Pages: [1]   Go Down

Author Topic: NEC PA271W SpectraView II Question  (Read 8184 times)

Ken W

  • Newbie
  • *
  • Offline Offline
  • Posts: 25
    • ioScapes
NEC PA271W SpectraView II Question
« on: April 19, 2011, 04:25:49 pm »

Last Fall I started using a NEC PA271W SpectraView II display. I used the NEC branded sensor to do the internal calibration. I assumed this would do a better job than the Spyder 3 Pro that I had used with my previous display. Out of curiosity I decided to do an internal calibration using the Spyder 3 Pro. I am not sure if I see any obvious differences. The display looks very nice using either calibration.  I compared some of the numbers that the Spectraview software provides after the calibration. I do not not understand a lot about these numbers and asked the NEC tech rep, who told me a lower Delta E is better and that anything between 1 and 3 is very good. Here is the comparison between the two sensors:

Nec branded: Delta E: 1.78
Spyder 3 Pro: Delta E: 0.48

Based upon this, it appears the Spyder 3 Pro is doing a better job? I wonder if this is true since the NEC branded sensor is supposed to be calibrated for the NEC wide gamut display. Is there any thing else to consider?
Logged

kers

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4391
    • Pieter Kers
Re: NEC PA271W SpectraView II Question
« Reply #1 on: April 19, 2011, 06:59:07 pm »

The question is : are the two numbers correct?

You would have tot test it with a reference high quality calibrated spectrophotometer...

Then you know what to believe...
Logged
Pieter Kers
www.beeld.nu/la

Ken W

  • Newbie
  • *
  • Offline Offline
  • Posts: 25
    • ioScapes
Re: NEC PA271W SpectraView II Question
« Reply #2 on: April 19, 2011, 10:27:11 pm »

Thanks for your response. If you made the assumption that the NEC/X-Rite iOne Display 2 colorimeter and the Spyder 3 Pro were operating properly within their standard specification, would these numbers be meaningful. If not, I wonder why NEC provides them after performing a display calibration? Also, does this mean 1. Properly functioning colorimeters are poor instruments for doing the calibration or 2. Properly functioning colorimeters do a good job of calibrating, but do not have the level of accuracy required to provide meaningful Delta E numbers?
Logged

WillH

  • Full Member
  • ***
  • Offline Offline
  • Posts: 144
    • http://www.necdisplay.com
Re: NEC PA271W SpectraView II Question
« Reply #3 on: April 20, 2011, 12:27:35 am »

The delta-e values will vary slightly from one calibration to the next. It takes only a very slight error to result in what appears to be a large delta-e value. Variations can be due to noise, slight drifts in the display and sensor as they warm up etc.

Keep in mind that the delta-e is comparing your Target (what you wanted to achieve) to the actual measurements after the calibration process is completed. Since the adjustment and result measurements are being taken by the same sensor, the delta-e value does not give any real indication of how "accurate" the sensor is. You could have a totally out-of-whack sensor, that when calibrated measures exactly what your Target intent was, and you would end up with a great delta-e value. However compared to what the "real" results are, as measured by a correctly functioning sensor, the delta-e would be huge.
Logged
Will Hollingworth
Senior Manager, Product Development
NEC Display Solutions of America, Inc.

Ken W

  • Newbie
  • *
  • Offline Offline
  • Posts: 25
    • ioScapes
Re: NEC PA271W SpectraView II Question
« Reply #4 on: April 20, 2011, 08:14:55 am »

Will, thanks for your post. I still have the same question, if the sensors are properly adjusted according to factory specs, are the numbers meaningful? According to NEC the "NEC/X-rite iOne Display 2-based colorimeter features factory custom calibration for improved color accuracy for NEC wide gamut displays when used with the SpectraView software". What exactly does this mean? Can I assume then at least that the NEC sensor that came with the PA271W is properly calibrated and that the numbers the SpectraView software gives are meaningful?
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20651
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: NEC PA271W SpectraView II Question
« Reply #5 on: April 20, 2011, 03:06:52 pm »

Keep in mind that the delta-e is comparing your Target (what you wanted to achieve) to the actual measurements after the calibration process is completed. Since the adjustment and result measurements are being taken by the same sensor, the delta-e value does not give any real indication of how "accurate" the sensor is. You could have a totally out-of-whack sensor, that when calibrated measures exactly what your Target intent was, and you would end up with a great delta-e value. However compared to what the "real" results are, as measured by a correctly functioning sensor, the delta-e would be huge.

Which is why I’ve been critical of these kinds of reports, from all nature of differing software products. I really think at best, they are a feel good report. At worst, it just confuses users.
The reports are kind of useful if you get a really awful dE value because, I don’t know, the puck fell off during the measurement (one would probably see that) or there was light leakage. And with such high values, you’d expect the previews to look awful. But over the years, more users have been sidetracked by this kind of ‘feature’ than anything else. It seems to be somewhat wasted engineering efforts unless someone can explain otherwise.

Now trending, that’s useful. I like to see that my delta’s have gone up because instead of calibrating once a month, I waited 2 months. Being able to see what the device does over time, using the same instrument of course is nice to see.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Ken W

  • Newbie
  • *
  • Offline Offline
  • Posts: 25
    • ioScapes
Re: NEC PA271W SpectraView II Question
« Reply #6 on: April 20, 2011, 03:33:27 pm »

Thanks Andrew for your comments. I like the idea of tracking over time. You mentioned liking to see the delta's having gone up? I thought a lower delta is better, or are you saying that it would be natural over time for the delta to go up?

