View Full Version : complementary 'color wheel' vs. mixing 'color wheel'

02-06-2014, 09:54 AM
In my quest to better understand color and how to best use and mix it, I stumbled across mixing vs complemetary colors. I think I have a certain understanding about color theory, the discussions about CMYK, RGB, YRMBCG .... etc but one thing I need to understand better.

I am still getting my feet wet with colored pencils and try to fit my wide range of pencils into an understandable and manageable grid.

When I now compare the placement of the colors on a CMYK wheel and the placement of pigments/colours in Bruce McEnvoy's wheel I see quite a 'displacement' especially when I try to figure out the best/most appropriate complementary color. Adding to that I try to see how the visual vs. mixing complements from handprint.com can be incorporated into my brain.

My question is:

Does CMYK represent the visual arrangement on the color wheel or does it more represent the mixing? And by the same token, do the RGB wheels (Itten et al.) fit into the frame of mixing or more visual. Or are these completely different and one has to find out by trial and error with what I have and learn from mistakes?

I desperately hope for enlightenment. Thanks much for considering.

02-06-2014, 10:52 AM
p.s. I received today my copy of Stephen Quiller's book: Color Choices with his color wheel.

02-07-2014, 07:49 AM
Reinhard, I'll try to answer but what "CMYK wheel" are you looking at? Also, when you say "the RGB wheels (Itten et al.)", do you mean RYB?

02-07-2014, 01:29 PM
Dr. Briggs, thanks so much for your kind offer to help.

As for CMYK wheels, I have
1. TheRealColorwheel by Don Jusko
2. http.englisch-verlag.de
3. Caran d'Ache based on Ostwald
4. Joen Wolfrom, assumedly based on Herbert Ives'
5. Küppers Farbensonne (as published and available on the Internet)

They all seem to have in common that YMC is places (and the basis of a 12 segmented wheel) on the positions 1(Y) - 5(M) - 9(C).

As for RYB (sorry for having not mentioned it clearly. I was still too much in my German mind frame).
1. Itten 1(Y) - 5(R- scarlet) - 9(B- cobaltblue greenish).

Whether these are 'mixing wheels' or 'visual wheels' is the reason for my question.

Then I have a printout of Bruce McEnvoy (2009) artist's color wheel - mixing complement hues and from Stephen Quiller 'Quiller Wheel 1989 as published in his book 'Color Choices reprint 2002, Watson-Guptill). Dudeen Color Triangle Dudeen 1977.

Did I forget something? The 'problem' I have is that colour mixing with colored pencils (Faber Castell Albrecht Dürer and Polychromos, and Caran d'Ache Pablo and Luminance) are FOR ME rather difficult the moment I want to mix more than 2 colours (let alone that I am still struggling to find the 'perfect' complement or companion colour). I hope you can imagine the quagmire I am in - maybe should have stayed with graphite :confused:

Thanks again for any advice you might be willing to give me.

02-07-2014, 02:05 PM
Actually, pencil strokes do not intermix all together to a homogeneous mixture as watercolor or oil paint can do.
Artists most often do not mix, they prefers to use pencil which has an exact color they want.
Excluding watercolor pencils.(Lyra, Koh-i-noor Mondeluz watercolor pencils) Adding some water you can intermix pencil strokes to a homogeneous color using soft watercolor brush.

02-07-2014, 03:12 PM
Gigalot, many thanks. I am painfully aware that the mixing with coloured pencils is a rather difficult affair. To some degree it seems to possible, hence my quest for the 'ideal'/acceptable/workable complements. I agree on both points with you. Mixing with water with the watersoluble pencils is definitely a way to go but I am still VERY reluctant to use water in my drawings. Once I will build enough courage to start using my acrylics (Lascaux) I hope to be able to find more answers to my mixing struggles - adding additional ones I am more than sure.

I would gladly use the 'correct' pencil but it seems that I never have the colour I have in mind, hence my mixing adventures.

02-08-2014, 07:37 AM
You can try a very small amount of water with dry, pressed brushes. :)

Pointillist used optical mixing in their paintings rather than physical mixing. At a distance of 3-4 meters colorful strokes merges into one color. This effect can be used in the pencil art.

