- This topic has 11 replies, 2 voices, and was last updated 6 years, 12 months ago by
.
-
Posts
-
Hello there,
My setup consists in a laptop and an external monitor. By default they’re very different and the controls are limited – Can’t set a color temperature, so matching both to a desired white level by hand seems impossible.
So what I’ve done is leaving both on factory defaults and calibrating them both to 6500k, 0.5 black level, 256 white level (the minimum and maximum between them both) and doing the very very long 6+ hours calibration.
They got somewhat close to each other, but there’s still a slight noticeable difference. Running the report, it says that my laptop screen is at 6459K and my secondary screen is at 6583K after calibration. Is that the reason for the difference? And is there anything I can do about it?
Another unrelated thing – I set those calibrations to be the default but the tray app doesn’t load them at startup even though it says it does. So I always need to right click, reset, then load again and then it works. Is there a way to make it load the calibration flawlessly at startup?
Thanks a lot!
Hi,
So what I’ve done is leaving both on factory defaults and calibrating them both to 6500k, 0.5 black level, 256 white level (the minimum and maximum between them both)
Ok, that seems reasonable.
and doing the very very long 6+ hours calibration.
Note that you’re probably not gaining much by doing this – “medium” calibration speed is usually the highest setting I would recommend.
They got somewhat close to each other, but there’s still a slight noticeable difference.
For the best visual match, you can use the visual whitepoint editor.
Another unrelated thing – I set those calibrations to be the default but the tray app doesn’t load them at startup even though it says it does.
Is “preserve calibration state” enabled? If it is, attach logs please (see the documentation for logfile locations).
Hi Florian,
Thanks for the quick response! I’ve attached the logs.
About the visual whitepoint editor, I tried it a couple of times but couldn’t really make sense out of it. Should I discard my current calibration and start it all over?
Thanks for the quick response! I’ve attached the logs.
You had “preserve calibration state” disabled.
About the visual whitepoint editor, I tried it a couple of times but couldn’t really make sense out of it.
The process is as follows, you pick one display as reference (which you don’t adjust), then you adjust the other’s white point until white matches visually. Then you measure the adjusted white, which will set it as calibration target, and start calibration/profiling on the display where you adjusted the white.
Hi Florian,
This is how it looks on my tray:
Preserve calibration state seems to always be checked but I still need to force it to load by hand.
So if I understand correctly – I should use the visual whitepoint editor to try to match one screen’s white to the other one, measure that and use that color temperature as the input? Will this be more precise than just setting both at 6500k?
The fact that I set 6500k as a target but neither of my screens really matched it, what might be causing that? A limitation of the screens themselves?
Thanks a lot!
Preserve calibration state seems to always be checked but I still need to force it to load by hand.
That’s not what the logs show.
So if I understand correctly – I should use the visual whitepoint editor to try to match one screen’s white to the other one, measure that and use that color temperature as the input? Will this be more precise than just setting both at 6500k?
Yes.
The fact that I set 6500k as a target but neither of my screens really matched it, what might be causing that? A limitation of the screens themselves?
Never look at the correlated color temperature when determining wether or not the target white has been matched – the only reliable measure is deltaE. A few dozen Kelvin difference from target has little meaning and can already be caused by normal fluctuations in instrument and display repeatability.
That’s not what the logs show.
So I guess there’s some sort of disparity between the tray utility and what the software is actually doing? Maybe a bug?
Never look at the correlated color temperature when determining wether or not the target white has been matched – the only reliable measure is deltaE. A few dozen Kelvin difference from target has little meaning and can already be caused by normal fluctuations in instrument and display repeatability.
Alright! But I don’t know how to make sense of deltaE, I guess I lack the theoretical knowledge to be able to properly understand this. Is there any material you could recommend me to study so I can understand this better? I’d really love to.
Thanks a lot!
So I guess there’s some sort of disparity between the tray utility and what the software is actually doing? Maybe a bug?
I suspect a graphics driver bug, although in terms of video card gamma table, this would be unusual for nVidia.
Alright! But I don’t know how to make sense of deltaE, I guess I lack the theoretical knowledge to be able to properly understand this
It is said that 1 delta E is the perceptability threshold, i.e. where you begin to see a slight color difference. Less than 1 delta E is in most cases a near imperceptible color difference. Above 1 gets increasingly more visible. Up to 3 delta E is usually an acceptable result for color (neutrals should not differ by much more than 1 delta E, because we are more susceptible to notice differences in achromatic colors). All of this based on the most refined delta E formula to date, which is delta E 2000.
So I’ve ran the large verification testchart, something I haven’t done before. And saw those results attached (left is external monitor, right is laptop screen).
Pretty garbage huh? I calibrated those with the same settings. Does this mean my laptop screen is incapable of displaying precise colors?
Also another thing that catches my eye is the ‘assumed target whitepoint’. Why is it so different if I tried to calibrate both to 6500k?
I’ve tried recalibrating both, but using sRGB instead of Gamma 2.2, 6500k and got the results attached.
Slightly closer whitepoints, got much more precision on the laptop screen… Though I still notice a orange cast on the laptop, green cast on the external monitor.
I’m trying to use the visual whitepoint editor right now but I’m finding it super confusing. Tried measuring both screens and got 6320k on the laptop one, 6750k on the external one after the calibration, which differs a lot from the image i’ve attached (respectively 6530k and 6576k). What’s going on?
Also, which one is more appropriate, sRGB or Gamma 2.2? I’m using this computer for game development.
Thanks!
I’m trying to use the visual whitepoint editor right now but I’m finding it super confusing.
I may be able to provide better help when you let me know what you’re struggling with.
Tried measuring both screens and got 6320k on the laptop one, 6750k on the external one after the calibration, which differs a lot from the image i’ve attached (respectively 6530k and 6576k). What’s going on?
How have you measured? The measurement during interactive adjustment disables calibration while adjusting, so that’s to be expected. The reports look fine from the numbers.
Also, which one is more appropriate, sRGB or Gamma 2.2? I’m using this computer for game development.
Use the all-around default, gamma 2.2.
