Talk:Complementary colors

This article is written in American English, which has its own spelling conventions (color, defense, traveled) and some terms that are used in it may be different or absent from other varieties of English. According to the relevant style guide, this should not be changed without broad consensus.

This article is rated Start-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects:

Color High‑importance This article is supported by WikiProject Color, a project that provides a central approach to color-related subjects on Wikipedia. Help us improve articles to good and 1.0 standards; visit the wikiproject page for more details.ColorWikipedia:WikiProject ColorTemplate:WikiProject Colorcolor articles High This article has been rated as High-importance on the project's importance scale. Visual arts This article is within the scope of WikiProject Visual arts, a collaborative effort to improve the coverage of visual arts on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.Visual artsWikipedia:WikiProject Visual artsTemplate:WikiProject Visual artsvisual arts articles

There are some fundamental problems with this article. Primary colors do not allow us to form the full range of human perceivable colors. Take a look at gamut and also Google it and you should be able to find that the gamut of a trichromatic TV or computer screen is far from the entire range of possible colors. The discussion of primary colors as a means of forming a color gamut doesn't belong in an article on complimentary color in the first place. Primary colors are only relevant to complementary colors from a historical standpoint where, for example, certain primaries/secondaries were picked for painting, etc. But from a color science standpoint, any color has a complementary such that when the two are added they will form gray (or the color of the illuminant in a subtractive system). It's irrelevant which colors are primaries.

If you want to understand primaries, the true authority is the CIE. Again, you will have to go outside of Wikipedia to get a thorough understanding of this topic. The short answer is that the only way to get the full gamut of human perceivable colors with three primaries is to make your primaries imaginary colors (or allow subtraction of colors in an otherwise additive system).

If this makes your brain hurt, you're not alone. I don't really feel completely authoritative on all this either. I will try to complete a complementary wavelength article to go with the dominant wavelength article when I have time (not for a while I'm afraid). This should cover complementary color except for the historical stuff, which we can keep here.

I'm really sorry to just complain and not help fix things, but I just don't have the time right now and thought I'd better at least point out the problems as I saw them. --Chinasaur 08:43, 4 Oct 2004 (UTC)

I agree with you on the inappropriateness of referring to the primary and secondary colours in this article, so I did your dirty work for you and deleted the references. I am looking forward to your article on complementary wavelengths. :-) --Heron 11:50, 4 Oct 2004 (UTC)

It should be recognised that here is a theoretical and a practical approach. The practical approach is based on the three types of colour receptors - cones - in the human eye: red, green and blue. This is the origin of primary colours. Thus it makes perfect sense to use them.

TV screens are three-chromatic for the above reason. As to the imperfect colour reproduction, it is not a limitation of the three-chromatic approach but was a technical problem. Engineers couldn't make a better phosphorus, later LEDs. The quality was limited by the technology. Yes, the still current sRGB is poor with it's 16 million colours. However, current digital movie cameras have 16 bit/channel colour resolution IE:2.814749767×10¹⁴ - 281 trillion - colours. This is important during the editing but not for the end product. A DLP cinema projector is offering 35.2 trillion colours. More than enough to exceed the limitation of the human eye - about ten times.

Now the author of the article needs to decide how to combine/separate/discuss theory vs practicality. Naturally, neither are right or wrong. Just different. — Preceding unsigned comment added by 115.70.29.185 (talk) 03:25, 4 May 2021 (UTC)[reply]

I tried to make the orthography of the article consistent. There were mostly "colour"s but some "color"s. But the article is named "color" so I changed it that way. If a great majority of people working on this article are British and have trouble remembering the American spelling, we might want to consider a move to Complementary colour, but otherwise I think this is the easiest solution. --Chinasaur 18:15, 4 Oct 2004 (UTC)

Colour is the spelling recognized around the world, color is the spelling in just one nation. I support formally changing the spelling of this word everywhere it appears. 66.46.152.30 18:36, 11 August 2006 (UTC)[reply]

