- This topic has 2 replies, 2 voices, and was last updated 5 years, 2 months ago by
.
-
Posts
-
Hi everyone.
I’m getting rather amusing results looking at Firefox 87, Chrome 89 and Safari 14, which seem to widely differ in behaviour depending on the display profile’s tone curve.
With the gamma 2.2, Firefox seems to leave CSS values alone, passing them to the display unchanged, while dramatically shifting colours in an sRGB-tagged PNG. What Safari does is less clear, as both CSS and image colours are somehow corrected, but they are corrected differently, resulting in a mismatch. Interestingly enough, Chrome appears to be the only one to successfully match identical colours from different sources under gamma 2.2.
Calibrating to the sRGB tone curve seems to completely cure Safari; Firefox remains true to its behaviour and still renders images and CSS colours differently. I’d expect changing the tone curve not to fix anything at all, so this one is especially baffling.
It’s tempting to write this off as ‘browser bugs’, but the plural number of browsers affected prompts me to start questioning my sanity and understanding of theory. Is there anything wrong with my profiling/calibration process that could cause this?
Both profiles are generated using X-Rite ColorMunki Design with the same settings, apart from the tone curve and calibration speed.
Thank you for your time.
-macos color management engine (and all apps using it) is extremely limited, use default settings in displaycal
-firefox is color managed but not by default. Google and enable full color management in about:config. This way HTML colors will be shown as sRGB.
Oh, it would never have occurred to me, if you didn’t mention it, that those flags in Firefox affect internal colours – always thought they were for untagged imagery. This explains everything – at least, as far as FF is concerned. Thanks a lot.
Regarding Apple CMM’s limitations, do you mean its inability to handle anything other than 1xCurve+MTX, or is there something else to keep in mind? I generally tend to stick to defaults, only deviating in the white point, level and, due to the specific instrument, black drift compensation. Those three couldn’t mess it up somehow, could they?
Just recalibrated using default settings, only enabling black drift compensation (I don’t think this can be avoided), but the Safari issue persists. Is there something else I’m missing?
Thanks again.
