I'm thinking about getting the NEC LCD2190UXi (21" S-IPS) monitor. I could buy just the monitor, or I could pay a little more and get the monitor plus NEC's SpectraView II color calibration software and a colorimeter. How important is that? Would I be able to calibrate the monitor manually without a colorimeter and special software?
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
Is a colorimeter necessary?
- Thread starter delerious
- Start date
Do you care about color accuracy? Given your monitor of choice, I guess you do. In that case you want a colorimeter (with the accompanied software) and color-managed applications.
Without a colorimeter, it is close to impossible to properly calibrate a monitor. And calibration is only half of the story (see the following link for an explanation): the other half of the story, profiling, is just as important. And profiling is possible only with a colorimeter.
http://www.hardforum.com/showpost.php?p=1034080993&postcount=76
Without a colorimeter, it is close to impossible to properly calibrate a monitor. And calibration is only half of the story (see the following link for an explanation): the other half of the story, profiling, is just as important. And profiling is possible only with a colorimeter.
http://www.hardforum.com/showpost.php?p=1034080993&postcount=76
Thanks for the info. Do I care about color accuracy? Well, I'm no graphics professional, so I may not need 100% accuracy, but obviously I would like the color to be fairly accurate. I'll be using the monitor for web browsing, programming, occasional photo editing and occasional gaming. The reason I am considering the NEC LCD2190UXi is because there really isn't much choice of 4:3 monitors anymore. I just got a 2007FP, but at 20" the text is too small. I think a 21.3" monitor will be much better.
dark_reign
2[H]4U
- Joined
- Nov 15, 2005
- Messages
- 2,314
You can calibrate the grayscale by eye. You may not get it perfect, but neither will a colorimeter.
vick1000
2[H]4U
- Joined
- Sep 15, 2007
- Messages
- 2,443
If you have to ask, you don't need it.
One thing I've read is that if you do calibrate, you should calibrate every month. Why does the color change so that it needs to be recalibrated so often? And if I don't calibrate at all, then does that mean my monitor will look pretty bad after 6 or 12 months?
If the 2190UXi is anything like the 2490UXi a calibration device is completely unnecessary IMO.
When you turn on Colorcomp on these panels you get an essentially perfect factory calibration.
Look near the bottom of this post to see the effect of colorcomp.
http://www.hardforum.com/showpost.php?p=1032929434&postcount=3
When you turn on Colorcomp on these panels you get an essentially perfect factory calibration.
Look near the bottom of this post to see the effect of colorcomp.
http://www.hardforum.com/showpost.php?p=1032929434&postcount=3
exe said:
So I guess I could just use Colorcomp to reset to factory calibration every few weeks?
That wouldn't accomplish anything. It is essentially a digital map, it isn't going to change. Resetting it will do nothing.
Also drift is greatly exaggerated. I first found out about colorcomp from someone with a couple of thousand hours on his panel and color comp was still producing a bang on calibration. 2000 hours is like 3 hours every day for 2 years.
The color curve likely won't change much for years when the backlight starts to fade and alters the basic color temperature.
The main change noticeable is a gradual drop in brightness over time, which unless you are doing professional color matching you don't need to worry about, just adjust brightness to suit what you like.
I don't calibrate my monitors much due to being lazy. Every 6 months or so, there is a definate color shift from my desktop and laptop monitors. I use them about 12 hours per day everyday. If color accuracy isn't a requirement for your work/job, then by all means don't bother with a calibrator.
How are you measuring this shift? I see/measure no such shift on my NEC. This discusion prompted me to update my review and do a new calibration. It has been 1400 hours since I last calibrated, 2680 hours since I first got it and so at least that long since the colorcomp factory calibration was entered. It is as valid today nearly 3000 hours later as it was at the beginning. LCD drift is heavily exaggerated. Certainly not the few weeks claimed above. Maybe a few years.
People can buy an NEC 2490 without a colorimeter and missing essentially nothing in terms of calibration. The factory calibration will be near perfect for years likely. This is one of the few instances where you can essentially have a factory calibrated monitor and not have to bother with calibration.
I updated page three of my review with the following info from today's calibration:
Calibration Update, at 2680 hours: It has been a while since I calibrated. About 1400 hours and there are no worries (at 2680 total hours now). Nothing visible has changed. Normal panel fading is slowly dropping brightness levels at the minimum setting. Here are the numbers and also the correction curves with NEC factory calibration and panel compensation turned on ( AKA colorcomp). As you can see with over 2600 hours on the panel, the factory calibration still remains near perfect.
