Talk:RYB color model

From Wikipedia, the free encyclopedia

This article is supported by the Color WikiProject, 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.
Stub This article has been rated as stub-Class on the quality scale.
High This article has been rated as high-importance on the importance scale.

Contents

[edit] incorrect by what standard?

"A primary color (or colour) is a color that cannot be created by mixing other colors in the gamut of a given color space. Primary colors may themselves be mixed to produce most of the colors in a given color space" -Wikipedia

"primary color n. A color belonging to any of three groups each of which is regarded as generating all colors, with the groups being: Additive, physiological, or light primaries red, green, and blue. Lights of red, green, and blue wavelengths may be mixed to produce all colors. Subtractive or colorant primaries magenta, yellow, and cyan. Substances that reflect light of one of these wavelengths and absorb other wavelengths may be mixed to produce all colors. Psychological primaries red, yellow, green, and blue, plus the achromatic pair black and white. All colors may be subjectively conceived as mixtures of these. See table at color." -Dictionary.com

1. Red Yellow and Blue make up all the colors in the RYB color space, so they are its primaries 2. in any event I fail to see how a color model itself could be deemed "incorrect" simply because it doesn't conform to three peak responsivities of the average human eye, especially since Wikipedia has not adopted a scientific POV

RYB is flawed because it was created by artists when color was not understood correctly. People continue to use it today because 1; it has become so well established that it is correct and 2; it's flaws are generally masked by the pigments of colors that it cannot generate, as well as the use of white. Vjasper 19:34, 22 July 2006 (UTC)

[edit] What happened in 2004?

As of 2004, scientists know that this set is incorrect, but it continues to be in common use in art. says the article. Was there some discovery made in 2004 which showed this set to be incorrect? What is the significance of 2004? - Jimp a.k.a. Jim 23May05
None, very odd phrasing. I've fixed it up. --W(t) 05:55, 2005 May 23 (UTC)

Well, how common in art do you think it will probably be as of...

  • 2014
  • 2024
  • 2034
  • 2044
  • 2054
  • 2064
  • 2074
  • 2084
  • 2094
  • 2104
  • 2114

Georgia guy 20:59, 23 May 2005 (UTC)

I've got this sneeking suspicion that it'll persist for sometime yet perhaps even indefinitely. For I'm not convinced that it is really incorrect. By the way, I've seen the same phrasing elsewhere on another colour page. I'll fix that up too if I find it. - Jimp 24May05
Can someone provide the source that this model isn't correct? Even the CMYK page notes that "dark blue reproduces poorly in CMYK, coming out rather murky (as it has to be approximated by darkening cyan with black and a little magenta)." Which, IMHO, doesn't provide credibility that the CYM model is better than the RYB model (however the example might be related to print which uses inks that might not mix as well, which would make CYM a theoretic model).
Basically I would request a vote to replace "model is incorrect" to "model is replaced by the CYM model". To state that it is incorrect would mean that all painters would have to mix in CYM instead of RYB. Felsir 07:12, 29 September 2005 (UTC)
Your fault here is assuming that all painters mix RYB to create all of their colors. Painters mix pretty much any color they have together to generate a new one, and they start with many colors already. Vjasper 16:59, 15 August 2006 (UTC)

I agree that it is not accurate to say that the RYB model is incorrect. In fact many printers use "cyan" inks that are quite bluish, and "magenta" inks that are quite pink/reddish (see image below) -- i.e. essentially a compromise between CMY and RYB. There are also printers that use CMYK+Red+Blue for an increased gamut. DavidHopwood 04:54, 19 January 2006 (UTC)

Approximate CMYK ink colors

Let me see if I can describe this image in words: The C and M (but not the Y) are about halfway between their RGB color and the color after them (yellow obviously not; it would look kind of lime green if it were.) Georgia guy 22:26, 7 February 2006 (UTC)

This is true! When you buy a tube of "primary red" paint, you got a rose. —Preceding unsigned comment added by 217.195.19.145 (talk) 11:22, 21 May 2008 (UTC)

[edit] factual accuracy

I've heard/read the substractive model being critiziced on the basis that additive model is "easier" and better relates to known physics and biologics. I don't think that there are any physical/biological claims included in the substractive model, so I fail to see how could such a model be "invalidated". Someone must know these issues better and could come up with a brief explanation? 62.220.237.74 11:58, 20 Jun 2005 (UTC)

