Spyder 5: different results with different LG monitors?

[repsak]

I recently bought a DataColor Spyder 5 and calibrated my two LG monitors directly with DisplayCAL: left monitor LG 27MU67, right monitor LG W2220P. I used the same options for both monitors: mode LCD (white LED), color temperature 6500 K, Gamma 2.2, Testchart 1134, the remaining options remained unchanged, in the second run I checked white & bleck level drift compensation. As you can see in the picture attached, the left monitor (27MU67) has a white color which goes a little bit into red  (I indeed was asked to add red at the beginning of the calibration in order to have all  RGB values in the middle & green) while the right monitor (W2220P) has a more white white in comparison (I was asked to add blue). As a novice to color calibration who is just enthusiastic about image and video editing I would have expected very similar results on both monitors, not such a clearly visible deviation. Is this a problem with Spyder 5  (which I could exchange for a´X-Rite ColorMunki Display but for 60 Euro ~ 65 USD more) or am I doing something wrong with the calibration? Perhaps it is also worth mentioning that the left monitor is quite new and with a 4K display while the right monitor has around 6 years now.  Many thanks!

Attachments:

You must be logged in to view attached files.

Calibrite Display SL on Amazon  
Disclosure: As an Amazon Associate I earn from qualifying purchases.

[Florian Höch]

Colorimeters like the Spyder5 use filters to be able to measure color. If the filter characteristics do not closely follow the CIE 1931 standard observer color matching functions (which are modeled after average human perception), as is the case with most of the cheaper colorimeters, then this can lead to the observed differences when measuring different displays, especially when different backlight technologies are involved (the W2220P has a CCFL backlight, the 27MU67 a white LED one). There’s several ways around this:

• Use generic colorimeter corrections that match the backlight technology for each display. The Spyder5 provides these as “LCD (White LED)” and “LCD (CCFL)” measurement modes.
• Use a calorimeter correction specifically made for your display(s) and Spyder5. This requires a spectrometer.
• Match the whitepoints visually (do not set a whitepoint target, match by eye during interactive display adjustment, ignoring the bars and dE readout)

The first two options can be combined with the last one.

[repsak]

Thanks, that helps a lot, I was not aware my W2220P has CCFL, I will try this option tomorrow first.

[p.dada]

You might want to leave the RGB sliders in the monitor menu alone if we’re talking about cheaper monitors with 8-bit internal processing, as it can introduce serious banding. Further, raising any of the channels beyond 50% can introduce clipping of the lighter tones. Based on my experience with consumer LG displays, I’d leave it at its factory defaults and choose User mode if f-engine options are available. Then I would disable the interactive white point adjustment and just leave the whole calibration process to my graphics card. If you have a recent Radeon graphics card, its internal LUT bit depth is 10 bits and you will get much smoother gradients. In any case, choose the right mode for your display as Florian already said, but don’t expect a dead-on match, especially with different backlights.

[repsak]

I changed to CCFL and did not use the interactive white point adjustment (RGB sliders are default); well, I still see a difference between the monitors, perhaps a little bit less than before but this is hard to judge. After all, however, the old CCFL monitor only covers less than 70% of sRGB according to my calibration results (while the new white LED monitor covers 98%) and I will use the old monitor only as additional space for control panels, not for viewing images. Thus, my expectations were that my new monitor is roughly “accurate” and that it leads to reproducible images with the same color calibration settings, but this is hard to test for me just having those two apparently very different monitors. However, I calibrated my new monitor more than once with slightly different DisplayCAL settings and the white always goes a little bit into red, not much or barely noticeable when viewed in isolation, but more visible when compared to the older monitor.

Thanks again for your help!

• This reply was modified 8 years ago by repsak.
• This reply was modified 8 years ago by repsak.

[Author]

Posts