I reverted your changes. Wikipedia practice is to keep whatever spelling was used first in the article; in this case, "color". See the Color talk page for the most recent of numerous lengthy debates on this topic. -- Laura S | talk to me 19:17, 11 August 2006 (UTC) Put also spelt Colour on the opening sentence. —Preceding unsigned comment added by 88.105.51.100 (talk) 12:50, 23 October 2010 (UTC)[reply]

This is ridiculous! This article says that yellow is the complement of purple, but at yellow it says that yellow is the complement of blue. 66.245.26.248 18:32, 27 May 2004 (UTC)[reply]

Good point. I have tried to fix the article. -- Heron 16:03, 7 Aug 2004 (UTC)

I will look over this article with an eye to both psychology, where red is the complement of green and to photography, where red is the negative of cyan. Artistically, both cyan and green make adequate complements, but to the photoreceptors in the eye, green is better. Photographically, red is the negative of cyan, not green. It's worth noting that the receptors for blue in the eye are very well suited to receiving violet. IOW, more difference is between the peaks for blue and green than the peaks for red and green. My main reference is a textbook that I no longer possess: Physiological psychology (I might even need to look at the University of Alberta calendar to remember the author). BTW, violet is a better word than purple for the complement of yellow, because purple can be confused with dark majenta or a 2:1 mix of majenta and cyan. Know what? Your monitor doesn't do violet.—Preceding unsigned comment added by Brewhaha@edmc.net (talk • contribs)

It would be better to start a new section than to respond to an old item from 2004. As for your concept of "negative colors", please provide a verifiable source. You seem to be taking this article into a direction that is unfamiliar to me, and I'm quite familiar with color science and its application in photography. I'm going to revert your edit for now, since I think this "negative colors" approach is rather non-standard. I recommend you start over with a new section below, and certainly back up any major changes with sources. Dicklyon 15:32, 30 March 2007 (UTC)[reply]

Neil R. Carlson calls it the 'opponent-color system', and it's based on action potentials in ganglia. I've been a wikipedian for quite a while, and I don't think I've ever reverted an article in entirety. The question that gets me is how the brain confuses violet with a mixture of blue and red, and I'm looking at graphs of light absorbance in pigment to put it in words. Brewhaha@edmc.net 11:35, 24 June 2007 (UTC)[reply]

Now, can anyone answer the question "Why are the 2 sets of colors different"?? Please read this very slowly and carefully:

In the summer of 2003, I stumbled across a site at http://www.gamecheetz.com/Rant.html that appears to explain very logically. It has a color section. When I stumbled across it, I thought it was logical in every way except one: it has not been widely adopted as standard. In February 2004, I edited various color edits to see how difficult the information could ever get widely adopted. Now, go to Talk:Color. The 7-section supgroup titled "What are the primary colors??" that is being discussed about the edits that I made sounded strange, but now I will reveal the truth: all I was trying to do is make the information mentioned in the above external link widely adopted, and it turned out that it can't. Three months later, I contacted the owner of the web site, convincing the owner that their information cannot be widely adopted, and the reason turned out to be that it was only the webmaster's opinion as of when the information was presented, and the opinion was changed. "The black-white scale and blue-yellow scales each have gray in the center, but the red-green scale has yellow. Thus, yellow is both a primary and a secondary color! No other color is both primary and secondary." (In contrast to the webmaster's previous belief as of when I stumbled across the page, which was all 3 scales contained gray.) Even later, I asked the webmaster how the red-green scale looks to someone who is red-green color blind and the answer was that the webmaster (contrary to what I thought when I stumbled across their site) is not an expert on colors.

