- This topic has 3 replies, 2 voices, and was last updated 3 years, 9 months ago by
.
-
Posts
-
I recently got a Dell XPS 13 7390 laptop that has a Sharp 4V0HY-LQ133M1 IPS panel (LED backlight). The panel is pretty great in terms of uniformity and absence of backlight bleeding but it has a horrible general yellow tint to it (it’s most likely pretty minor but I am very sensitive when it comes to yellow tinted screens). It is advertised as 100% sRGB capable and colorimeter readings seem to confirm that, as I always get close to 100%. As colorimeter I grabbed a i1Display Studio.
I also own an EIZO monitor, factory-calibrated to sRGB and boy do I love the colors on this thing, they are spot on! White on the EIZO is what I consider a pure white and colors are well distributed. So I have good reference point. After calibrating the EIZO, almost nothing visibly changes (except for minor contrast shifts in certain areas), which further confirms the EIZO’s good default calibration.
After the first calibration of the XPS (6500K and Gamma 2.2 like the EIZO) the result was pretty poor in terms of still being too warm while additionally having a red tint to it.
I then began using the visual whitepoint editor and tried eye-balling the whitepoint to match the EIZO. After about 10 calibrations – successively adjusting the R and G steps in the whitepoint editor – I am getting pretty close in terms of general color distribution in comparison to the EIZO.
However, after each calibration, the overall whitepoint is ever so slightly back into the yellowish range, producing whites, greys and blacks that have a faint yellow/brown touch still to them, making them look more dirty. I can’t get the whites (esp. greys between black and white) to be as “pure” as on the EIZO.Did I expect too much from calibration, i.e. is that screen ‘beyond repair’ and I won’t ever get rid of that warm/yellow touch of the panel entirely?
I mean this as a general question aside from all the details about correction profiles, observers, tone curve etc. – is it even physically possible to entirely shift a tinted screen’s overly yellowish whitepoint to fully match another’s cold/pure one via videoLUT if both provide full sRGB gamut?
Might poorly QC’ed backlight LEDs produce a yellowish light that can’t be sufficiently corrected through the LC layer and its calibration?Calibrite Display SL on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.I recently got a Dell XPS 13 7390 laptop that has a Sharp 4V0HY-LQ133M1 IPS panel (LED backlight).
After the first calibration of the XPS (6500K and Gamma 2.2 like the EIZO) the result was pretty poor in terms of still being too warm while additionally having a red tint to it.
U need to know laptop’s backlight and find a suitable correction for it. Otherwise your i1d3 variant will not measure properly. With backlight I mean spectral power distribution, saying it’s LED actually means nothing.
However, after each calibration, the overall whitepoint is ever so slightly back into the yellowish range, producing whites, greys and blacks that have a faint yellow/brown touch still to them, making them look more dirty. I can’t get the whites (esp. greys between black and white) to be as “pure” as on the EIZO.
In laptop or monitor without HW cal all greys are corrected usieng GPU LUTs. Some are better than the pthers in this task. Laptop’s iGPU LUT are known (at least in the past) for low bitdepth and lack of dithering, hence banding & posterization after calibration in a grey gradient is unavoidable.
You’ll have to live with it. Maybe just profiling without grey calibration and using an editor with high bitdepth + dithing processing when rendering to screen may improve things (AFAIK just LR in develop module, ACR and CaptureOne)Did I expect too much from calibration, i.e. is that screen ‘beyond repair’ and I won’t ever get rid of that warm/yellow touch of the panel entirely?
It’s GPU fault.
Thank you so much for your quick reply Vincent! It’s most appreciated!
So if I get this right, as long as a panel is not near-perfect from factory (which from what I’ve read is almost never the case for any recent consumer-grade IPS panel that is not extensively factory-calibrated), because of inherent iGPU limitations it is not possible to completely correct all of its color distortions via calibration due to the LUT-capabilities of the iGPU not being fine-grained/powerful enough.
For instance, the XPS colors get pretty close to the EIZO now but blues are still not as deep and tend to shift more towards cyan, depending on the tone. If I force the whitepoint more into blue, other colors begin to suffer. Guess I am at the limit of what is possible with this hardware.
I pondered whether I should return the XPS and participate in the hardware lottery again. Since the screen is otherwise pretty great (contrast, uniformity, no bleeding) I was hesitant to do so and instead opted for calibration. Judging from what you wrote, even with a replacement device I might likely be struggling again to get rid of whatever fancy tint the next one might have (e.g. red, blue or again yellow) and even get additional bleeding or uniformity issues into the mix. Seems like it is not worth the gamble and I should learn to live with this one instead.
Thank you so much for your quick reply Vincent! It’s most appreciated!
So if I get this right, as long as a panel is not near-perfect from factory (which from what I’ve read is almost never the case for any recent consumer-grade IPS panel that is not extensively factory-calibrated), because of inherent iGPU limitations it is not possible to completely correct all of its color distortions via calibration due to the LUT-capabilities of the iGPU not being fine-grained/powerful enough.
No just grey ramp, it’s the only thing you can correct with GPU calibration “system wide”. This is where such errors arise (bandig / grey ramp coloration)
For instance, the XPS colors get pretty close to the EIZO now but blues are still not as deep and tend to shift more towards cyan, depending on the tone. If I force the whitepoint more into blue, other colors begin to suffer. Guess I am at the limit of what is possible with this hardware.
That is profile related (color management), not related to iGPU although some rounding errors can happen due to limited precision.
I’ll point to my 1st quote: use an accurate correction for your i1d3 colorimeter to measure both Eizo and XPS (one for each display backlight tech).
Eizo’s factory grey ramp can be great (it’s very likely that you can use it’s non color managed grey ramp as “neutral reference”), but the profile you made for it may be not accurate (You cannot use EIzo as overall reference in color managed apps unless the profile that describes display is a match)I pondered whether I should return the XPS and participate in the hardware lottery again. Since the screen is otherwise pretty great (contrast, uniformity, no bleeding) I was hesitant to do so and instead opted for calibration.
Then it is good, just profile it properly. White point matching/grey issues are difficult to avoid in a laptop.
