Unsolved
27 Posts
0
5620
U3014, Spyder 4 Elite, DisplayCAL, calibration issues
I have just found this thread and I would like to thank you for your input.
I have been battling with my Dell monitors for a while now. First with U3011 (which is dead now), and then with U3014. I have never been fully happy with calibration on these monitors. I have Spyder 4 Elite.
I see green tint to my grays on U3014 using the Spyder 4 software. To battle this, I have recently installed DisplayCAL, but the results are very poor. The white point seem to be wrong and the whole display looks very yellow. I have tried to calibrate it using multiple settings in DisplayCAL software but the results are always the same. Here is what the Verification tool in DisplayCAL tells me after the monitor was calibrated using DisplayCAL, Spyder 4 Elite and Nvidia 1080 GPU.
I have read the other thread and noticed that you recommend the X-Rite i1Display Pro. I will be trying the workaround you mentioned to use corrections, but you mention that AMD card is also required. I have Nvidia 1080 GPU.
I would like to ask whether anything changed since your last comment there? Do I really need to change my colorimeter and my GPU to get accurate results? Could you please also explain also, why AMD card is better for this?
Additional question, if I was meant to buy a different Dell (or other) monitor now (I am a photo retoucher), which one would you recommend (30" or bigger)?
Help!
yumichan
721 Posts
0
September 19th, 2018 02:00
ArgyllCMS >1.9, GeForce 10 series, >8bpc configured in Nvidia controls panel and a monitor with 10-bit input "should" be able load band less calibration in GPU.
Spyder family has not changed, the hardware is not accurate and their instrument agreement is very poor.
If you want accuracy get an i1d3. If you are going to use only DisplayCAL, then a Munki display will work (it's a slow i1d3 with different unlock code) but since your U3014 has hardware calibration features, but an i1Display Pro and use it with Dell hardware calibration tool (DUCCS, I would get the latest one 1.6.5).
DUCCS may fail to get the best from your monitor. If that happens you could run a DisplayCAL graphics card calibration on top of your "good/average, but not perfect" DUCCS' calibration in CAL1/CAL2 OSD modes. Since you cannot change RGB gains in CAL1/CAL2 when prompted to "equal" RGB bars just click in continue, they will be corrected in GPU calibration.
PS. Anyway... try a GB-LED spectral correction with your Spyder4 and DisplayCAL while in Custom Color mode, you have nothing to loose. You can also try bundled RGBLED correction for Spyder (top right corner in DisplayCAL once you have imported Spyder software corrections) instead of generic GB-LED CCSS corrections means for i1d3 family.
RebelFL
27 Posts
0
September 19th, 2018 03:00
Thank you. I will try your workaround suggestions tonight and will report back. If that won't help I will have to fork out for i1 Display Pro it seems. In the meantime, is Dell UP3216Q calibrating well? What should I replace my U3014 with? I don't think I can stratech to 30"+ Eizo.
RebelFL
27 Posts
0
September 19th, 2018 08:00
@yumichan
I have calibrated using the built in Wide Gamut LCD (RGB LED) mode, but the results are exactly the same, ie yellow.
In the other thread you have mentioned RG Phosphor file, which can be used as a correction. I have been looking for it and can't find it anywhere to download from. Do you know where I can find it please?
I don't have a RG Phosphor option in my dropdown box. Also I don't have any correction files in my correction dropdown box (only "None" and "Auto (none)").
I have looked online for RG Phosphor file, but other than a few people mentioning it, I don't see where I can download it from to use as a correction.
BTW I have also top of the line Dell Inspiron 7570 2018 laptop with the 4k full coverage Adobe RGB touchscreen straight from the XPS model, which I believe is also wide gamut, calibrated with Spyder 4 Elite and the results are very good.
yumichan
721 Posts
0
September 19th, 2018 09:00
You can get them from Xrite setup executables, then imported to DisplayCAL/ArgyllCMS. DisplayCAL is GUI, Argyllcms (oeminst) is command line. Maybe it is possible to get them from DisplayCAL in an automated way, but it may be instrument dependent.
Get setup exe for Xrite i1Profiler, and use DisplayCAL/argyllcms to get CCSS files from it.
"BTW I have also top of the line Dell Inspiron 7570 2018 laptop with the 4k full coverage Adobe RGB touchscreen straight from the XPS model, which I believe is also wide gamut, calibrated with Spyder 4 Elite and the results are very good."
And that means... nothing.
That model is very likely to use new widegamut backlight type and by chance your very inacurate spyder's filters seem to be better positioned (always from your subjective POV, your statement lack of a 3rd party "reference" instrument to support it) to capture that backlight as an standard observer would do... at least untill they fade. That backlight dependency/variability is what i1d3 and spectral corrections solve to a very high degree.
Also a premium laptop (small screen) is very likely to have better uniformity than a low cost widegamut like U3014 or whatever widegamut model from del, asus, benq, viewsonic, lg... partial screen tinting may affect your subjective evaluation of your screen whitepoint.
RebelFL
27 Posts
0
September 19th, 2018 09:00
@yumichan
Thank you. I found the RG Phosphor. It appears I have unchecked the iProfiler option during DisplayCAL installation and it never installed corrections. I am calibrating now using Phosphor correction in Custom Color profile, as advised. It will take around 30 min.
I wasn't trying to be a smarta** by mentioning the laptop. I was just curious why the results are so different.
Also tge reason why I asked about a different monitor is that if I am to spent £200 on new colourmeter, maybe I should think about a better monitor instead. I've had the u3014 for a good while, and maybe it is time to think about its replacement.