There are two kinds of red-green color blindness. One is called protanopia and represents when the red cones are filled with green pigment. Deuteranopia represents the transverse. Tritanopia is about a complete absence of blue pigment. It's very rare. In protanopia, the subject would see darkness towards the green end. Someone with the second kind of anopia would see darkness toward the red end. Brewhaha@edmc.net 11:35, 24 June 2007 (UTC)[reply]

Now, are any of you registered Wikipedians able to answer the question with your best knowledge: "Why are the 2 sets of primary colors in paint and light different??" 66.32.241.40 02:18, 8 Aug 2004 (UTC)

As for that rant, the guy is seems to have mulled over the issue, but came to the wrong conclusions because he just doesn't know how the brain perceives different colors. The most comprehensive site on color I have found is http://www.handprint.com/HP/WCL/wcolor.html, and if you are interested I suggest you check it out. The reason all this information isn't discussed is because I think it's out of the realm of wikipeida (as you can see there there is a TON of info... too much to include in one encyclopedia article).

FYI the reason there are 2 sets of primary colors is because paint absorbs light, and colored lights emit it (that's why the primary colors of light are called "additive primary colors" and the primary colors of pigment are called "subtractive primary colors". Ctachme 05:04, 8 Aug 2004 (UTC)

No. That's not the question. The question is whether green or cyan makes a better complement for red. A paint artist will tell you green. A neurobiologist will tell you green. Another question is whether violet makes a better complement for yellow than blue. It's a shame that someone simply reverted my edits. What am I supposed to do? I would resolve the controversy by distinguishing complementarity from negativity. Brewhaha@edmc.net 11:35, 24 June 2007 (UTC)[reply]

Yes, Ctachme, that's true, but painters use a third set of primary colours: red, blue and yellow. I think these were chosen because they were once the only colours that could be manufactured in a stable form suitable for paint pigments, and modern painters still refer to these as the "primary colours". Traditionally, yellow was made from orpiment (an arsenic compound), blue from lapis lazuli (a gemstone), and red from Chinese cinnabar or vermilion (a mercury ore). The complements of these pigments are red/green, blue/orange and yellow/purple. -- Heron 16:18, 8 Aug 2004 (UTC)

But shouldn't the stance be that the so called "painter's primary colors" are only worthy of historical note because as you just said, it was based only on available pigments, not on the actual colors that the eye considers primary? I see no reason to pretend this historical note is still fact. The three subtractive primary colors are yellow, cyan and magenta with compliments of blue, red and green, and the additive primary colors are just the opposite (blue/yellow, red/cyan and green/magenta). Ctachme 02:38, 9 Aug 2004 (UTC)

The controversy rayjez even today between technicians and artists. I've been led to distinguish negative colors from complementary colors. Maybe I'm trailblazing. Nothing is new in that. It's worth noting that cyan is next to green and violet is next to blue on the spectrum. The neurobiology serves as the initial stage of what allows some people to find a number between about one and sixteen for a hue. Brewhaha@edmc.net 11:35, 24 June 2007 (UTC)[reply]

OK, I concede that there is no value in promoting the RYB 'primaries'. My point is that there are still many books out there that cling to the old theory, and that people should be able to find that theory somewhere on Wikipedia, even if it is only in an historical note. Would you accept, for example, a note under blue saying "Blue is an additive primary colour. It was once considered a subtractive primary colour, the complement of orange, but this theory has now been superseded by the more accurate cyan/magenta/yellow system of subtractive primary colours"? -- Heron 08:36, 9 Aug 2004 (UTC)

Sure, that would seem fine to me, as long as all of the colors were updated to reflect this. Ctachme 11:53, 9 Aug 2004 (UTC)

Changelog:

• I explained the function of additive complementary pairs.
• Gray is (literally) a shade of white; changed "shade of gray" to "shade of white".
• RGB section ended up being adupe of additive section; I removed it and will copy the formula from it to the additive section at a later date, with a better example than black and white.
• Looking at a chart is not a good way of remembering... Some might consider this cheating.