[edit] confused

If it's incorrect, why does it work? -anon

Because you can still mix in white to get cyan and magenta from blue and red, respectively. The "incorrect" part is the assumption that red and blue are primary colors. In the more accurate CMY model, they are secondary colors. I suppose if you include white as part of the RYB model (perhaps call it RYBW) it might be a viable alternative to CMYK, but CMY is simpler as it only has three base colors and they are all primary; RYBW requires four colors. Also the status of white as a primary color in this system would be unclear. If I buy a paint that is a light red, would it be a primary color, or would it be red mixed with white? This RYB/RYBW model is certainly not useless or impractical for painting or coloring with traditional media, but it's not well-defined enough for, say, computer work, where CMYK would be much better for things designed to be printed, RGB (or HSV or HSL) would be better for things designed to be displayed on the monitor, etc. - furrykef (Talk at me) 07:47, 20 September 2005 (UTC)
Actually I'm not quite sure what I wrote above is accurate, with white being needed for magenta and such, but I think the general gist is correct. - furrykef (Talk at me) 06:38, 24 September 2005 (UTC)

Any subtractive color model necessarily includes white, either explicitly or as the absence of colorant. A typical CMY[K] printer depends on the background color of the media being white, for example. Nor is there anything particularly wrong with additive models that require 4 or more colors. The costs of that need to be balanced against the increase in gamut, for any particular application. DavidHopwood 05:03, 19 January 2006 (UTC)

Yes, of course, you're correct. I don't know what I was thinking. - furrykef (Talk at me) 08:42, 1 April 2006 (UTC)

[edit] history

Since RYB is a "historical" color space how about adding a history section to the article? Anyone have the data? 129.42.208.182 19:11, 23 February 2006 (UTC)

It's not really historical, just established. Vjasper 16:57, 15 August 2006 (UTC)

[edit] Just to clear a things up...

You cannot create every color using the RYB color model. Red, Yellow, and Blue are the primary colors for the RYB color space, however that does not mean that the RYB space can be used to create every color.

A common mistake that people make (which is also made on Wikipedia) is thinking that pigments are subtracted. Pigments are NOT subtracted, they are multiplied. If pigments were subtracted then you would be able to produce negative light, which you cannot. Because light values are between 0 and 1 (0 being absolute dark and 1 being absolute light) when you multiply them they get darker. You can multiply different values of Cyan, Magenta, and Yellow, in order to generate all possible colors, however you cannot do the same with RYB. That is why RYB is not a correct color model/space/thingy. Vjasper 18:07, 22 July 2006 (UTC)

To elaborate on this just a little bit, you might ask yourself why RGB and CMY are "better" than RYB. Here's the reason.

There are two kinds of displayed light, direct and indirect. Direct light is percieved as it's color value, ie a flashlight, or the sun. If you combine direct light their colors add. Indirect light is stuff like reflections. When you reflect light off of a surface the result is the pigment color (or the inverse of the pigment color, depending on how you're doing things). When you combine pigments they multiply.

You can represent all the colors that can be produced by light and seen by the human eye by combining different values of Red, Green, and Blue. You can represent all the colors that can be produced by combining pigments and seen by the human eye by combining different values of Cyan, Magenta, and Yellow. Red, Yellow, and Blue do not represent anything specific, that is to say they are an abstract color model. It has no real practical use (you could say it's practical for artwork, however the bottom line is that it would be much better to use CMY).

EDIT: forgot to tag, sorry. Here you go. Vjasper 19:46, 22 July 2006 (UTC)

