Home › Forums › General Discussion › Differences between calibration with Lg True Color Pro and Datacolor
- This topic has 42 replies, 2 voices, and was last updated 1 year, 11 months ago by Vincent.
-
AuthorPosts
-
2022-04-16 at 20:27 #35121
AnonymousInactive- Offline
Report of displaycal
2022-04-16 at 21:27 #35123Can you explain to me in a very simple way what DWMLU is and how to use it?
Reply To: Differences between calibration with Lg True Color Pro and Datacolor
2022-04-16 at 21:36 #35124I ran another calibration with lg true colour pro and a check with displaycal and the values are much better, is it due to the fact that during the calibration process with lg true colour pro I chose 3×3 matrix instead of lut table?
a matrix profile assumes that display is well behaved and can be described by primaries coordinates & grey. If it behaves betetr now it would mean that LG software cannot measure it properly or build a suitable mesh from measurements (it cannot build a suitable “table” type profile).
Grey is still very bad behaved => you’ll need GPU calibration from displaycal or DWMLUT since LG HW calibration cannot correct it
Shifts in a* (pink-green) axis and even bigger shifts in yellow-blue (butthey may be less disturbing)(LG HW in monitor “could” store a proper calibration, likely to be 256-1024 entry per channel LUT… just happens that LG does not take enough measurements to capture all the issues in that low cost LG panel => if it canot be measured, can’t be corrected => if we had a public SDK you could upload a DisplayCAL grey calibration to HW and it will be fine)
Also, could you explain to me what the wording Measured vs. assumed target whitepoint weighted ΔC’00 is?
If white is white (regardless of CCT blue-yellow axis), if it is a natural white with no pink-green tint.
Measured vs. assumed target whitepoint ΔE*00 =~5dE
High Measured vs assumed => not white, some pink or green tint according to Spyder X White LED mode.Measured vs. display profile whitepoint ΔE*00 = ~4dE
If profile white and measured white matched it would mean that LG software and DisplayCAL agree on measurement … so LG is not using the same correction (maybe none at all, garbage software)But since Sptder X is not accurate at all… both measurements can be wrong => get an i1DisplayPro.
Calibrite Display Pro HL on Amazon SpyderX Pro on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.2022-04-16 at 21:46 #35127
AnonymousInactive- Offline
At the moment I can’t afford an i1DisplayPro, could you tell me why as Assumed target whitepoint I have 6300k and not 6500? Is there any way to change this?
2022-04-16 at 23:49 #35128At the moment I can’t afford an i1DisplayPro, could you tell me why as Assumed target whitepoint I have 6300k and not 6500?
Because it is the closest one to measured:
Measured whitepoint:
xy 0.3168 0.3255 (XYZ 97.34 100 109.89), CCT 6303KLG software failed to get to desired white due all of its errors: wrong correctuon and all other simplificatioojs explained in Dell or Benq HW cal threads.
Is there any way to change this?
No and you do not want to.
White is not white and if you compare it to the CLOSEST natural white (assumed, 6300K CCT AND in daylight locus, 6300K alone means nothing since its a set of several colors, some pink, some green, onlye 1 “natural white”, explained in wikipedia fro example) is at X dE distance (5 in your case).
Distance to any other daylight locus in +-100K increment will be further away.
If you got let’s say 1.4 dE to 6300K it will mean that it is white although a little warmer (yellow) than D65 but white.
In the same way you can get 6500K CCT and have a pinkish awful white (high dE to assumed 6500K and daylight locus.CCT in kelvin alone means nothing because you cannot address white point with just 1 coordinate in a plane, you need two, thus LG report is useless (also Dell , Benq and lots of reposrt from other tools and reviews.
You need two coordinates: CCT = b* axis ; color tint pink-green = a* axis,
dE to assumed white tries to give you a hint of such “tint” in pink -green a* axis.2022-04-17 at 0:39 #35129
AnonymousInactive- Offline
Why, for example, with a measured whitepoint of 6405K and an assumed target whitepoint of 6400K do I have an error of 4.43 in Measured vs. assumed target whitepoint ΔE*00? Shouldn’t the error be minimal? And why in Measured vs. display profile whitepoint ΔE*00 do I have an error of 4.12 if the profile whitepoint is 6469? I’m sorry, but I just don’t understand why these errors are so large and not small.
2022-04-17 at 12:46 #35130Explained several times in this thread, including last message. CCT means nothing and gives you no information about whitepoint.
2022-04-17 at 14:20 #35131
AnonymousInactive- Offline
OK, I get it. Why put it in the displaycal report though? Doesn’t it get confusing? Having said that, if this is unnecessary what do I need to note in a report if I want to know if I have a good white point?
2022-04-17 at 14:36 #35132OK, I get it. Why put it in the displaycal report though? Doesn’t it get confusing?
Quite te opposite. It is the only way. Your LG report saying wp is 6556K CCT is useless. Displaycal report saying distance to closest daylight locus white gives you actual information on whitepoint color.
Having said that, if this is unnecessary what do I need to note in a report if I want to know if I have a good white point?
-Closest distance to daylight locus (or blackbody, whatever you aim to, usually daylight)
-Warm-coolness (CCT)
In that order2022-04-17 at 14:44 #35133
AnonymousInactive- Offline
Can I ask you if all the programmes I open will be displayed correctly once I have calibrated with displaycal? Will the profile created be applied to all applications or only to some? Will brave and obs have the profile applied to them or will these apps ignore it and so I will see (when using these programs) wrong colours as if I hadn’t done the calibration?
2022-04-18 at 10:00 #35134Your display is sRGB so most sRGB content will show “close to what it should be” on that monitor even without color management, but if content has another colorspace or TRC you’ll need a color managed imave/video viewer as with any other display.
2022-04-18 at 13:14 #35135
AnonymousInactive- Offline
Why does the measured white point of my screen change throughout the day, is this normal? How much does it usually change? Does it change anything else? Considering this when should I do the calibration? How long after switching on the monitor?
2022-04-18 at 17:23 #35136
AnonymousInactive- Offline
72% NTSC=100% sRGB? What happens if I go to calibrate a monitor with 72% or more NTCS with displaycal? And between the two colour spaces, which one is better for films, series, video games and other entertainment products?
2022-04-18 at 17:46 #35137Why does the measured white point of my screen change throughout the day, is this normal? How much does it usually change? Does it change anything else? Considering this when should I do the calibration? How long after switching on the monitor?
Because your measurement device is paperweight, or because you have some app in your computer to change WP in GPU shaders (on embebed in some clock in monitor firmware under some low blue light configuration), or because monitor is boken, or all of them together at the same time. You’ll have to check one by one.
A common good nor premium monitor should have stabilited greatly its wp after 30min-1h warmup.
- This reply was modified 1 year, 11 months ago by Vincent.
2022-04-18 at 17:52 #3513972% NTSC=100% sRGB?
Usually if a display is advertised as 72% NTSC it means common 9x%-100% sRGB monitor with little to none colorspace outside sRGB.
What happens if I go to calibrate a monitor with 72% or more NTCS with displaycal?
Nothing special. DisplayCAL calibrates grayscale, it cannot do more.
Then after calibration display behavior is measured and a ICC is created with such data (and grey calibration embebed into VCGT in ICC file).And between the two colour spaces, which one is better for films, series, video games and other entertainment products?
If you are asking for a jack of all trades configuration: sRGB only monitor like yours, use gamma 2.2 as display TRC.
-
AuthorPosts