- This topic has 3 replies, 2 voices, and was last updated 9 years, 6 months ago by
.
-
Posts
-
Hi everyone,
I’m a newbie in color-management but have finally decided to get into it. So I bought a Colormunki Display and used it throughout the week w/ DisplayCal to generate different profiles for my Macbook Pro (I neither liked the software it came with nor the profiles that it generated ;-)).
My goal is to get my MBP late 2013 close to 6504k, 100cd and Gamma 2.2.
I generated various profiles. I came up with the attached profile1. By looking at the verification report I thought that it’s actually pretty decent for a MBP.
But then I looked at various Black Level test charts (like http://www.lagom.nl/lcd-test/black.php) and realised that I can’t distinguish anything dark (I think that’s called black crush?). In the attached test image I see samples >= #5 (ambient light around 75lux). I tried many different approaches and ultimately came up with profile2 (black point correction set to auto and gammut mapping for perceptual intent set to perceptual). With that I can see samples >= #2.
I’d greatly appreciate if somebody experienced could tell me a) if either of these profiles seem to be ok for a laptop of that kind and b) whether or not my adjustments for profile #2 make any sense.
Best,
DavidCalibrite Display SL on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.Hi,
But then I looked at various Black Level test charts (like http://www.lagom.nl/lcd-test/black.php) and realised that I can’t distinguish anything dark (I think that’s called black crush?)
Note that looking at such test images is ill-conceived in many ways – you are NOT supposed to see the first four or so steps on a gamma 2.2 display with good contrast (and sites that tell you otherwise are simply wrong!) unless you view the test image in isolation (no UI elements or webpage surrounding it) on a black background in a dark room, because the luminance increments are so low. I.e. RGB step 1 (of 255) on a 100 cd/m2, 2.2 gamma “output offset” display is only 0.000005 cd/m2 above black, which is imperceptibly low (unless viewed in isolation). The next three go up to 0.00011 cd/m2 above black, which is still too low to be visible in typical lighting conditions unless viewed in isolation.
If the browser color manages the image and assumes sRGB as the source colorspace, then the situation is a little different and it is likely you’ll make out even the first steps (under good conditions).Thank you Florian. That answer is enlightening and will make me go with my first profile without the black point correction and the gammut mapping. Any objections or any recommendations to make my verification report any better? Out of curiosity, what would be a good test to find out whether or not blacks get crushed?
Out of curiosity, what would be a good test to find out whether or not blacks get crushed?
Depends a bit on the display black level and the instrument. The ColorMunki Display can go as low as 0.003 cd/m2, so it should be able to read a (tiny) difference between RGB 0-0-0 and 1-1-1 on a display that does not have an even lower black level. On LCD screens, due to their viewing angle dependency you can also look at the display from the side and see the difference in levels more clearly than when looking directly on it from the front. But clipping is not a common problem on (decent) computer monitors to begin with: If the display’s contrast control hasn’t been changed (usually the only setting where it doesn’t distort the digital signal is the factory default), the chance that the display is actually clipping near-black values is relatively low from my experience.