I'm still not pleased with the complementary subtractive section. I don't think purple and yellow or red and green pigments will combine to produce the illuminant color. But it's 2:25am and I get up in 6 hours for work so, another time. Ayeroxor 06:27, Jun 23, 2005 (UTC)

Yes, sorry, I think the "color of the illuminant" was something of a red herring for this discussion. Sorry because I think I put that in at some point. I think the current article is now at least roughly accurate, but not very readable. Hopefully more color science people and artists will help correct technical errors while lay people can make it things more readable in WP's iterative process from here. --Chinasaur 22:20, 28 December 2005 (UTC)[reply]

I'm taking those color pairs back out. Could you justify including them? They don't make much sense to me and I doubt they have much formal validity in color science. They are more relevant to CMKY printing than anything else I can think of. --Chinasaur 09:21, 18 January 2006 (UTC)[reply]

They make a lot of sense to most people :o) There is very strong empirical evidence that the three primaries and the three secondaries form the main concepts by which colour perception in general is ordered; certainly RGB sensors are the main components of the sensory apparatus. It is thus useful to order them in complementary pairs and mention them explicitly, the more so because it is very common to do so and we as an encyclopedia must reflect this social practice. It is not our place to endorse or condemn. However we might add that the value of this ordering in colour science is very limited. Of its "formal validity" I know little; but a material validity is present ;o)--MWAK 12:31, 25 January 2006 (UTC)[reply]

If you think about it, there should be a shade of grey that when inverted, will produce the same colour (ie; it is exactly halfway between black and white). Does anyone know what its hex is equal to, and if so, could you add it to the article as a piece of trivia?Tuck99 04:21, 25 February 2007 (UTC)[reply]

The closest you can come, in a typical 8-bit space, is 127 or 128, which are inverse of each other. It's too trivial to be worth a mention. Dicklyon 05:06, 25 February 2007 (UTC)[reply]

I'd say it depends on what you mean by "inverted", and on factors like amount of ambient light, etc. “Complementary” usually just refers to hues, in my experience, not even really to particular colors. --jacobolus (t) 22:19, 10 August 2007 (UTC)[reply]

The article currently claims:

In the RGB color model (and derived models such as HSV), primary colors and secondary colors are paired in this way: Red and green, yellow and purple, blue and orange.

With all the people obviously knowing the differences between the various color wheels and theories, how can something as commonly known as the RGB color model and it's complementary colors still be wrongly described? In the RGB color model, red is the opposite of cyan, green is opposite of purple and blue is opposite of yellow. The first image in this article even confirms this. Unless anyone can justify why the piece of text I just quoted is correct, I'll edit the above text. JoaCHIP (talk) 18:39, 1 March 2009 (UTC)[reply]

You are right. I think this is obvious enough that I have finally edited the text. The opposite of red (255, 0, 0)is obviously (0, 255, 255), which is cyan, so you are obviously right. You can use any RGB color scale to prove this. Same with blue with yellow [(0, 0, 255) with (255, 255, 0)] and green with MAGENTA (not purple) [(0, 255, 0), (255, 0, 255)]. (255 can be substituted by any number, really, but you get the idea). I think this justifies editing that piece of text, and so I did. T.c.w7468 (talk) 23:25, 5 May 2009 (UTC)[reply]

Oh yeah and by the way... http://en.wikipedia.org/wiki/File:RBG_color_wheel.svg . There's some proof. T.c.w7468 (talk) 23:30, 5 May 2009 (UTC)[reply]

This was also labeled incorrectly in the afterimage section. The afterimage opposite of red is also cyan, so I've edited "green" to say "cyan" as it was clearly wrong, and any image if an inverted American flag will use cyan, and not green, as its red opposite. MXVN (talk) 05:52, 6 December 2011 (UTC)[reply]

I noticed that a few professional sports teams, such as the Mets, Knicks and Islanders (blue & orange), along with the Lakers (purple & gold), use complimentary colors as their primary colors. Is that relevant information? Mr. Brain (talk) 04:09, 4 May 2009 (UTC)[reply]

transparent is something that lets light right through

translucent is something that makes light scatter

opaque is something that does no let light through

Some editors are removing large swaths of this article and replacing material with unsourced information and conjecture. Perhaps before making further dramatic changes, the issue should be discussed here. Taroaldo (talk) 19:23, 18 July 2010 (UTC)[reply]

I just removed a clarification needed tag from afterimages, as i think it was pretty clear how the afterimage is formed. I added a bit more clarification just to make it more clear.Larryisgood (talk) 16:41, 26 January 2011 (UTC)[reply]