Prior to trying out the Spyder 3 Pro a few days ago I had not looked at these numbers. I was very happy with the display using the supplied NEC/X-rite iOne Display 2 sensor for calibration. My curiosity got the better of me and in a moment of relative down time and I decided to try calibrating with the Spyder 3. Since I wasn't sure if I saw a visual difference, the numerical data caught my attention. Now I have learned some things I did not know.

Thanks and best wishes
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20651
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: NEC PA271W SpectraView II Question
« Reply #7 on: April 20, 2011, 03:37:21 pm »

I like the idea of tracking over time. You mentioned liking to see the delta's having gone up? I thought a lower delta is better, or are you saying that it would be natural over time for the delta to go up?

So the idea is, how often should you recalibrate? You do this once a month and the trending shows you that between calibration, the dE is 3, maybe a bit too high. So you then do this every week, the value goes down to dE2. Great. Calibrate more frequently. Then you go every other week, you find you are still at dE2. Once a week is over kill. Or you find that over time, the values go up. Maybe you are trying to hit a target that isn’t reproducible as the display ages.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Ken W

  • Newbie
  • *
  • Offline Offline
  • Posts: 25
    • ioScapes
Re: NEC PA271W SpectraView II Question
« Reply #8 on: April 20, 2011, 05:29:11 pm »

This is very interesting. I always assumed, I guess incorrectly, that these flat panel displays are fairly stable and I wondered if it was even necessary to recalibrate every month. I can see that by referencing the numbers the SpectraView software provides after calibrating, that the trend will give me an idea of how stable the display is and the frequency that I need to calibrate. I am thinking correctly?
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20651
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: NEC PA271W SpectraView II Question
« Reply #9 on: April 20, 2011, 06:18:39 pm »

This is very interesting. I always assumed, I guess incorrectly, that these flat panel displays are fairly stable and I wondered if it was even necessary to recalibrate every month. I can see that by referencing the numbers the SpectraView software provides after calibrating, that the trend will give me an idea of how stable the display is and the frequency that I need to calibrate. I am thinking correctly?

If that were the case, we wouldn’t have to calibrate them at all (assuming the calibration in effect is one we desire and provides a print to screen match).
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Ken W

  • Newbie
  • *
  • Offline Offline
  • Posts: 25
    • ioScapes
Re: NEC PA271W SpectraView II Question
« Reply #10 on: April 20, 2011, 07:58:37 pm »

I was thinking fairly stable, so maybe it wouldn't be necessary to calibrate monthly, but perhaps every three months. Since the calibration process is so simple on the NEC display, I guess I will calibrate every 2 to 4 weeks.

thanks for your help
Logged

julianv

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 55
Re: NEC PA271W SpectraView II Question
« Reply #11 on: May 06, 2011, 07:25:47 am »

Ken, I am curious to know how you switched between the NEC and Datacolor calibration products.  I have used both, although not installed at the same time on the same computer.  I suspect that both products install background processes that monitor the state of the LUTs on the video card.  If I was running comparisons, I would be careful to uninstall one product and reboot before installing the other.  Otherwise, there might be some sort of interaction between them.

Can anyone provide more authoritative info on this?
Logged

Ken W

  • Newbie
  • *
  • Offline Offline
  • Posts: 25
    • ioScapes
Re: NEC PA271W SpectraView II Question
« Reply #12 on: May 07, 2011, 07:08:31 am »

I do not have the Datacolor Spyder 3Pro software installed. So I do not believe there is a conflict. My understanding is that the NEC SpectraView II software does not modify the graphic card LUT to obtain proper color calibration, but instead directly adjusts the monitors internal settings.

When I plug the Spyder 3Pro in to my computer, instead of the NEC/X-rite sensor, the SpectraView software automatically recognizes the Spyder 3Pro and uses it to measure the display and adjust the monitors internal settings. The SpectraView software does create an ICC profile, that I assumes is loaded when the computer boots up and works in conjunction with the displays internal calibration. Only one ICC profile can be utilized at a time, so I do not believe there is a conflict.

Perhaps some one else has a better understanding of this?
Logged

julianv

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 55
Re: NEC PA271W SpectraView II Question
« Reply #13 on: May 07, 2011, 07:15:49 am »

The NEC software does load the graphics card LUT - but not in the usual way.  It loads a linear ramp in the card, and handles the (calibrated) gamma curves in the monitor.  I did not realize that you were using the Spyder with the NEC software.  That should work OK, assuming that NEC allows for it, and did their homework correctly.  I thought that you were switching between the NEC software (with their sensor), and the Datacolor software (with their sensor).
Logged

Ken W

  • Newbie
  • *
  • Offline Offline
  • Posts: 25
    • ioScapes
Re: NEC PA271W SpectraView II Question
« Reply #14 on: May 07, 2011, 09:46:47 am »

The NEC website lists SpectraView II compatible sensors, and the Spyder 3Pro was one they listed, so I tried it out.
Logged
Pages: [1]   Go Up