- This topic has 3 replies, 2 voices, and was last updated 2 years, 11 months ago by
.
-
Posts
-
Some time ago I bought Samsung Q80T. According to for example Rtings I set picture settings to “Movie” with color tone set to “Warm 2” as it should give the best reference results in terms of white balance and generally colors – according to RTings warm2 gives very good 6500K white balance. I knew, that of course it will look a bit more yellowish than standard settings, but I had still a feeling that it is even too warm. So I ran some measurments through X-Rite i1Display Studio and DisplayCal and… my doubts seem to be right. I attach two screenshots which show very strange behaviour of white balance and RGB measurments in function of luminance. It is strongly varying and we can see, that reds are pushed strongly in highlights and blues are pushed in darker areas.
Of course such behaviour makes it much harder to calibrate and I cannot just make simple shifts of colors. As this is TV I also can’t just create ICC profile, I have to use only TV menu adjustments. There is 20-point colors correction, but with those differences I have to sometimes use maximum shift available. This does not look ok:)
So I’d like to consult is it normal behaviour that white balance varies so much between darker and brighter areas? Or maybe I should give up with calibration and just contact Samsung support.
Thanks in advance for any hints:)
Just for the record:
Measurments were made with local dimming set to low, all picture “boosters” (dynamic contrast, picture clarity, shadow detail etc.) were off. Gamma was set to 2.2, but on BT.1886 results are very similar and reds are also boosted.
Calibrite Display SL on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.Check if it has 10/20 point grayscale color correction on OSD menus (check manual). All theses big TV should have that. Then use DisplayCAL’s cousin HCFR to fix greyscale color.
If greyscale corrections exceed maximum value change white point setting and try to aim to “closest daylight” whitepoint, cooler but white (no green pink tint). Then repeat grayscale correction.
Indeed, I have 20-point grayscale color correction, however it’s hardly enough to get close to correct white balance. And that is what worries me: if rtings is right, I shoudn’t have to struggle so much with it:)
And additional problem: with so extreme corrections on 20-point color correction also other aspects start to go a bit crazy, for example gamma, so it’s quite hard to get everything in place:)