Colored shadow redirects to this article, but the term is not explained, here. Shouldn`t that be changed?! --Hans Dunkelberg (talk) 08:45, 3 July 2011 (UTC)[reply]

It is a grave heresy, unless we essentially redefine "red" with a deviation towards magenta and "green" with a deviation towards cyan. Which sources speak about this and what exactly they say? Incnis Mrsi (talk) 10:48, 19 January 2013 (UTC)[reply]

Russian Wikipedia refers to books of Bride M. Whelan. Does anybody read his books? Incnis Mrsi (talk) 11:03, 19 January 2013 (UTC)[reply]

Well, we should look at it this way: under the Wikipedia NPOV rule we must try to give an exposition of both systems: the traditional one and the scientific one. We give them due weight by indeed saying that the one is scientific and the other traditional and mention the latter last. Of course reality is a bit more complicated as is shown by works as those of Mr Whelan who take a sort of "in between" position: they take a traditional colour wheel and then squeeze in cyan and (not very saturated) magenta somewhere. That makes such "systems" severely incoherent but that doesn't matter much as they are not even popular science books: they are just popular. They aim at giving the reader a happy colour experience and perhaps help him a bit in decorating his house (should the poor man try to actually mix paint according to the precepts he is in for a very unhappy surprise, obviously). Such books cannot be used as serious references.--MWAK (talk) 09:28, 21 January 2013 (UTC)[reply]

I don't understand the problem with red–green complementarity in the RYB system. I'll add a book source. Dicklyon (talk) 03:26, 8 June 2013 (UTC)[reply]

Do you want to say that a painter’s red and green can be mixed (in appropriate proportion) to produce something perceived nearly as a neutral grey? Incnis Mrsi (talk) 04:57, 8 June 2013 (UTC)[reply]

I probably wouldn't go that far; they'd probably make a muddy brownish color, like this guy shows. But they're still considered complementary in that system, and if you used a red slightly on the magenta side and a green toward the cyan side a little, as you note, they might make a neutral mix. If you mix appropriate amounts of the three primaries red, yellow, and blue, you should be able to make a neutral; the relative amounts of blue and yellow in that mix will be a green of sorts, no? Of course, it does depend on what the absorption spectra of those primaries look like. Here's a page where red and green paints mix to make a fairly neutral blackish color. Dicklyon (talk) 05:15, 8 June 2013 (UTC)[reply]

I agree that a non-spectral (i.e. without a dominant wavelength) “red” and a non-standard wave-shortened “green” (something like emerald) can be perceptually complementary. So named (ironically) Natural Color System really makes namely this choice. But it has no sense (except for making linguistic constructions) to draw a triangle (or hexagon, or rhombus) diagrams and to label vertices and midpoints with some words like “red”, “yellow”, “green”, and “blue”, without specifying their meaning. Is RYB color model presented in a more intelligent way by B.M.Whelan s and most Wikipedians? One can draw an arbitrary color triangle (such as yellow–blue–black), make custom “secondary color” definitions (such as white=yellow+blue, brown=yellow+black), and then say that blue and brown are complementary in the YBK color model. Incnis Mrsi (talk) 06:04, 8 June 2013 (UTC)[reply]

---

P.S. I concluded that the source of confusion is conflicting definitions of “red”. I tried to explain something of it at b:Color Theory/Misconceptions – feel free to correct if needed. Incnis Mrsi (talk) 13:36, 15 June 2013 (UTC)[reply]

I spent 4 days without Internet, and today surprisingly see all references to them disappeared from the article. It is not good. Incnis Mrsi (talk) 13:10, 3 June 2013 (UTC)[reply]

Plese don't worry, the historical section isn't finished. Goethe will return shortly, along with Maxwell and Young. I just need to line up the correct references. SiefkinDR (talk) 14:40, 3 June 2013 (UTC)[reply]

Why were they removed? I think you need to work more slowly and carefully. Dicklyon (talk) 05:17, 8 June 2013 (UTC)[reply]

