I’m not sure what the math is used to represent the data in DT as RYB on the scope but I did come across this paper… some nice figures comparing the representation of the two spaces…
http://nishitalab.org/user/UEI/publication/Sugita_IWAIT2015.pdf
That is the whole point of colour harmonies - change colours to make them look nicer. If you don’t want to do that, why even start the discussion?
Of course, when dealing with skin, that is always prioritised. Painters would often choose clothing/prop/decor colours with harmonies in mind. As photogrpahers, we have the chance to do this both before capture (mise en scene) and after (editing). If all you want to do is capture reality as is without interference, then everything in this thread is irrelevant.
Hi, my name’s Ron Francis, I’ve been a professional artist (painter) for around 45 years, have a deep understanding of colourspaces, including Munsell, and have a reasonable history of computer coding.
I responded to Bruce Williams’ youtube video on colour harmony, and rather than make my case again, I will cut and paste my response here.
[quote]This is a neat idea and well implemented, but colour harmonies based on RYB don’t make sense. I think it comes from the Itten book of colour, which is responsible for a lot of misinformation.
RYB is a colourspace for pigments, so that mixing complementary pigments should produce a neutral grey. But this is wrong on a few levels. Pigments vary so much that often multiple colours on different parts of the colour wheel will act as a complementary to one particular pigment.
Also, the RYB colourspace is inferior to CMY for pigments, as CMY will produce a significantly larger gamut than RYB when mixing. In short, RYB is an old idea that is/should be now redundant, yet many artists hold on to it because they were taught that they are the primary colours in paint in their youth.
Anyway, colour harmony should be based on light, not pigment, so RGB should be used instead. The idea being, the complementary to red is cyan, so if you want to add a colour to 'balance the red", cyan will add the other colours of the spectrum, leaving your eye not wanting to see another colour as the whole spectrum is present. These are called optical complements.
Really, I see absolutely no reason to use RYB to find colour harmonies, and I have read no evidence as to why it should work. (What reason is there that green should be the complement of red?)
At any rate, if this system is going to be worthwhile, there should at least be a toggle to use RGB rather than RYB, but in my view, RYB should be relegated to a historical oddity.[/quote]
Just a side note, pigments mixing is partitive, not subtractive. Some of it is subtractive, but on a superficial level, light is bouncing directly off particles on the surface and those wavelengths are mixed together in the eye. Mixes can be reproduced in sRGB, but I believe the code would be very complex and based on large data sets of spectrophotometer readings.
15 Likes
thats fine for video, where a whole bunch of footage is put together plus a lot of color grading is done and consistency of skin colors is more important than a correct reproduction of the skin color of the artist. Photography usually is a quite different thing…
Commercial photography isn’t.
This is such an important side-note.
Subtractive mixing should be relatively easy/straightforward to implement in the digital domain. Real world pigment mixing and I believe a good part of analog film dyes do not mix simply subtractive. Hue skews are abundant when mixing pigments (and dyes to a degree).
In that context I wonder if RYB color harmony is an attempt to recreate more complex color-mixing effects?
1 Like
Colour Wheels may be great fun — but when it comes to (pigment) mixing,
things may be a bit complicated. Do pure primary pigments really exist?
The brusque answer is: No, they do not exist in real life.
excerpt.pdf (743.5 KB)
Have fun!
Claes in Lund, Sweden
Here’s an algorithm for just that: http://scottburns.us/subtractive-color-mixture/
There’s also some interesting information on the difficulties of modelling paint mixing.
I believe RYB is an old idea, and was never correct in the first place. Maybe it came about through a lack of pigments, and even now we don’t have pigments all that close to magenta and cyan, but we do have some more modern ones that are much closer.
I just don’t think pigment complementaries should have anything to do with colour harmony theory.
1 Like
Color harmony is a perceptual phenomenon, for human observers. There is no a priori reason for it to conform to any particular color space that we find convenient for other reasons.
At the end of the day, color is an artistic choice like many other things. There are some rules of thumb, but in themselves they are not sufficient to guarantee good results in either painting or photography.
In practice, RYB may be as good enough as a starting point for photography as anything else.
1 Like
I believe that my Huawei phone (and maybe others) is using RYB.
As I understand it … Y allows through the filter more light than G … and … G can be a product of Y + B.
YCbCr…which I am not sure is the same…???
There is only one reason that I can think of that makes any sense of these colour harmony systems.
That is, the colours when added together will create an achromatic colour. That’s what adding complementary colours does, whether they are directly opposite, or split into three or four separate segments. I don’t understand why complements would be used at all otherwise.
But pigment complementaries only work for pigments, and even then there is much variation.
The after image of looking at a colour will be its optical complement, not pigment complement, so if the idea is to not leave your eye wanting to see another colour to complete the spectrum, optical complements is the only system that can achieve it.
As I said elsewhere, RYB is an old idea that has been debunked anyway, so I don’t know why it is used for anything at all.
2 Likes
Many photographers (and of course other artists) find color harmony useful because it gives them a framework to produce colors with a certain effect. This does not necessarily require that colors satisfy any mathematical relationship, the various color arrangements are just a starting point.
From this perspective, a mathematical model for individual hues is much too simplistic, as color perception of an area is affected by many factors in an image (the size of the color patch, surroundings, illumination, etc). Perhaps that’s why artists don’t pursue models of color harmony with strict mathematical properties, they miss key effects.
As for photography, you color grade the image, not its mapping to any color wheel or color space. Just like you don’t work on the histogram, but the image itself, even though it is useful as a tool.
5 Likes
Thanks Tamas - that is the most insightful comment about colour harmony that I have read for a long time!
1 Like