- This topic has 19 replies, 4 voices, and was last updated 3 years, 11 months ago by
.
-
Posts
-
After a lot of searching around these forums and elsewhere, I’m still a bit confused about the “whitepoint” settings.
I’m running an Eizo and a macbook pro as a dual monitor setup. I do most of my color critical work on the Eizo, but I would prefer to have both monitors well calibrated.
When I calibrate the Macbook with Whitepoint set to “as measured” and then run a measurement report, I get a whitepoint of around 7300K. As a novice to this, it was a bit of a surprise (I assumed it would be closer to 6500).
It seems like my options are to
- Leave as is. I get a “NOT OK” warning on the Measured vs. assumed target whitepoint in the measurement report, but I could effectively ignore this. The whitepoint on my side by side monitors will be different, but perhaps this is not critical?
- Set the whitepoint to 6500 when calibrating (under the calibration tab). My concern with this is that I read some posts saying that calibrating to a “non native” whitepoint can be detrimental to the profile. Not sure if pushing it from 7300K “native ” to 6500K is that big of a deal?
Would love any insight anyone might have on this. Thank you!
If you get a NOT OK on “Measured vs. assumed target whitepoint” it is caused by display whitepoint not being white (daylight white).
It is not caused by being a cool white arround 7300K. A cool white <1dE to daylight 7300K (intersection between those 2 curves) will be reported as perfectly fine.
Assumed = close to reference curve, whatever CCT it is.If it is not a “white” whitepoint it coud be caused by:
-poor factory calibration. Native white= max contrast of that panel for that backlight, so most manufacturers leave display white point at native white.
-user error while measuring, for example setting a wrong spectral correction for their colorimeter, or not choosing one.
-poor accuracy in measurement deviceThere is a 3rd option between “native white” and D65 white for a laptop. It is choosing as whitepoint the closest daylight white to native white. That white would be white (no green, no pink) but may be cooler than D65. The contrast drop by using that white should be small.
A 4th option would be one of the daylight whites between closest daylight to native and D65, if 3rd option white looks too blue.A new macbook after display warmup would be close to daylight curve (measured vs assumed = OK, or less than a few dE00) and about 7000K or more.
Since there is no data from you, no further help could be provided.Thank you for the response!
There is a 3rd option between “native white” and D65 white for a laptop. It is choosing as whitepoint the closest daylight white to native white. That white would be white (no green, no pink) but may be cooler than D65. The contrast drop by using that white should be small.
Ha ha, sorry I’m a little confused. So many versions of white! In practical terms, would this mean choosing a whitepoint color temperature of, say 6800K on the calibration tab before calibrating?
Does my option 2 (Setting to 6500 so it will match Eizo), seem like a bad idea?
Thanks
First of all, verify this:
user error while measuring, for example setting a wrong spectral correction for their colorimeter, or not choosing one.
When you are sure that it is ok, then you can choose whatever whitepoint you want.
Same applies to your Eizo if you wish to verify how it behaves with DisplayCAL… you’ll need specific configuration for its backlight if you use a colorimeter. ColorEdges calibrated with Color Navigator may arise some surprises -CIE observer- when verified with Displaycal.
Since there is no data from you, no further help could be provided. Check all this stuff first
Well you are right! I definitely fall into the category of “not using one” for spectral correction.
The dropdown tab was a little confusing to me, so I had left it set to “Auto”
Can you help me understand how to make this selection?…
I see a drop down menu of about 15 choices. They appear to reference specific monitors (EG “Macbook Pro Retina 2016” or “Dell U2413”) None of the small handful of choices actually match with my monitors (Macbook or Eizo) as far as I can tell.
I suspect the right answer might be to select the technology that matches my display. How would I figure this out? I did some searching on my particular Eizo model. (CG223W) and couldn’t determine it exactly. (on the other hand, I’m guessing the only Phosphor LED IPS for the 2016 Macbook would do for my 2015 model?
Thanks!
We do not know which colorimeter you have. I assume that is an i1d3 because of the options.
CG223W in an older widegamut CCFL, older than PA271W. Proper spectral correction would be a mix between widegamut CCFL (PA241W/PA271W) for Blue and Green and stanradr sRGB CCFL for Red, and a synth white if you add the three.
Maybe there is a sample in colorimeter database(CCSS):
https://colorimetercorrections.displaycal.net/
If ther is not such community correction… use WGCFFL (which is not exactly the same)… or make your own although it is not noob friendly (Excel, or Numbers spreadsheet, take each channel as I said, then add them to get white, row order I think it’s WRGB ).
There were other similar Eizos and NECs to yours, maybe there are corrections for them, use displayCAL CCSS correction spectral plot to identify them: Red = Standard CCFL; green blue = WG CCFL like PA241W.For mac, if your mas is sRGB-like mac use White LED (common displays), if your mac has a P3 display, on the one labeled for mac P3 displays.
If you do not know, check 2D gamut boundaries in your mac profile information in DisplayCAL: sRGB=>White LED, P3=>the one labeles WLED PFS phosphor P3 Retina etc…Calibrite Display Pro HL on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.Thank you! It’s gonna take me some time to digest/re-profile the monitors. Hopefully this gets me going in the right direction. I’ll report back as soon as I’m able to.
