, September 30, 2013 - link
The current version of CalMAN, at least one of the higher end versions, has support for calibrating a set of displays to look identical. It might mean none of them are perfect, but they all look about the same. I've never tested it out, as it's meant for commercial installations, but that might help with this issue if you already have the monitors.1Angelreloaded - Saturday, October 5, 2013 - link
instead of using 3 monitors, have you ever thought about moving to a large format display they make professional based ones for 40inch and higher now at 4k levels with IPS tech, that's a lot more realestate than say 3 1080p s side by side.nathanddrews - Friday, September 27, 2013 - link
I think I already know the answer to this: "too much".How much would it actually add to the cost of a monitor to have it pre-calibrated using this method at the factory? Obviously a trained human would be expensive and time consuming, but what about a series of sensors on the assembly line that tune the display before packaging? Panels that meet the requirements for uniformity and gamut get one price, displays that can't make the cut go off to a different bin. Doesn't seem too ridiculous to me.
Senti - Friday, September 27, 2013 - link
The problem is the transportation after that will likely screw everything in unpredictable way.nathanddrews - Friday, September 27, 2013 - link
How, exactly? I've never heard of shipping or transporting affecting display calibration...foxalopex - Friday, September 27, 2013 - link
Professional calibration software like Spectraview is designed to ask for a recalibration every 2-4 weeks. So yes by the time it gets to you it's not as calibrated as it could be. Keep in mind this is a professional monitor so they're nothing like a normal user. If you're more on a budget, a good $600-700 IPS monitor would probably work for most folks but for folks who insist on the best, you can't go wrong.DanNeely - Friday, September 27, 2013 - link
That drift is due to components aging with; a good factory calibration should still be good when you unbox it and will stay better than what we normally suffer with for a while since the bad one is drifting too.Some tablet vendors have been doing it for at least a year now. I'd be interested in seeing how much the screens on them have drifted if Anandtech used any of the tablets they did color calibration testing a year ago as daily drivers.
bobbozzo - Friday, September 27, 2013 - link
see above:cheinonen - Friday, September 27, 2013
...the out-of-box experience is also very good and it is still more uniform than any other display tested to date. The calibration does not affect the uniformity.
cheinonen - Saturday, September 28, 2013 - link
Just as an FYI, measuring the uniformity to the degree that I do takes close to two hours per display. That's with a fast meter and moving it between locations as quickly as possible but still allowing time for each sample to settle and be accurate. Adding that kind of testing to every display will increase the price a lot. Most companies wouldn't see any return on investment there, since the majority of consumers still don't care enough. Even if you only add $25 to the cost, that's a lot for most people unless you're talking about $1,200 displays.dushyanth - Friday, September 27, 2013 - link
which would you choose: Eizo CS230 or this