I use the Gretag Colormunki and its associated software. AFAIK, color shifts is due to... physical degredation of the components of the monitor. Just because your NEC has stabilized, that doesn't apply to every other monitor made out there or does it? If you believe in what you say, then everyone that needs a calibrated monitor should just go and borrow one, set it once and be done with it 'for years'. AFAIK, you're the only person I know that uses a calibrator that is suggesting that.
It didn't just stabilize. The color curve has been stable for ~2700 hours since the beginning. I don't think it is a fluke either. What components do you think are degrading? The matrix itself is a stable component.
The backlight is the main part that wears down. But it isn't going to color shift week to week, it is gradually going lose brightness over time. Even more gradually you can expect a small shift in color temperature of the backlight.
What I meant when I asked how you know it is shifting. Is how does the software show you the change? I would like to see a visual of these shifts you are seeing. Also be a aware that screens are not that uniform and slightly different reading location will shift readings (try measuring your white balance across the screen). So you need to be careful with positioning if you are trying to measure consistently. Also back lights can take a full hour to warm up to full stability, so again it needs care if you are trying to compare measurements over time.
If you use different amount of warm up time and position the puck differently, you could see as big a shift in one day as you do in 6 months. Because you aren't actually measuring degradation, you are measuring panel inconsistencies.
To answer your question. If someone can borrow a colorimeter, IMO they are good for a year or longer, if they can be sure nothing will ever reset the graphic card LUTs(not an issue with the NEC).
I don't think anyone needs monthly calibration unless they are working in professional setup with complete with light controlled viewing station to view prints in. The people who need to calibrate often are the people with controlled soft proof lighting and similar setups. http://www.gtilite.com/soft-proofing.htm
I plan to keep tracking mine and adding it to my review. Could you do similar. I would like to see the trend.
http://www.hardforum.com/showpost.php?p=1032929434&postcount=3
So far. No real change in the color curve. But some drop in overall brightness, but again, unless you are working with softproofing under controlled lighting, your brightness should be set to what you find most comfortable for your conditions, so brightness changes do not require recalibration.
The backlight is the main part that wears down. But it isn't going to color shift week to week, it is gradually going lose brightness over time. Even more gradually you can expect a small shift in color temperature of the backlight.
What I meant when I asked how you know it is shifting. Is how does the software show you the change? I would like to see a visual of these shifts you are seeing. Also be a aware that screens are not that uniform and slightly different reading location will shift readings (try measuring your white balance across the screen). So you need to be careful with positioning if you are trying to measure consistently. Also back lights can take a full hour to warm up to full stability, so again it needs care if you are trying to compare measurements over time.
If you use different amount of warm up time and position the puck differently, you could see as big a shift in one day as you do in 6 months. Because you aren't actually measuring degradation, you are measuring panel inconsistencies.
To answer your question. If someone can borrow a colorimeter, IMO they are good for a year or longer, if they can be sure nothing will ever reset the graphic card LUTs(not an issue with the NEC).
I don't think anyone needs monthly calibration unless they are working in professional setup with complete with light controlled viewing station to view prints in. The people who need to calibrate often are the people with controlled soft proof lighting and similar setups. http://www.gtilite.com/soft-proofing.htm
I plan to keep tracking mine and adding it to my review. Could you do similar. I would like to see the trend.
http://www.hardforum.com/showpost.php?p=1032929434&postcount=3
So far. No real change in the color curve. But some drop in overall brightness, but again, unless you are working with softproofing under controlled lighting, your brightness should be set to what you find most comfortable for your conditions, so brightness changes do not require recalibration.
my photospectrometer software shows the shift from the last time I calibrate my monitor in the form of a curve graph. In addition, I get a button to see the before and after calibration, and I can see the change.
I don't know what 'matrix' you are talking about and my comments have never been about only your NEC monitor, as I do not own one. I don't know what exact components are degrading, but I can only assume all components degrade, just by a different degree. I have owned over a dozen panels over the last 10 years and all but one exhibit the color shift I experience. Some are worse than others. My warm up time is usually well in the day of use (+ a few hours) and my puck is positioned around the same area, I'm not anal enough to mark it off with a ruler. My brightness is always set to the same cd/m value between calibrations.
If you are convinced that color shifts are rare, then I can only assume you can advise these monitor makers to start sampling one of their monitors, takes several color profile, and then market them as color calibrated. Heck, they even have 'active' contrast ratios and many other gimicky features, why not add color calibration to the mix? I'm sure it would sell well, and they would only have to use a sample of one, practically a free cost.