Just getting moving on this, and I’m having trouble figuring out if my Macbook is P3 or not.
I checked profile information in Display Cal (the little “i” by the “settings” options at the top of the application). It gives a lot of information, but nothing that says “srgb” or “WLED” or indicates “2d gamut boundaries”.
(I was hoping an alternate solution might be able to search by model type on the web, but that has proved fruitless as well.)
Would you be able to help instruct a bit further on finding the 2d gamut boundaries for my monitor?
Thanks for helping this newbie iron things out!
Just getting moving on this, and I’m having trouble figuring out if my Macbook is P3 or not.
The model and year of the mac is visible in “About this mac”. Have you read the help text on the main tab in DisplayCAL? It should be very easy to figure out, no need to mess with profile info and gamut boundaries.
Thanks Florian, I had not seen the tab you are reffering too, but it’s much simpler now that I do!
Hi All,
Finally getting a chance to look at this more in depth. Your feedback was great, and I’m sure it’s a step up for me to now be using corrections and to know which ones to use with my monitor.
However, I’m still confused about the item that originally brought me here: Understanding whitepoint.
My Eizo now gives back a calibration report with all items being “ok”/green.
However, My Macbook Monitor still says that my measured vs assumed target whitepoint is “NOT OK”
What is further confusing is that looking at the actual data, it seems the variation is very nominal (6798K vs 6800K)… see screenshot
Am I safe to ignore this “not ok” warning, or is it a sign that I’m still doing something wrong?
However, My Macbook Monitor still says that my measured vs assumed target whitepoint is “NOT OK”
What is further confusing is that looking at the actual data, it seems the variation is very nominal (6798K vs 6800K)… see screenshot
Am I safe to ignore this “not ok” warning, or is it a sign that I’m still doing something wrong?
That variation si not small. If you exclude brightness White point is defined by 2 coordinates x & y from CIE xyY (“Y” is brightness). It’s a point in a 2D plane.
Color correlated Temperature is just a number. A PROJECTION in a curve length of a point in that plane.
it gives you information about where is the point on the curve placed closes to that point.
But it gives you no information about actual distance from white point to daylight cuve of color temperature.DisplayCAL is saying that your whitepoint is not close to daylight curve. It projects onto this curve at 6798K position, but it is not close to that curve in a 2D plane. It is “Pinkish” or “Greenish”, not “white”, not close to “daylight whites” curve.
When you measure color difference between your white point and 6800K daylight white it gives you ~8dE
Fast MS Paint diagram explaining it.
If you have a P3 Mac display (so spectral correction is OK), 8dE is big enough to do not ignore it.
Calibrate in DisplayCAL to D65 white or 6800-6700K CCT. Your display contrast is high enough, you are going to loose just a bit.
But if you evaluate its current white as “visually white”… you can ignore it.Aha, I do think I understand. Thanks for the detailed explanation (and drawing!).
If I am using only that monitor, I do think my eyes quickly adjust and I would call it “visually white”
However, I do see a blue shift between it and my calibrated Eizo (seems not to be a surprise, as the Eizo clocks in at : whitepoint: xy 0.3148 0.3321 (XYZ 94.77 100 106.32), CCT 6372K
I’ll give a shot to d65 for the Macbook and report back.
Again, thanks for the nuanced feedback!
However, My Macbook Monitor still says that my measured vs assumed target whitepoint is “NOT OK”
What is further confusing is that looking at the actual data, it seems the variation is very nominal (6798K vs 6800K)… see screenshot
Am I safe to ignore this “not ok” warning, or is it a sign that I’m still doing something wrong?
That variation si not small. If you exclude brightness White point is defined by 2 coordinates x & y from CIE xyY (“Y” is brightness). It’s a point in a 2D plane.
Color correlated Temperature is just a number. A PROJECTION in a curve length of a point in that plane.
it gives you information about where is the point on the curve placed closes to that point.
But it gives you no information about actual distance from white point to daylight cuve of color temperature.DisplayCAL is saying that your whitepoint is not close to daylight curve. It projects onto this curve at 6798K position, but it is not close to that curve in a 2D plane. It is “Pinkish” or “Greenish”, not “white”, not close to “daylight whites” curve.
When you measure color difference between your white point and 6800K daylight white it gives you ~8dEFast MS Paint diagram explaining it.
If you have a P3 Mac display (so spectral correction is OK), 8dE is big enough to do not ignore it.
Calibrate in DisplayCAL to D65 white or 6800-6700K CCT. Your display contrast is high enough, you are going to loose just a bit.
But if you evaluate its current white as “visually white”… you can ignore it.How can we verify that the display is on or very close to the daylight white curve when calibration is complete ?
You can see it in profile information, the text info on the right. Example:
Media white point
Illuminant-relative XYZ 95.06 100.00 109.80 (xy 0.3118 0.3280)
Illuminant-relative CCT 6560K
ΔE 2000 to daylight locus 0.17 <<<<<<<<<<<<< daylight curve
ΔE 2000 to blackbody locus 4.50 <<<<<<<<<<<< blackbody curveAlso you can measure it by Tools, report, calibrated display report, look for “CDT” (correlated daylight temperature) on log window.