It's a POV. And i don't share it.
"Because light values are between 0 and 1" This is already a model which admit a theorical "maximum light" which is unreal. This maximum came from the screen model not from reality. In theory, absolute white is lethal! Try to ligh a fire with RGB and a loupe… You can have pure red color of various illuminances in reality, not on a screen. RYB reprensents something specific. CMYK can NOT represent all the colors. Because this is pigment, light and chimistry, a specific colored ink can't be strictly replaced by a quadri-color. A combination of pigments will never reply to light as another one. To my knowledge, fluorescence can't be represented by any actual color model, can be obtained with specific ink, can hardly be simulated by screens. Because the red and blue of the RYB model are not the red and blue in a CMYK model, you can't compare like if they were. By a primary red tube of pain and you got rose (reddish magenta). CMYK and RYB are not complete. How do you process white on non white canvas? You need white ink. C+M+J=K only in theory, so, CMY doesn't exist. Same for the RYB model, the name do not include all the needed colors to use the model, white and black are missing. RGB and CMYK are not better, only different and associated with specific usages. Screen and white paper are not the only media.
Every model is abstract and proceed to approximations. Like RGB not displaying violet but purple instead. RYB is intuitive and may follow luminances. in Red yellow green cyan blue magenta, yellow got the highest luminance, blue the lowest. Red is not in the middle with theorical luminance but on perceptive POV, it can be in the middle. Red is the color that the brain see the most. Not following sciences, RYB is still a perfectly working model, useful and intuitive. And is rarely used "as is". RYB usage is mainly oral, not precise, doesn't fit industrial usages, used in unstrict recipes. This color are often used as other paints modificators.
Lacrymocéphale
PS: Now, what I want to do, is to face children and models. I asking myself which model children find out to be the most logical. I don't think that they will find out that R+G=Y so may RGB will be in a bad position as a non-intuitive model. B+Y=G is more credible I think. I got to ask kids. —Preceding unsigned comment added by 217.195.19.145 (talk) 12:27, 21 May 2008 (UTC)

[edit] CMY

The article text shouldn't really be comparing RYB with RGB, since they're completely different models, RYB being subtractive and RGB additive. It should instead contrast RYB with CMY, which is a fairer comparison.

[edit] Thinking logical

As I understand things: a color is a color. A color cannot be be subtractive nor additive. Well you might if you would mean that a color is produced by additive or subtractive process. But again: you cannot tell from the color. Subtractive and additive colorspaces don't exist. What would the difintion be to call a colorspace subtractive /additieve? Additive means lightness goes up when mixing colors and with subtractive lightness goes down. Mixing green and red goves you lighter yellow in a additive process and darker yellow in an subtractive system yet the hue (yellow) does not change. Mixing paint doesn't add or subtract lightness, the same lightness is preserved. If printer prints with dots next to each other lightness is preserved. So both processes are not additive of subtractive. Subtractive printing process means you lay layers of paint on top of eachother. Light gets absorbed and lightness goes down. The RYB colorwheel is something completely different than a RYB color space. RYB colorwheel is about mixing colors and telss you nothing about lightness you could make it a colosrpace by adding L and S for saturation: RYBLS -space. Now it is obvious that RYB color space which in corporates lightness is something different.


So: "RYB is a historical set of subtractive primary colors." is nonsense; "The RYB space receives criticism for not being able to produce all perceivable colors." is nonsence as a colowheel doens't change it's lightness so it's obvouis that it can't get all colors. It does is capable a representing all hues yet not all saturations. And probably "In particular, several bright shades of Green, Cyan, and Magenta are not producible from any combination of Red, Yellow, and Blue (and are absent from the RYB color wheel)." would be changed to saturated in stead of brightness (which is somewhat like lightness). "In the RGB color space the colors are added, thus you start with levels of dark colors which are added to produce lighter colors." Is nonsence too. Again probably a RGB colorwheel is meant so don't link to RGB colorspace. The colorwheel nor the space are additive or subtractive. Again what would the properties be of a colorspace that is additive or subtractive? Well I'll stop.. it's to bad.BartYgor 13:58, 13 August 2007 (UTC)

I already made several changes to the summary of this article which was put at primary color, to make it more accurate (for instance changing "bright shades" to "colorful" (which is IMO better than "saturated" as a description). In any case, some of what you call nonsense is perfectly reasonable. The RYB color model is not just a color wheel... it includes all colors which can be mixed from varying amounts of red, yellow, and blue paints on white paper/canvas (and possibly including black and white paints as well, for convenience). It's not a particularly *good* (read: useful, accurate) model of color, but that is beside the point. Anyway, a "RYB" color wheel is explicitly part of a subtractive color model, and intermediate colors on the wheel (between RYB) are made by mixing primary colors, so I don't think there's any problem in calling the model subtractive. Anyway, this article does indeed have a great deal of nonsense, and should be cleaned up. --jacobolus (t) 23:02, 13 August 2007 (UTC)

Thanks. First if the article wants to describe a colormodel that uses only the three 'primary' colors RYB and it want's to describe it for mixing paints. Then all the colors you can make from these three using paint would be represented by the surface of a triangle (2D) in CIExy colorspace (3D) as you can see the number of colors would be very limited. If the RYB primaries would be of the same lightness all mixtures would have the same lightness. So you would not need white or black for convenience: you would really need them to alter lightness / saturation; to make more colors. But then you are talking about RYBWK space, which is something different than the article. Of course things are only a matter of definition I would not call RYB a colorspace because that would make you think to much of RGB spaces (http://en.wikipedia.org/wiki/RGB_color_spaces) which is not a space that represents all the colors you can make with RGB paint, but all the colors you can make with RGB light and where you can adjust the lightness of RGB independently (of course this would widen your gamut extremely as you can change lightness). So an RGB space is related to monitors. With paint you can't adjust the lightness of the primaries (or you would be adding black - a fourth color). So if you would imagine a RYB-space (as defined above:paint + 3 primaries) to be somewhat like the mentioned RGB 3Dspace for monitors you're verry wrong. You're RYB space is really only a 2D figure (very limited colors). That's why I insist that the word spaces is used for 3D color models. Colorwheels are in fact 1D (a circle of colors). But sometimes colorwheels have the interior filled op (representing all the possible colors one can mix from RYB) so that would be 2D. And please explain me when would you call a certain space additive / subtractive? What would it's properties be? Again to me ther is no such thing.--BartYgor 19:20, 14 August 2007 (UTC)

“Then all the colors you can make from these three using paint would be represented by the surface of a triangle (2D)” – No, this is wrong. The paper starts out white. So by adding paints, you selectively darken sections of the paper. If you have three primaries, which when combined together form a blackish color, then you can form all the colors in a sort of double three-sided pyramid (i.e. point at the white top, point at the blackish bottom, and a triangle in the middle area), which is a three-dimensional gamut. Adding white or black paints just makes a) true black possible, and b) makes mixing the colors much easier/more convenient. --jacobolus (t) 23:19, 14 August 2007 (UTC)
“With paint you can't adjust the lightness of the primaries” – you certainly can adjust the density of paints (for instance by adding white/black paint to them, or thinning them out by adding whatever base the paints use). --jacobolus (t) 23:22, 14 August 2007 (UTC)

At last an intelligent discussion, thank you! Good remarks! But, we first have to agree on what the article is about. Because yes you can adhance the gamut if you would apply tranlucent paint on a white surface (and translucent is very important here; and I did not read in the article that it want's to specifically describe this) but then again your talking about RYBW space in my opinion. Also the blackish color you would form in mixing paint (i'm specifically not talking about putting translucent paints on top of each - becaus in that case if you apply enough yellow in enough layers it well get black too) would be in the centre of the 2D figure (some gray with the same lightness as RYB if these three were of the same lightness, else it would be some mean lightness of the primary lightnesses) adding then more of the one primary would shift you just again a little bit away from that center towards the primary you just added, but you would stay in the 2D surface. And if you were even a bit right you surely can't make white with the primaries! --BartYgor 08:38, 15 August 2007 (UTC)

Adjusting the densitities seems logical. So you would be able to create lighter colors. That would give you a piramid (not a double piramid). And of course you can understand the words "RYB-model" anyway you want. But this is wiki and it is about what is generally understood when talking about "RYB-model". THis is what I would like to have cleared up in the first sentence of the article. Something like "the RYB-model describes all the colors one can make using RYB-pigment and a solvent on a white surface by mixing them or applying them as translucent layers on top of eachother". This is something completely different than "the RYB-model describes all the colors one can make by mixing RYB-paint". So this is paramount: what is generally understood when talking about "RYB-model"? In my opinion when talking about RYB-model they are talking about http://www.digitalanarchy.com/theory/theory_ryb-rgb.html which in fact are 2D colorwheels they are not talking about lightness (in fact there even is a colorwheel picture in the this wiki article). So they are not talkeing about additive/subtractive . So my conclusion is: you are not talking about 3D spaces! You're talkeing about colorwheels and the specifics about this RYB colorwheel (colorwheels are (like colorspace) colormodels) is that hues are more compressed between Y and B than for instance a RGB colorwheel (again i'm talking about RGB colorwheels not the RGB colorspace used for monitors -something completely different). So in my opinion the heading of the article should read RYB colorwheel. RYB-model is not incorrent but like with RGB colormodel you could mean the RGB colorwheel or the RGB colorspace used for monitors. So to avoid confusion: speak in more specifi terms (colowheel, color space) instead of colormodels --BartYgor 08:38, 15 August 2007 (UTC)