02-08-2014, 08:17 AM
Many thanks. I think this might be an approach to further follow.

However this brings me back to my original question of visual vs mixing complements since creating colour by a pointilistic style would require the how to best mix knowledge. Bruce McEnvoy's mixing colour wheel (using pigments vs colours) is, as far as I can make out, pretty far away from the usualla available wheels. Plus, in his table of which pigments mix best to create neutral tones, observing mixing curves, makes this all even more complicated for my poor brain. I am not even starting with the topic that coloured pencil manufacturers - with the exception of Caran d'Ache for their Luminace 6901 line - are rather reluctant to absolutely unwilling to share the pigments used in their products, making it another difficult guessing game for a colour beginner like me.

02-08-2014, 11:39 AM
Pigment colors are not an exact. Actually, pigments are a mixture of many crystalline forms, particle size and shapes, doped with various chemical elements. Pigments color are unstable. Pigment mixing is not predictable. Detailed mixing recommendations do not work in practice.
It might be better to use computer programs or special web sites ( http://hexday.com/color/FFB3AA or http://www.color-hex.com/color/e54e1b for example) to determine exact complementary color for existing pencil colors. You can try a primitive color wheel to look some opposite colors:

02-08-2014, 12:43 PM
Many thanks for the great help. The wheel I have saved and will print out to see how it compares to what I have. Using the computer to find the complementary colour I had tried some time ago but somehow 'forgot', but many thhanks for reminding me.

The links are really great and helpful. ColorHex fascinates me especially since they already have included tints and shades. I am afraid that for all the reading and studying and trying to learn I might find no more time to draw. What a luxury problem. :D

02-08-2014, 01:50 PM
When colors align themselves opposite each other on a two-dimensional color wheel by having been plotted in their locations by performing a scientific color analysis and placement with a color-measuring instrument, those colors, when mixed, will create neutral.

I don't know whether those who subscribe to the "mixing complements/visual complements" theory, or concept would care to use one of those terms to describe the reason that neutral is produced by such a mix, but it's been working well for me for over 45 years.:D

But, please keep in mind, such a placement on a color wheel of true, scientific, complementary colors cannot be accomplished [accurately] by guesswork, intuition, or visual approximation or comparison.

02-08-2014, 02:36 PM
William, thanks for your thoughts and help. The first time I read about 'visual' vs 'mixing' was Bruce McEvoy's in handprint


Can I interpret your comment in the direction that using the computer to invert the colour might not be a too bad approach. I know we had the discussion about colour wheels and if I recall correctly, you wrote that Don Jusko's RealColorWheel has the colours positioned correctly.

I see that I need so much more to learn about colour. Let me warn you, I'll be in Phoenix in June and might use that chance to try to impose myself on you. Regina and I lived there 71-73 and we will want to rehaunt old places.

02-08-2014, 06:26 PM
An important thing to keep in mind is that visual complements (i.e. visual mixing or averaging) can be established by their color alone. But ascertaining pigment mixing complements cannot. Two red pigments might look the same, and one might be a perfect mixing complement with PG7, but the other might mix a brownish or blueish black instead. Because they are made from slightly different spectral profiles (the phenomenon of metamerism).

You can make an accurate visual color wheel, but in making a subtractive (paint mixing) wheel, the complementary relationships deviate somewhat from a visual wheel...often in slightly weird, unpredictable ways. The difference between dealing with color in a general or abstract sense, and actual physical pigments.

02-08-2014, 07:03 PM
Patrick, thanks so much. If I understand your helpful comment correctly it means that I have to sit down and do the colour/mixing studies myself with the pencils I have because of the matter itself and the very specific situation with the media I am employing/learning/struggle with/want to break and throw into the trash more than once.

I was afraid it came down to that but obviously can't be helped, especially with coloured pencils where nobody knows what they comprise and how they react in mixing.

02-08-2014, 11:45 PM

The difference between additive and paint-mixing complements has "been working well" for science for a century and a half, since it was investigated and explained by Helmholtz, Bezold, