Pages: [1] 2   Go Down

Author Topic: Should I Calibrate to Monitor's Native White Point?  (Read 24197 times)

CarolynC

  • Newbie
  • *
  • Offline Offline
  • Posts: 39
Should I Calibrate to Monitor's Native White Point?
« on: October 08, 2010, 09:37:48 pm »

I'm still trying to understand the best way to calibrate my new NEC monitor.  (PA241W)  I have some test prints here and it looks like I was getting the best color temp matches at 4500K (which I don't know if anyone ever goes that low.)  Then tonight, I read online that Andrew Rodney recommends calibrating at your monitors native white point.  Should I do this and why?  Also, if and when I do this and the color temps on screen don't match the temps on my prints, how should I adjust the color...by using the Visual Match feature in SpectraView?  Or, on my monitor itself?

Until reading Andrew's advice tonight on using the native white point, I've just been creating new calibrations...going down by 500k each time until the temps seemed to match my prints more.  Thanks.
« Last Edit: October 08, 2010, 11:04:13 pm by CarolynC »
Logged

Roy

  • Full Member
  • ***
  • Offline Offline
  • Posts: 196
    • http://
Re: Should I Calibrate to Monitor's Native White Point?
« Reply #1 on: October 08, 2010, 11:15:45 pm »

You should leave your new NEC monitor with its very accurate factory calibration. Use the multiprofiler app to choose the colour space the monitor is working in and to set the correct display profile on your computer. It is very unlikely that you can improve on the factory calibration for a new NEC PA monitor.

If you have such big mismatches with your prints, something is wrong with your overall workflow. The monitor calibration is the last thing you should question.

Logged
Roy

shewhorn

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 537
    • http://
Re: Should I Calibrate to Monitor's Native White Point?
« Reply #2 on: October 09, 2010, 01:31:32 am »

If you have such big mismatches with your prints, something is wrong with your overall workflow.

Well that's just not true.  ;D If you have mismatches with your prints, then your calibration is off, or you're not softproofing (or both... I guess the latter could be considered part of the workflow).

I suspect Andrew's recommendation to calibrate to the native white point has to do with the fact that this is the least destructive in terms of what changes need to be made to the profile. Profiling to the native white point retains the maximum resolution and thus helps to avoid issues with posterization and such. The further away you get from a monitor's native color temperature, the more likely you are to experience anomalies due to bending it away from its native point.

If you're just adjusting RAW files then the color temperature doesn't make a big difference due to something known as chromatic adaptation (it's one of those little evolutionary things that allows us to visually identify that the food we're about to put in our mouth is rotten). If however, you are trying to match your screen to a print it makes a HUGE difference. The color temperature to which you adjust your screen to is going to depend upon the light source you use (you'll want to be using a full spectrum light source such as those offered by GTI and SoLux). If you're using a regular store bought bulb it's unlikely you have something that outputs a full spectrum (even if it says it does) and that means some colors won't reproduce at all, and some will be too hot. Anyhow... when you put a piece of blank white paper next to your screen and light it up with your light source and you display a blank white page on your screen, the temperature of the two should look pretty close. Note that you should also be using the soft proofing profile from your lab.

Depending upon what puck you have, the best match may very well be at 4500ºK for your puck but just because a puck reports a color temp of 4500ºK doesn't mean it actually is 4500ºK. Just using different software, even with the same puck is enough to get a different value reported back. Colorimeters are a lot of things but when it comes to reporting objective numbers for color temp, accuracy is not something they're known for. I have a DTP94 and a Spyder 3 and they're 1800ºK apart from one another for the exact same color temp. If you have a spectrophotometer (ColorMunki, Eye One Pro, etc.), then you're probably getting a fairly accurate reading. The number doesn't matter though... they could call it bananas for all I care. If you have a spectrophotometer and some more advanced software tools you could take an ambient light reading of the color temperature getting bounced off of your paper and use that as the basis for your monitor color temp. Eyeballing it should get you there too though.

Cheers, Joe
Logged

probep

  • Full Member
  • ***
  • Offline Offline
  • Posts: 149
Re: Should I Calibrate to Monitor's Native White Point?
« Reply #3 on: October 09, 2010, 01:59:21 am »

1. The white point color temperature on a new NEC PA display is 6500 K. And Roy is right: PA-series displays are very accurate factory calibrated.
2. I am with shewhorn - the color temperature difference between one of my Spyder3 and one of my i1Display2 reaches 1800-1900 K. For example, sensors show:
 i1Diasplay 2 - 5709 K
 Spyder3 - 7507 K
 i1Pro - 6338 K
 i1Pro UVcut - 6377 K
 NEC custom calibrated i1D2 - 6350 K
« Last Edit: October 09, 2010, 02:19:22 am by probep »
Logged

shewhorn

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 537
    • http://
Re: Should I Calibrate to Monitor's Native White Point?
« Reply #4 on: October 09, 2010, 02:45:18 am »

1. The white point color temperature on a new NEC PA display is 6500 K. And Roy is right: PA-series displays are very accurate factory calibrated.

For clarity I agree with this as well. In addition I may be misunderstanding what Roy is saying.

Quote
Use the multiprofiler app to choose the colour space the monitor is working in and to set the correct display profile on your computer.

Roy, I'm not quite sure what you mean there with regards to the italics. There is no preset in multi-profiler that will provide an accurate screen-to-print match because mutliprofiler has no idea what color temp your light source is. You can of course create a custom preset in MP with a color temp that will work.

Cheers, Joe
Logged

CarolynC

  • Newbie
  • *
  • Offline Offline
  • Posts: 39
Re: Should I Calibrate to Monitor's Native White Point?
« Reply #5 on: October 09, 2010, 10:47:12 am »

Thank you for all of the information.  

You should leave your new NEC monitor with its very accurate factory calibration. Use the multiprofiler app to choose the colour space the monitor is working in and to set the correct display profile on your computer. It is very unlikely that you can improve on the factory calibration for a new NEC PA monitor.

How do I get back to the factory calibration?  If I simply calibrate at 6500k is that good enough?  Or, there are a couple factory setup calibrations to select from in SpectraView as follows...

Photo Editing - D65, Gamma 2.2, Intensity 140 cd/m, Contrast Ratio: Monitor Default
Print Standard - D50, Gamma 1.8, Intensity 80, Contrast Ratio: Monitor Default

Do I choose one of these or should I choose one, then click Edit and select Native from the White Point drop down menu?  I'm confused because I did this this morning and when I chose Native, it didn't allow me to enter an Intensity value that I wanted, I could only select Maximum Possible.  


Onto my lighting situation -

I know most of you might cringe at me saying this but I'm trying to get screen-to-print matches in a room that is very uncontrolled as far as lighting goes.  During the day, it's very bright in here, no direct sunlight in the windows but it's extremely bright because of a lot of windows.  The room doesn't even have shades or curtains and to be totally honest, I used all the money I DON'T have to get this NEC monitor.  Now reading that I need special lights and realizing I need to buy shades or curtains too is overwhelming me because I don't have any extra money to spare now.  I also don't have an extra room I can move to, this is the only space I have to set everything up.  I think what I'm telling myself is I'm just trying to get as close screen-to-print matches as I can for now.  My goal is to get an online shop going to try and sell some fine art prints.  THEN I would have the money for lights, etc.  See how BACKWARD my crazy plan is??!?  I thought the monitor was the last thing I'd have to buy in order to make this happen, besides of course paying for lab prints.  (WHCC)  I'm new to this whole color management world and I guess I'm just hoping to get very close screen-to-print matches for now somehow.

The thing that's driving me nuts is I measured the ambient light in the room I'm in and it changes all throughout the day.  This morning, it was like 4700k, then yesterday at 4pm, it was around 8000k.  I even started thinking, "well, maybe if I see the ambient light is around 5000k at one time in the day, then I can trust viewing my prints at that time of day when comparing them to the monitor."  I know, crazy.  Since I'm broke, is this nuts to do this temporarily??  Trust what my prints look like if the ambient light in the room is at a certain temperature during the day?
« Last Edit: October 09, 2010, 11:17:50 am by CarolynC »
Logged

TKTeo

  • Newbie
  • *
  • Offline Offline
  • Posts: 6
Re: Should I Calibrate to Monitor's Native White Point?
« Reply #6 on: October 09, 2010, 11:08:49 am »

2. I am with shewhorn - the color temperature difference between one of my Spyder3 and one of my i1Display2 reaches 1800-1900 K. For example, sensors show:
 i1Diasplay 2 - 5709 K
 Spyder3 - 7507 K
 i1Pro - 6338 K
 i1Pro UVcut - 6377 K
 NEC custom calibrated i1D2 - 6350 K
Is it possible to use the NEC custom calibrated i1 Display 2 on non-NEC monitors?
Logged

shewhorn

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 537
    • http://
Re: Should I Calibrate to Monitor's Native White Point?
« Reply #7 on: October 09, 2010, 11:38:53 am »

Thank you for all of the information.  

How do I get back to the factory calibration?  If I simply calibrate at 6500k is that good enough?  Or, there are a couple factory setup calibrations to select from in SpectraView as follows...

As suggested by Roy, try multi-profiler.

http://www.necdisplay.com/MultiProfiler/downloads/ 

Quote
Onto my lighting situation -

I know most of you might cringe at me saying this but I'm trying to get screen-to-print matches in a room that is very uncontrolled as far as lighting goes.  During the day, it's very bright in here, no direct sunlight in the windows but it's extremely bright because of a lot of windows.  The room doesn't even have shades or curtains and to be totally honest, I used all the money I DON'T have to get this NEC monitor.  Now reading that I need special lights and realizing I need to buy shades or curtains too is overwhelming me because I don't have any extra money to spare now.

The reason we cringe is because getting a screen-to-print match without controlled lighting simply put is not terribly practical. Why (ETA: haha... you've figured it out already)? I already mentioned the issue of peaks and dips with regular lighting. Then you have natural light. Natural light changes color temperature throughout the day. I use SolUx MR16 bulbs, specifically the 4700K 50 watt black back bulbs with a 36º spread. They're a bargain at $16 bucks (the GTI booths typically cost over $1000). The bigger trick to the MR16s though is finding a fixture that accepts them.

https://www.solux.net/cgi-bin/tlistore/soluxbulbs.html

You can get some vinyl light blocking shades from target or just do what I did (which is EXTREMELY ghetto because I was too lazy to go to target when I moved into my new place)... cut up a giant cardboard box and stuff it in the window!  :D


Quote
The thing that's driving me nuts is I measured the ambient light in the room I'm in and it changes all throughout the day.  This morning, it was like 4700k, then yesterday at 4pm, it was around 8000k.  I even started thinking, "well, maybe if I see the ambient light is around 5000k at one time in the day, then I can trust viewing my prints at that time of day when comparing them to the monitor."  I know, crazy.  Since I'm broke, is this nuts to do this temporarily??  Trust what my prints look like if the ambient light in the room is at a certain temperature during the day?

Haha... hadn't read this part... started replying first. Well yeah, you've figured out the problem. I haven't tried that myself so I'm not sure how practical that is. My concern would be how big a of a window (in terms of time... not your actual window  ;D) do you have and how much ambient light is in that room? A larger concern though would be losing shadow detail due to the bright ambient light in the room. You could overcome that to a certain extent by increasing the luminance but then highlights start to suffer. You'll have to increase the luminance most likely anyway in order to match that of your print.