I don't know what 'matrix' you are talking about and my comments have never been about only your NEC monitor, as I do not own one. I don't know what exact components are degrading, but I can only assume all components degrade, just by a different degree. I have owned over a dozen panels over the last 10 years and all but one exhibit the color shift I experience. Some are worse than others. My warm up time is usually well in the day of use (+ a few hours) and my puck is positioned around the same area, I'm not anal enough to mark it off with a ruler. My brightness is always set to the same cd/m value between calibrations.
If you are convinced that color shifts are rare, then I can only assume you can advise these monitor makers to start sampling one of their monitors, takes several color profile, and then market them as color calibrated. Heck, they even have 'active' contrast ratios and many other gimicky features, why not add color calibration to the mix? I'm sure it would sell well, and they would only have to use a sample of one, practically a free cost.
No I wouldn't suggest that. The NECs are individually factory calibrated and compensated. Manufacturing variations will make for slightly different properties for each monitor. But once you calibrate a screen, it is going to stay true for quite some time.
Since the OP was looking for the similar 90uxi series monitor I have and I have seen another NEC 2490 doing the exact same thing after a couple of thousand hours, I feel fairly confident in telling him that he doesn't need the calibration device and the 2490 doesn't show drift week to week. Note the 2190 does have a different panel type than the 2490 so I can't be certain it doesn't have the same solid stability of mine. But I wouldn't think NEC would bother if it had stability issues.
Something you can try. Try doing the calibration test the very next day after you calibrated it at three slightly different locations and save the difference curves(but don't calibrate). Then check again just before your next calibration interval and see if you can actually spot a trend, or is this merely measurement variation. Also can you see the overall correction curve at the end of calibration? Can you save these to spot long term trend?
Like I said. I will keep reporting mine in my review thread. Since there is a built in calibrated baseline in the monitor, I should be able to spot when a trend emerges. I will try to calibrate the NEC every 1000 hours or so, to see how it is changing over time.
If you aren't a working graphics professional and you are buying the only sRGB factory calibrated monitor I know of (NEC 90uxi sRGB panels), there is essentially no need to buy the Spectra View version unless you want another toy to play with. You will NOT see a visible difference calibrating/profiling it. It will not shift week to week.
Since the OP was looking for the similar 90uxi series monitor I have and I have seen another NEC 2490 doing the exact same thing after a couple of thousand hours, I feel fairly confident in telling him that he doesn't need the calibration device and the 2490 doesn't show drift week to week. Note the 2190 does have a different panel type than the 2490 so I can't be certain it doesn't have the same solid stability of mine. But I wouldn't think NEC would bother if it had stability issues.
Something you can try. Try doing the calibration test the very next day after you calibrated it at three slightly different locations and save the difference curves(but don't calibrate). Then check again just before your next calibration interval and see if you can actually spot a trend, or is this merely measurement variation. Also can you see the overall correction curve at the end of calibration? Can you save these to spot long term trend?
Like I said. I will keep reporting mine in my review thread. Since there is a built in calibrated baseline in the monitor, I should be able to spot when a trend emerges. I will try to calibrate the NEC every 1000 hours or so, to see how it is changing over time.
If you aren't a working graphics professional and you are buying the only sRGB factory calibrated monitor I know of (NEC 90uxi sRGB panels), there is essentially no need to buy the Spectra View version unless you want another toy to play with. You will NOT see a visible difference calibrating/profiling it. It will not shift week to week.
Last edited:
Thats intresting, one more time we are able to see/hear about NEC 90 series,superiority
useful info no doubt! After this suggestion what made me spend 200 euros more is the "hand picked" panels used on SV versions,for EU as many users reported
if i already decide to spend more then 1000 euros for this monitor i won't
refuse to burn 200 more and in some way feel more "secure" that i won't get bad unit.(I'm generally unlucky ) even when we talk about NEC 90 series. Don't want to risk at all tomorrow ill visit the other shop to check price info. The one i contact today
offer me 1550 euros price
useful info no doubt! After this suggestion what made me spend 200 euros more is the "hand picked" panels used on SV versions,for EU as many users reported
if i already decide to spend more then 1000 euros for this monitor i won't
refuse to burn 200 more and in some way feel more "secure" that i won't get bad unit.(I'm generally unlucky ) even when we talk about NEC 90 series. Don't want to risk at all tomorrow ill visit the other shop to check price info. The one i contact today
offer me 1550 euros price