Thank you again for your input.
yumichan
721 Posts
0
September 19th, 2018 10:00
DisplayCAL's whitepoint with the same spectral correction that Spyder software uses is the same as Spyder software whitepoint.
If you do not get the same, then you are doing it wrong.
An easy approach is to use Spyder softwre, get whetever whitepoint you want.
Then use DisplayCAL to measure xy coordinates of that WP with some suitable correction (for example the same bundled correction that you used in spyder software).
Further DisplayCAL calibrations should use the same incorrect xy coordinates as white target.
But as I said above if you don't get the same white... you are doing it wrong. It's user's fault... maybe related with the correction you choose in Spyder softare not being equal to the one you use in DisplayCAL
yumichan
721 Posts
0
September 19th, 2018 10:00
You'll need an i1DisplayPro for the new monitor if it is a widegamut (and it's very likely that for a sRGB one too), so IMHO buying an i1d3 is the first step in you way.
Since HW calibration features are very common nowadays so if your are going to buy a widegamut don't buy the munki.
RebelFL
27 Posts
0
September 19th, 2018 10:00
@yumichan
Just finished calibration and the results are identical. Still very strong yellow tint to the whole picture. As it is, I cannot work on the monitor. Please see the image bellow to see the difference between primary monitor (u3014) and secondary monitor (generic Iiyama monitor, only used to display Photoshop tools). Both calibrated with the same Spyder 4 Elite and DisplayCAL.
I am going to have to uninstall DisplayCAL and install Spyder software back. Using Spyder software the white point is OK, but I'm getting green tint in greys, especially visible on B&W image. Not ideal, but at least I don't have a very strong colouration to the whole image, like I do now. As it is, I cannot work on the monitor, so the calibration needs to go.
I really need to consider my options. Thank you for your help.
RebelFL
27 Posts
0
September 19th, 2018 11:00
I am also angry at myself for going with Spyder all these years ago. I am always very careful and always do research before purchasing something like that. And I know I have done the research when I was dropping £200 on the Spyder 4 Elite... but somehow I have missed that the Xrite is so much better.
RebelFL
27 Posts
0
September 19th, 2018 11:00
Using DisplayCAL profile loader I have loaded my ICC profile created by Spyder for this monitor. Side by side with the same secondary monitor, there is a big difference between the calibration. While Spyder's software still created a warmer overall look than my secondary monitor, it is significantly less yellow than what DisplayCAL created.
I have spent the whole evening yesterday picking settings and calibrating using DisplayCAL. I am honestly out of ideas.
What I have tried so far in Display CAL:
- default calibration settings (as measured) with LCD (generic) mode
- default calibration settings (as measured) with Wide Gamut LCD (RGB LED) mode
- 110 cdm2, 5500K, Gamma 2.2, with Wide Gamut LCD (RGB LED) mode
- 110 cdm2, 5500K, Gamma 2.2 with LCD (generic) with spectral correction: LCD RG Phosphor - I cannot pick Wide Gamut LCD with correction selected.
Every single time I get exactly the same strong yellow tint. There is no variation to the picture at all. If I load the profiles from each of these calibrations, they all look identical.
I am out of ideas really. If you see a mistake somewhere, please do point it out. This is all very frustrating and I feel like banging my head on the wall.
yumichan
721 Posts
0
September 20th, 2018 02:00
"Every single time I get exactly the same strong yellow tint. There is no variation to the picture at all. If I load the profiles from each of these calibrations, they all look identical.
I am out of ideas really. If you see a mistake somewhere, please do point it out. This is all very frustrating and I feel like banging my head on the wall."
There is a visual whitepoint editor in DisplayCAL, use it. It is close to color temperature selector.
You'll set whitepoint "by eye match" to other source (your other monitor).
Then DisplayCAL will calibrate greys to match (in color tone) that white you choose visually, an to desired gamma set as targte (AFAIK "Y" value from CIE XYZ readings, I mean cd/m2 reading, will be taken as "true"/"accurate" for gamma)
RebelFL
27 Posts
0
September 20th, 2018 10:00
I have reverted to the ICC profile created by Spyder. It's not ideal, but better. There is no point messing about with it if I cannot get an accurate calibration with the instrument anyway. This one is OK, not too bad, so I will just keep working on it for the time being. I have already started eyeing up i1display pro, so I don't think it will be long until I actually buy it. Thank you
Thinh Tran
10 Posts
0
September 21st, 2018 19:00
After calibration, you must add the profile made by DisplayCAL to the Color Management of Windows (It stays in Control Panel window) and restart your computer. I got the same situiation. When i loaded the profile calibrated by DisplayCAL, my UP2516D became over saturated : very yellow, darker ... At first, i thought the reason was the changing UP2516D's native white point (7100k) to the new one (6500k). It was a misunderstand. The true reason was Windows still used the default of sRGB profile in spite of DisplayCAL profile Loader had run already. After i added the calibrated color profile to Color Management, everything became OK.
RebelFL
27 Posts
0
September 28th, 2018 23:00
Thanks. I just ordered iDisply Pro so hopefully that will be the end of my problems.
RebelFL
27 Posts
0
October 5th, 2018 07:00
So my i1Display Pro arrived and the results are identical to Spyder 4 Elite with DisplayCAL. Looks extremely yellow What I have tried so far: normal i1 Display Pro calibration in DisplayCAL; RG Phosphor correction with i1 Display Pro in DisplayCAL both cases produced exactly same results. What I still want to try: calibration with i1 Display Pro software (which I have not even installed yet); hardware calibration using DUCCS CD cover for white balance. Check out the colour under the calibrator. It should be white