Cheers, Joe
Logged

shewhorn

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 537
    • http://
Re: Should I Calibrate to Monitor's Native White Point?
« Reply #8 on: October 09, 2010, 11:39:19 am »

Is it possible to use the NEC custom calibrated i1 Display 2 on non-NEC monitors?

Yes.
Logged

probep

  • Full Member
  • ***
  • Offline Offline
  • Posts: 149
Re: Should I Calibrate to Monitor's Native White Point?
« Reply #9 on: October 09, 2010, 11:55:41 am »

Is it possible to use the NEC custom calibrated i1 Display 2 on non-NEC monitors?
NEC Colorimeters do work with 3rd party programs as regular out-of-shelf X-Rite i1Display 2 colorimeters.
NEC colorimeters are custom calibrated only for SpectraView II and NEC displays.
Logged

CarolynC

  • Newbie
  • *
  • Offline Offline
  • Posts: 39
Re: Should I Calibrate to Monitor's Native White Point?
« Reply #10 on: October 09, 2010, 12:01:59 pm »

Quote
I use SolUx MR16 bulbs, specifically the 4700K 50 watt black back bulbs with a 36º spread. They're a bargain at $16 bucks (the GTI booths typically cost over $1000). The bigger trick to the MR16s though is finding a fixture that accepts them.

How many do you think I need and what kind of fixture do you use??  Do I have to buy the track lighting or do some people on a tight budget maybe just use one or two of these bulbs in some other kind of fixture?  

Quote
You can get some vinyl light blocking shades from target or just do what I did (which is EXTREMELY ghetto because I was too lazy to go to target when I moved into my new place)... cut up a giant cardboard box and stuff it in the window!  :D

HA!  Gotta love the ghetto style.  I guess since I'm a girl and into the way my house looks, I'd rather get the vinyl shades from Target (which are still gonna look bad to me I'm sure) than use cardboard!

Quote
Haha... hadn't read this part... started replying first. Well yeah, you've figured out the problem. I haven't tried that myself so I'm not sure how practical that is. My concern would be how big a of a window (in terms of time... not your actual window  ;D) do you have and how much ambient light is in that room? A larger concern though would be losing shadow detail due to the bright ambient light in the room. You could overcome that to a certain extent by increasing the luminance but then highlights start to suffer. You'll have to increase the luminance most likely anyway in order to match that of your print.

Yeah, I hate the idea of limiting myself to such a narrow window of time to edit each day...don't what to work that way.  So does anyone ever edit photos and get good screen-to-print matches without shutting the shades??

Quote
Cheers, Joe

Thank you so much for helping me, I need it!  I'm feeling very lost here and just need some guidance.  I just want to get this all straightened out so I can move on and get to the fun stuff again.   :)
Logged

shewhorn

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 537
    • http://
Re: Should I Calibrate to Monitor's Native White Point?
« Reply #11 on: October 09, 2010, 12:24:40 pm »

NEC colorimeters are custom calibrated only for SpectraView II and NEC displays.

They aren't custom calibrated "only" for SVII and NEC displays. They are custom calibrated to provide better accuracy in terms of white point (at least I've noticed they tend to be a lot more consistent against a spectrophotometer vs. regular off the shelf i1D2s) and they've supposedly tweaked them to work better with wide gamut displays (perhaps they've replaced the filter with something of a higher quality and made some firmware tweaks?). These changes would be of benefit to ALL displays. There's nothing special about NEC displays that would cause the NEC i1D2 puck to not work with any other display. NEC uses the same panels that many other manufacturers do so there's nothing really unique about them. What sets them apart from other manufacturers is the uniformity of the backlighting, and the 14 bit monitor LUTs (and with the PA series, the factory characterization).

Cheers, Joe
Logged

shewhorn

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 537
    • http://
Re: Should I Calibrate to Monitor's Native White Point?
« Reply #12 on: October 09, 2010, 12:37:39 pm »

How many do you think I need and what kind of fixture do you use??  Do I have to buy the track lighting or do some people on a tight budget maybe just use one or two of these bulbs in some other kind of fixture?  

I use just a single bulb for most stuff. It depends on the size of the print. Re: fixtures I bought their basic task lamp. Track lighting fixtures may actually be the cheapest option if you can find an MR16 track at the local hardware store.

Quote
HA!  Gotta love the ghetto style.  I guess since I'm a girl and into the way my house looks, I'd rather get the vinyl shades from Target (which are still gonna look bad to me I'm sure) than use cardboard!

:D We don't let people into the office. It's a disaster area!

Quote
Yeah, I hate the idea of limiting myself to such a narrow window of time to edit each day...don't what to work that way.  So does anyone ever edit photos and get good screen-to-print matches without shutting the shades??

Sssshhhh.... don't tell anyone (looks around to make sure nobody is looking).... [WHISPERS] I don't screen-to-print proofing for everything [/WHISPERS]  I think you'll find that once you get to know a certain lab or printer/paper combination you'll get a good feel for what it's going to do. When I'm working with a new paper, I do A LOT of soft proofing and testing so I can get a good understanding for how a device and paper combo behaves. After I know that "hey, this particular paper doesn't really do well reproducing shadow detail below 6,6,6" then I can do what I need to do to deal with that. I'll still soft proof if I'm want to see what a certain area of the print is doing but that doesn't always require a print and if it does, there's a chance that the print is going to be off.

Now, that said most of what I do is wedding and portrait photography. I'm concerned with skin tones (most of the time, not always). Once I have one right, I can go off that one for everything else. If I was doing advertising or other commercial work where color accuracy is extremely important then I'd be far more critical.

Cheers, Joe
Logged

probep

  • Full Member
  • ***
  • Offline Offline
  • Posts: 149
Re: Should I Calibrate to Monitor's Native White Point?
« Reply #13 on: October 09, 2010, 12:47:13 pm »

They aren't custom calibrated "only" for SVII and NEC displays. They are custom calibrated to provide better accuracy in terms of white point (at least I've noticed they tend to be a lot more consistent against a spectrophotometer vs. regular off the shelf i1D2s) and they've supposedly tweaked them to work better with wide gamut displays (perhaps they've replaced the filter with something of a higher quality and made some firmware tweaks?). These changes would be of benefit to ALL displays. There's nothing special about NEC displays that would cause the NEC i1D2 puck to not work with any other display. NEC uses the same panels that many other manufacturers do so there's nothing really unique about them. What sets them apart from other manufacturers is the uniformity of the backlighting, and the 14 bit monitor LUTs (and with the PA series, the factory characterization).

Cheers, Joe
1. The quote from NEC FAQ:
Quote
QUESTION: Can I use the new MDSVSENSOR2 color sensor included in the new SVII-PRO-KIT with other 3rd party calibration applications?
ANSWER: Yes, however the custom calibration for NEC wide color gamut displays will not be available. Only the standard calibration is available.
2. I've tested NEC i1D2 colorimeters in basICColor display, ColorEyes Display Pro, X-Rite ProfileMaker, ArgyllCMS, Quato iColor Display, X-Rite i1Match, LaCie blue eye pro. Yes they are more accurate than out-of-shelf i1Display 2 colorimeters but are still  far off i1Pros.
And only in SpectraView II the NEC colorimeters and the i1Pros show almost the same values.

Addition. I've tested NEC i1D2s on wide gamut displays: NEC 2690WUXi2 and HP LP2475w
« Last Edit: October 09, 2010, 01:04:01 pm by probep »
Logged

shewhorn

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 537
    • http://
Re: Should I Calibrate to Monitor's Native White Point?
« Reply #14 on: October 09, 2010, 01:07:20 pm »

Thanks for the clarification. I misread your post as saying the MDSVSENSORs would not work with other displays because of NEC changes.