Please answer my question about "additive/subtractive color models"? --BartYgor 10:09, 15 August 2007 (UTC)

I don't think you very well understand the gamut of colors which can be created with three paints on white paper. I suggest you try to learn to paint for a week, and come back if you still don't understand. As it is, I feel we're talking in circles. (also note, whoever wrote the digitalanarchy.com link you posted doesn't know what he's talking about) --jacobolus (t) 16:25, 15 August 2007 (UTC)

Maybe. And I will be the first to acknowledge if true. But I did try mixing RGB and it gave me some brown (not dark! - maybe you are mistaking that a less saturated color appearce somewhat darker yet is still is of the same lightness). And how would i tell my printer to print equal amounts of CMY? I'm sorry but i'll believe you when you explain things to me. I'm just talking about basic fysics here. On this page http://www.handprint.com/HP/WCL/color5.html#theorysub it is explained how yellow and magenta is mixed (scroll down a bit you will see the wavelength figure). From the figure it is clear that lightness of the mix is the mean of the lightnesses of the base paints. I'll explain more basic. Imagine a red and green sheet both lettre size. I cut them up in small pieces mix the pieces put close together so no of the table colour gets through. What you are basically saying is just by mixing up the pieces lightness (the power of the light so to say, will change). I say lightness stays the same: the same amount of light is reflected if pieces are mixed or not. THis reasoning seems logical to me. Were do I go wrong? Or maybe you'll say this website is also wrong. The whole world is wrong...?--BartYgor 21:30, 15 August 2007 (UTC)

[edit] CMY and RYB

This article could use some well cited info on the similarities and differences between RYB and CMY, as it is the "limitations and eccentricities" section is just comparing apples and oranges since RGB is additive and cannot be used in the same applications as RYB Velps 19:11, 16 August 2007 (UTC)

Such explanations of differences would IMO be better placed at subtractive color. --jacobolus (t) 00:42, 17 August 2007 (UTC)

[edit] Real Color Wheel

(on file)@bellsouth.net I have found your site by accident mostly. I was looking up color theory, to understand it. I have been an art student for years. I am 56 years old and teach art only recently. I returned to college to learn computer art and added art education to the mix, because I love art I thought I would like to pass that on to high schoolers along with an understanding how to paint, draw and create commercial art and photography. In my last semester of my return year to college, I heard the phrase color theory. I had been taught color absolutes in the sixties, and I attended art schools in the New York City area. I learned the color wheel with red, blue yellow primaries, etc. I had been taught to use black to shade, learning many years later that green made a more lively darkening effect for red than black and so every other color used its compliment. I never could reason why the light, pigment and commercial ink color wheels were not the same. I confused me and I never thought to investigate why they did not work in unison in my mind. I just accepted these contradictions, because I thought it was scientific fact, not theory. Looking up why it is called color "theory", brought me to your site. Thanks for the study, curiosity, and work that brought so much information. I appreciate your corrections for my misunderstandings and faulty perspectives. It will help me greatly with my teaching. Dianne

This email dated 3-16-8 prompted the start of a Wikipedia article as a way to help those who want more information about color and the Real Color Wheel. As of 3-16-8 this site, (http://www.realcolorwheel.com/) is getting 2,935,392 hits per month. http://www.realcolorwheel.com/WikipediaFiles.htg/3-15-8UsageStats.jpg I really don't know where this article should be placed, I thought "Color Theory" but that is crowded with what I call the "old" color theory. Here is my original theory page for the Real Color Wheel. http://www.realcolorwheel.com/colorwheel.htm I'm very new to Wikipedia so this will read as an email. I'm sorry about that, I hope to improve.

