Pages: [1]   Go Down

Author Topic: NEC SpectraViewII & LCD3090  (Read 3907 times)

walter.sk

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1433
NEC SpectraViewII & LCD3090
« on: April 11, 2010, 11:38:01 am »

I have been having a great time with my NEC 3090 and the Spectraview software together with my Eye-one Display II older version colorimeter, using the Automatic setting where SpectraView determines whether to use the factory values for the primaries or not.  Since I use a viewing box with dimmable brightness, the prints I get with homebuilt profiles from my Z3100 are close enough for jazz, so to speak.

I do have one question, though.  I calibrate and profile the monitor monthly, using D65, 120cd/m2, gamma of 2.2, Best Grayscale, 32 steps, extended stabilization time and the setting for luminance uniformity.  On the first run, I get a Delta E of 1.5 or so with the White Point, a luminance setting of as low as 117 instead of 120, and a Delta E of greater than 1 in the color accuracy.  I know these are acceptable values, but I find that if I then do a second run, all of the differences are much, much smaller.  I have experimented many times with the same results.  I always do the calibration and profiling after using the monitor for a couple of hours, so I know it has nothing to do with warmup of the monitor.  Could the colorimeter have to warm up by being used?  I can't think of any explanation for this other than chance, but the differences are always in the same direction.
Logged

Paul Sumi

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1217
NEC SpectraViewII & LCD3090
« Reply #1 on: April 11, 2010, 11:59:09 am »

Quote from: walter.sk
I always do the calibration and profiling after using the monitor for a couple of hours, so I know it has nothing to do with warmup of the monitor.  Could the colorimeter have to warm up by being used?  I can't think of any explanation for this other than chance, but the differences are always in the same direction.

Walter,

I am experiencing the same phenomenon (2690 Mk1 and X-Rite DTP94 puck), getting a lower Delta E on the second run.  I plug in the colorimeter for at least 30 minutes before calibrating.  I don't know if this warms the puck up or if only actual use makes any difference.

In any case I still get very good calibrations for my prints so I've not pursued this any further.

Paul

Logged

bossanova808

  • Newbie
  • *
  • Offline Offline
  • Posts: 47
NEC SpectraViewII & LCD3090
« Reply #2 on: April 11, 2010, 11:03:37 pm »

Calibration software iteratively measures, adjusts, re-measures.  It does this for a finite amount of time, then finishes.

It stands to reason on the first run through, you're further from perfection.  Each run through that follows you get closer given the same amount of effort.

This is actually mentioned in the SpectraView I documentation that comes with the Euro models (i.e. best results are obtained by calibrating twice or more).

It's not an issue so much as a reality - if you want a reasonably quick calibration, there's a limit to what can be achieved.  If you want better results, you need to run through it more than once.  It would be better if there was a way in the software to say 'just sit there and work for an hour and then be perfect!' but there isn't...
« Last Edit: April 11, 2010, 11:04:23 pm by bossanova808 »
Logged

WillH

  • Full Member
  • ***
  • Offline Offline
  • Posts: 144
    • http://www.necdisplay.com
NEC SpectraViewII & LCD3090
« Reply #3 on: April 12, 2010, 12:15:41 am »

Quote from: walter.sk
I have been having a great time with my NEC 3090 and the Spectraview software together with my Eye-one Display II older version colorimeter, using the Automatic setting where SpectraView determines whether to use the factory values for the primaries or not.  Since I use a viewing box with dimmable brightness, the prints I get with homebuilt profiles from my Z3100 are close enough for jazz, so to speak.

I do have one question, though.  I calibrate and profile the monitor monthly, using D65, 120cd/m2, gamma of 2.2, Best Grayscale, 32 steps, extended stabilization time and the setting for luminance uniformity.  On the first run, I get a Delta E of 1.5 or so with the White Point, a luminance setting of as low as 117 instead of 120, and a Delta E of greater than 1 in the color accuracy.  I know these are acceptable values, but I find that if I then do a second run, all of the differences are much, much smaller.  I have experimented many times with the same results.  I always do the calibration and profiling after using the monitor for a couple of hours, so I know it has nothing to do with warmup of the monitor.  Could the colorimeter have to warm up by being used?  I can't think of any explanation for this other than chance, but the differences are always in the same direction.


What is happening is that any change you make to the luminance takes time to re-establish thermal equilibrium. Changing the luminance during calibration changes the amount of power going into the backlight, which in turn changes the temperature, which in turn changes the conversion efficiency (electrical power -> light output). By re-calibrating you are starting off at a point closer to what your final target is, so there is less change.
Logged
Will Hollingworth
Senior Manager, Product Development
NEC Display Solutions of America, Inc.

walter.sk

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1433
NEC SpectraViewII & LCD3090
« Reply #4 on: April 12, 2010, 03:53:04 pm »

Quote from: WillH
What is happening is that any change you make to the luminance takes time to re-establish thermal equilibrium. Changing the luminance during calibration changes the amount of power going into the backlight, which in turn changes the temperature, which in turn changes the conversion efficiency (electrical power -> light output). By re-calibrating you are starting off at a point closer to what your final target is, so there is less change.
Thank you, Bossanova & WillH. This was not an issue for me, just a question, and I am grateful for the answer.  WillH:  would I get better results were I to raise the number of calibration and profile steps from my currently selected 32 steps, up to the maximum of 52 steps?
Logged
Pages: [1]   Go Up