I question whether the tweaks made are ONLY available to NEC monitors. This could easily be done in software when SVII takes ownership of the puck thus preventing other apps from benefiting however (and I think you mentioned this although I'm not sure if the comment was related only to SVII or all the other apps you tried as well) it does seem that they are a bit more accurate in terms of reporting color temp and luminance.

Cheers, Joe
Logged

CarolynC

  • Newbie
  • *
  • Offline Offline
  • Posts: 39
Re: Should I Calibrate to Monitor's Native White Point?
« Reply #15 on: October 09, 2010, 01:16:30 pm »

I use just a single bulb for most stuff. It depends on the size of the print. Re: fixtures I bought their basic task lamp.

Is this your task lamp?  https://www.solux.net/cgi-bin/tlistore/draftinglight.html

I replied to you in my other thread about the NEC MultiProfiler.  I just read somewhere on this forum that instead of setting Picture Mode to AdobeRGB, I should set it to Full, use the Preset Color Gamut, "Native Gamut."  Then I chose the Preset White Point, 6500k.  Are these the right choices?  Also confused on what I should set my Brightness and Black Level at and if I should check Ambient Light Compensation.  I guess on this last screen, where I choose brightness, black level and ambient light comp, I can't really choose the right levels until I get my window shades and lamp, true?

By the way, is the lamp's sole purpose for viewing your prints or does the lamp also have an affect on how you see colors on the screen?  If so, where do you place the lamp in relation to the screen?
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20646
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Should I Calibrate to Monitor's Native White Point?
« Reply #16 on: October 09, 2010, 01:29:33 pm »

IF you have a run of the mill, non high bit display, start with Native White Point. If you deviate from that, you can find more banding as a result. If however you find the screen to print matching (after nailing the luminance, that’s most critical) is too warm or cool, you can adjust WB to taste understanding that you are going to get more banding as you move farther from the native WP.

If you have a high bit display panel, then you can get away moving away from Native WP without the banding issues. Start with Native (or D65 which should be a good starting point), again decide if you need to season to taste. All this of course depends highly on the illuminant you use, next to the display to view the prints!
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

CarolynC

  • Newbie
  • *
  • Offline Offline
  • Posts: 39
Re: Should I Calibrate to Monitor's Native White Point?
« Reply #17 on: October 09, 2010, 01:37:10 pm »

IF you have a run of the mill, non high bit display, start with Native White Point. If you deviate from that, you can find more banding as a result. If however you find the screen to print matching (after nailing the luminance, that’s most critical) is too warm or cool, you can adjust WB to taste understanding that you are going to get more banding as you move farther from the native WP.

If you have a high bit display panel, then you can get away moving away from Native WP without the banding issues. Start with Native (or D65 which should be a good starting point), again decide if you need to season to taste. All this of course depends highly on the illuminant you use, next to the display to view the prints!

My NEC PA241w is a high bit display, correct?  Sorry for my extreme ignorance.
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20646
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Should I Calibrate to Monitor's Native White Point?
« Reply #18 on: October 09, 2010, 01:39:10 pm »

My NEC PA241w is a high bit display, correct? 

Yup, 14-bit.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

CarolynC

  • Newbie
  • *
  • Offline Offline
  • Posts: 39
Re: Should I Calibrate to Monitor's Native White Point?
« Reply #19 on: October 09, 2010, 01:50:03 pm »

Yup, 14-bit.

Cool.  So it really isn't a problem to have my White Point set at 4500k?  I think what is confusing me is when I read online that you suggest keeping things at their Native White Point but I think I understand now why you tell people that.  I do want to avoid banding for sure but sounds like I won't have a problem on this monitor going further away from the native white point.  (i hope)

I have a question for you digital dog - Since I'm using SpectraView to set my white point at 4500k, what option should I choose in MultiProfiler in Picture Mode, Adobe RGB or Full?
Logged
Pages: [1] 2   Go Up