Originally Posted by MarcusSwe
Just to get a rough estimate and try for myself, what are the "attributes" of these two corrections? I.e. if you measure them afterwards without corrections, how do the R/G/B-values line up? (For example, 100% R, 100% G, 110% B etc.) I can't use any file corrections a.t.m. and have to rely on under/over-adjusting R/G/B.
I think it's worth trying a Judd-Voss white point (x0.3067,y0.3180). How blue or not this becomes depends on your
So naturally my results may be different than yours. Different panels are different, different meters are different, and our different viewing angle and height, etc also need to be taken into consideration for perceptual matching. That's alot of things to consider, but wait there's more!
The type of polarization/filter or whatever of the display's glass is really important, as this will tint parts of the screen and there's nothing you can do about it except give yourself time to adjust to it. This is very important to understand for perceptual matching as the best options, IMO, you will have to counterbalance this in your head or use a very small window and get closer and centered vertically+horizontally on that pattern (a few inches outside this sweet spot is when tinting can be obvious).
You can toss up a full screen field, like say 70%, and then move around and pay attention to areas close but not at edges of the screen. You can see how it changes between tints of red, green and blue.
If you want, you can factor this into your judgement, so you could go for something that looks more blue when perfectly aligned with the center of the screen if that spot looked red at your primary viewing position.
I don't think comparing the RGB balance to a single setting to be a good metric for comparison (which is what harlekin did). I did this originally, and instantly dismissed non-refresh with a judd-voss derrived white point, yet in practice I found this to not be the case. As such, I didn't compare their relative differences to a single configuration like harlekin. I also had slightly different results than he did (dschilic1's matrix and the denon+B6 were less similar to each other).
I spent the most time with SillySally's CCSS
(EF9500), so that's what I've been comparing things to.
First description is mostly relative to at 100% white, but additional descriptors are mostly for 30% and 75% after giving myself time to adjust to the new color temperature (Watched Young Frankenstein, Tetsuo and/or Eraserhead every time as I'm very familiar with them).
Non-refresh: Too cool (blueish), slightly sickly (greenish)
Denon-B6: Cooler but not as cool as Non-refresh, more sickly
SillySally CCSS: This had similar red and green to Non-refresh, but needed much less blue. Sickly once I adjusted
Non-Refresh + dschilic1: Slightly cooler and slightly sickly
Non-Refresh + RichB: Almost the same as dschilic1, but slightly magenta at 30% (hard to describe)
10k CCSS: This was the first one I tried and I didn't like it at all. It was enough that I didn't check it again though (was sickly then magentaish or vice versa).
Denon-B6 + matrices: Obviously too cool and magentaish.
SillySally+matrices: Obviously sickly, slightly magentaish.
Non-refresh + Judd-Voss: My current choice. This is actually warmer than straight up non-refresh for some reason. White point addressed sickly tints. My ColorMunki Display must really like this combination.
Denon-B6 + Judd-Voss: Nope. Cold!
SillySally CCSS + Judd-Voss: Not that bad at 100% but the rest was slightly sickly-yellow.
Non-Refresh + RichB/dschilic1 + Judd-Voss: NOPE. Cold and purplish
10k CCSS + Judd-Voss: Didn't bother to try.
Denon-B6 + matrices + Judd-Voss: Same as before, but more obvious
SillySally+matrices + Judd-Voss: My second choice. A hair less warm than Non-refresh + Judd-Voss.
Again, your results may be different with your meter, and I hope this helps you in some way. I wonder what kind of results you will get.
Also, dschilic1's correction is strange. I set each set to 120 nits, but when using his what should be ~120 nits is now ~92 nits. Perhaps my meter really hates his, or this is a great example of display/meter variances.