New Article: Real Color wheel was first designed in 1995 by Don Jusko, a location artist matching nature's colors to pigments. image = http://www.realcolorwheel.com/final.htg/OriginalRCWpainting1-500.jpg The conventional colorwheel based on the red, yellow and blue and it's subsequent secondary colors was incorrect because it couldn't be used to darken any color using only it's opposite color. They just made a brown hue when mixed equally and never black. The pigment black was used to make shades. image = http://www.realcolorwheel.com/mypigments.htg/makebrowns400.jpg Today we have permanent transparent colors that were never available before 1940, These new colors are primary in that they can mix the red and blue hues that were once considered pure and primary. Newton and all color theorists in the past never saw or recorded the real primary pigments transparent yellow, transparent cyan and transparent magenta, as cyan and magenta can't be seen in the rainbow or prisms.

The RGB computer model colorwheel and CMYK printer colorwheel are alike in that they both use black, less light in the RGB and black ink in the CMYK. The Real Color Wheel is different and colors get dark differently, matching the color element's crystal as the way to get darker mixed colors instead of using black pigment. The yellow to brown crystals show yellow, orange and red darken to brown before black. http://www.realcolorwheel.com/crystal.htm#13,%20CHRYSOBERLE http://www.realcolorwheel.com/crystal.htm#19_RUTILE The cyan to blue crystals show cyan, cobalt blue and blue darken to blue before black. http://www.realcolorwheel.com/crystal.htm#28,%20CALCITE http://www.realcolorwheel.com/crystal.htm#50,%20BERYL Magenta and green go straight to black.

With the Real Color Wheel all pigment color oppositions mix to a neutral black and all light oppositions mix to a neutral white.

I would like feedback as to how to make this article better, if I'm on the right page. Don (talk) 09:03, 17 March 2008 (UTC)

The information you have written here about the “Real Color Wheel” reads like advertising copy, and is also simply wrong, so I for one won’t be creating an article out of it, and if you do try to turn it into an article it will likely be deleted, unless its content is made accurate and cites reliable sources. I am glad you find Wikipedia useful however. Please feel free to stick around and improve the articles you find here whenever you can. (much of the information about color-related topics currently on Wikipedia is full of inaccuracies as well) --jacobolus (t) 10:16, 17 March 2008 (UTC)

Thank you for your prompt reply Jacobolus. Your right, it is putting forth new information, which could be construed as promotion which is advertising. I don't know how else I could present original information. I re-read my copy and can't see anything as wrong. I am currently working on getting University collaboration for Wikipedia. 66.233.93.165 (talk) 07:30, 19 March 2008 (UTC)

As far as I can tell, it's not particularly new information. It is however inaccurate information. If you are looking to better understand the way human color vision works, and how paint mixing fits into that picture, the best online resource I can recommend is http://handprint.com/HP/WCL/wcolor.html. Alternately, there are a number of good books about the subject. But in general, the answer to your question of how to present original research on Wikipedia is simple: don't. See WP:OR for details. --jacobolus (t) 09:04, 19 March 2008 (UTC)

I'm not looking for a disagreement Jacobolus, my color wheel information is original and can't be found anywhere else, that is an indisputable fact. It's not fair to denounce my work as unoriginal or inaccurate with out proof.

"The policy that governs the issue of original research is Wikipedia: No original research (WP:NOR). It says: "Wikipedia does not publish original research or original thought. This includes unpublished facts, arguments, speculation, and ideas; and any unpublished analysis or synthesis of published material that serves to advance a position."

I would like some help to go forward. Currently there are 392 sites linking to http://www.realcolorwheel.com/ 2,935,392 hits per month has to count, something being done right. So what should I do?

2-24-08, In the past thirty days (on file) Northern Arizona University, BYU, Glenn Carleton College, Pittsford Sutherland High School, Imperial College London, University of New South Wales, Kennesaw State University, York College of The City University of New York and Lausanne Collegiate School have requested the Real Color Wheel course or Real Color Wheel. Don (talk) 23:18, 19 March 2008 (UTC)

Get it published in a peer-reviewed scientific journal, mentioned in a major newspaper, or similar, and then come back and make an article about it. --jacobolus (t) 00:36, 20 March 2008 (UTC)