U3014, calibration issues

Question

Hi,

This question is for yumichan. Hopefully you can help.

I have a U3014 that seems to have a poor contrast ratio and a high native white point. I did a factory reset and ran "Report on uncalibrated display device" in dispcalGUI on the Standard OSD preset, and these were the results:

23:13:40,190 Black level = 0.4067 cd/m^2
23:13:40,192 50%   level = 61.42 cd/m^2
23:13:40,193 White level = 272.79 cd/m^2
23:13:40,195 Aprox. gamma = 2.15
23:13:40,196 Contrast ratio = 671:1
23:13:40,197 White chromaticity coordinates 0.3004, 0.3258
23:13:40,198 White    Correlated Color Temperature = 7245K, DE 2K to locus = 9.9
23:13:40,200 White Correlated Daylight Temperature = 7241K, DE 2K to locus = 6.6
23:13:40,201 White        Visual Color Temperature = 6758K, DE 2K to locus = 9.5
23:13:40,203 White     Visual Daylight Temperature = 6939K, DE 2K

Then, I used DCCS 1.5.3 and calibrated Cal 1 to sRGB and Cal 2 to AdobeRGB. I returned to dispcalGUI to verify the results. The average and maximum ΔE*00 numbers were fine for both calibrations, but the "Measured vs. assumed target whitepoint ΔE*00" was 7.11 and 6.7 respectively. There is also a weird reading of the "Correlated Color Temperature" on the AdobeRGB profile at 5%.

Here are links to the full reports:

https://dl.dropboxusercontent.com/sh/f4dbqzn9bhfygfn/AAD-xf9yh_KGpWIGnTgZpSn0a/U3014_sRGB.html
https://dl.dropboxusercontent.com/sh/f4dbqzn9bhfygfn/AACMWOfI8Qx-GB42qkSTqGK5a/U3014_AdobeRGB.html

Am I doing something wrong, or should I get a replacement?

Also, you recommended setting a gamma of 2.0 when creating a custom profile back in this thread:
http://en.community.dell.com/support-forums/peripherals/f/3529/t/19502841

Is there a particular reason why you suggest using 2.0 instead 2.2?

Thank you in advance for your time!

yumichan · Accepted Answer

I meant 2.2, thanks for finding the typo. Now it's fixed. Such high dE in assumed WP means whitepoint far from daylight curve (and balckbody curve too). Make sure you do not have applied GPU driven color enhancements like 'Digital Vibrance' and such. If no GPU driver color enhancements are applied, then it's DCCS fault (one of its many faults), device is measuring it properly since it is stored properly on ICM file (profile WP vs measured WP, check 'show additional statistics') Unless new DCCS comes out for 2016 2k & 4k models, I would not expect a solution or a update from Xrite or Dell. Your best choice right now is to have an AMD GPU (Radeon or Firepro) or a nvidia Quadro. After DCCS calibration (gamut emulation + gamma) is done, just recalibrate CAL1 and CAL2 in GPU with ArgyllCMS to D65. Your grey is neutral, AMD or Quadro GPUs won't create banding and 6dE greenish-cyan white won't make you loose more that 5% greys values due to GPU calibrationContrast issue cannot be fixed if it has low contrast even in native mode (Custom color mode OSD, RGB gain 100% x 3 channels, offset 50% x 3 channels)

tenderchkn · Answer

Thank you for your reply. I have a Nvidia GeForce GPU, so software profiling is not an option for me at the moment. I just checked my GPU options, and everything is at their defaults. Digital vibrance is set to 50%. If I pull it down to 0%, the image is grayscale, so I think 50% is the neutral setting.

I did another "Report on uncalibrated display device" in Custom Color mode and the contrast ratio was 775:1. Since this display should be capable of at least 9xx:1, I think I will go for a replacement.

Should I also expect a better native white point from a replacement as well? What is the acceptable range that it should show in dispcalGUI?

Also, can you explain the difference between "Measured vs. assumed target whitepoint" and "Measured vs. profile whitepoint"? In the AdobeRGB report, it reads:

Profile whitepoint XYZ (normalized): 111.53 121.63 126.88 (91.7 100 104.32), CCT = 6591K
Measured whitepoint XYZ (normalized): 110.15 120.71 125.35 (91.26 100 103.85), CCT = 6610K
Assumed target whitepoint (XYZ): 6600K daylight (94.99 100 110.26)

The Kelvin values between all three are similar, but I see that the XYZ values on the assumed target is different from the other two. Does that mean that the calibrated color temperature (blue-yellow) is correct, but that there is a shift towards green? Is this a hardware issue that varies from unit to unit, or is it entirely a software issue caused by DCCS, like you said? If that is the case, then I guess I will have to get an AMD card and recalibrate in GPU.

FInally, if I were to recalibrate with DCCS in the future, do I need to perform another factory reset, and unassociate ICC profiles first? Or do I just set the OSD preset mode to either CAL1 or CAL2 and let it start from there? Thank you again for your help!

yumichan · Answer

After calibration, program measures CALx calibrated behaviour and stores that info on a profile.

Profile white point (DCCS info from i1d3 + RG-phosphor correction) and measured white in Argyll (id3+RG_phosohor) are almost the same. DCCS IS ACTUALLY READEING PROPERLY, but fails computung a LUT3D that gets you close D65.. or may be it's caused by after calibration brightness setup.

Yes, shift towards cyan-green in white. It's DCCS issue since you can calibrate to exactly D65 in GPU, but with banding in greys sue to your nvidia.

DCCS 1.5.x resets CALX slot and only that slot, so no factory reset is needed. Those issues were in DCCS 1.0.x that "first buyers" suffered from the begining (like me).

tenderchkn · Answer

Thank you for your input. I will ask Dell for a replacement, and I hope that it will produce better measurements. I know what contrast numbers to look for, but can you tell me what to look for in the native white point?

More specifically:

1. Should the test be done in Custom Color Mode with 100% gain and 50% offset?

2. dispcalGUI lists four white point measurements. Which ones are the important ones?

3. What would be considered a "good" Kelvin and dE reading?

Also, out of curiosity, I ran a verification on the factory calibrated sRGB preset, and the dE between profile and assumed target whitepoint was 8. Do they use DCCS at the factory to calibrate as well? Or is there a chance that there is something wrong with my i1d3?

yumichan · Answer

1- Yes, but standard mode is ussually the same setting

2- Daylight correlated color temp "CDT", usually.

3- dE<2, usually 6500K for general use.

Factory AdobeRGB and sRGB aims for a mean error @ brightness 50% (>200cd/m2). White and is neutral greys could be wrong...

Your i1d3 seems to be OK. white with cyan -green cast should be spotted by sight. Calibrate in GPU to D65 using custom color mode and RGB gain and you'll se what white should be. You have to use RG_phosphor correction, it's important

Monitors

U3014, calibration issues

Was this post helpful?