Incnis, you, too. I now see that in this edit you added an unsourced assertion that the widespread concepts of complementarity in the RYB system are just wrong. You don't get to decide that. I always revert unsourced assertions of errors and confusions when I see them, and I'm sorry I missed this. It is not WP:NPOV. We need to respect and fairly represent even those color systems that predate the more colorimetrically accurate concepts. Dicklyon (talk) 05:26, 8 June 2013 (UTC)[reply]

When I edited the article, the only reference was Theory of Colours (van Gogh’s perverse Café was absent at the time), and Goethe did not write a Whelanian crap, he actually examined and researched the perception. Theory of Colours does not describe the complementary of green as “red”. Maybe van Gogh did, but Goethe did not. Incnis Mrsi (talk) 06:15, 8 June 2013 (UTC)[reply]

Theoretically it is true, as rules for mixing differ, but actually (not terminologically) this difference is tiny. The citation intended to suggest (although it does not explicated) that van Gogh was right with his red–green “complementarity” but in his (pigment) system. No, he was not, and this pair was likely a misconception appeared somewhere in 19th-century.

There is almost no difference between RGB’s and pigments’ complementaries in actual colors (not names). The case with orange: actually, the same color, but referred to as “blue” (except in heraldry) historically and as “azure” since 20th century. These words were nearly synonymous and diverged with establishment of modern color terminology in mid-20th century. Just look at the “blue” of van Gogh’s paints.

The complementary of yellow, nowadays called the (color wheel) blue, was historically known as indigo: read the latter article to find a citation. It has no big difference with violet, which Goethe named as yellow’s complementary. BTW, Goethe’s “violet” is hardly different from (modern) “indigo”. Goethe did not make a noticeable error with these two pairs.

Finally, subtractive complementary to red is known to be cyan, never green. So, “RYB complementaries” are a misconception having little to do with any (either physiological or pigment mixing) color process. But the article retouches this consideration down. Incnis Mrsi (talk) 20:24, 8 June 2013 (UTC)[reply]

The traditional role of red and green as complements in the RYB system that painters use is probably more related to a color opponent process than to the idea that complements mix to give a neutral. Either that or the colors that they call Red and Green are a bit off from what we call red and green in the RGB system. But trying to explain the traditional RYB system in terms of modern colorimetry and the RGB system is a bit of a losing game in my opinion. Let it stand on its own and you'll be less bothered by it. Dicklyon (talk) 01:34, 9 June 2013 (UTC)[reply]

A bit off, really? Get van Gogh and check whether he used a-bit-offed NCS so-named “red“ (actually crimson) and ”green” (actually emerald), or standard (both RGB and CMYK) ones. Incnis Mrsi (talk) 03:37, 9 June 2013 (UTC)[reply]

I wouldn't know how to get a hold of him. Dicklyon (talk) 04:51, 9 June 2013 (UTC)[reply]

It's true that Van Gogh said he was particularly crazy about using a crimson lake, which seems to have been alazarin crimson and thus close to magenta, and emerald green, closer to cyan, but for him those were red and green. SiefkinDR (talk) 06:07, 9 June 2013 (UTC)[reply]

BTW commons:Category:Le Café de nuit shows a great divergence in colors: one photo even shows an unambiguous cyan instead of ”green”. It’s not good. Incnis Mrsi (talk) 06:52, 9 June 2013 (UTC)[reply]

This book Blue and Yellow Don't Make Green has a good discussion of the different variants of red and green, and other pairs, that can be used as complements. Dicklyon (talk) 18:34, 19 June 2013 (UTC)[reply]

Can we please come up with a better image for the top of this article? The picture of the pilot flying blind is confusing and adds no information at all about complementary colors. Most people who come to this article simply want to know what colors are complementary. I believe we need a color wheel or chart that shows people the complementary colors. Thank you! SiefkinDR (talk) 13:45, 9 June 2014 (UTC)[reply]

What are complementary colours 102.221.220.128 (talk) 16:44, 22 June 2022 (UTC)[reply]