![]() ![]() “Cerulean” is just a name for various pigments, so I’d say the claim that cerulean is “closest to cyan” makes no sense. That’s the terminology you should be using when explaining something to someone else, so that they can read up on it independently. I indeed overread that you were referring to “cerulean” as “cyan”, which is confusing to anyone who expects proper color terminology when talking about primary colors. The reason I did not call it Cyan is because on the computer screen the ideal cyan would be far greener in hue(Like aquamarine), which would have further confused people. But I’ll promise I’ll give you pretty picture to show why when I get my little project done.īeerbaron, if you actually read the post, you would have seen I specified Ultramarine blue as the blue that is closest to the blue of screen RGB and Cerulean blue as the color that should be mixed with, which, in real life paints is the most chromatic pigment closest to cyan(and indeed used as the cyan of CMYK). Actual hsv mixing would probably feel a bit extreme and unwieldy for most users. Krita current algorithm’s just doing avaraging on the channels btw. I am currently working on this exact part of Krita, but given the confusion over color here despite all the pages, wikilinks, videos etc linked, I’ll just wait until I have something tangible to show rather then attempting to explain what I hope on doing. The reason why we don’t do linearising before mixing by default in this day an age is because the colors from gamma corrected mixing is what people are used to, and the amount of time we’d need to spent on answering the majority is far smaller than just telling the few people who want correct maths to go to a linear space.(This is also why 16bit is defaultly set to linear in Krita). Similarly, perceptual mid-grey is actually 18% in a linear space, and it feels really awkward to work in despite the pretty colors. (A lineary mixed white-to-black gradient feels overexposed to the human eye, and is thus not efficient for display, meanwhile, darker greys in an 8bit linear space look banded because we can more readily tell the difference between these colours).Īn uncorrected, or linear space doesn’t have these problems and you get lovely colors out of them, though there’s some other problems, like for example, that you must work in a high bit depth, because 8bit just doesn’t have enough room for smooth gradients in dark values. The gamma correction is necessary to make most efficient of the relatively small 8bit per channel space most computers can display. This is because the mixing doesn’t account for the gamma correction done on the channels. No, the problem he is talking about is that mixing between colors in a gamma corrected space like the default sRGB of all painting programs leads to a slightly darker color. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |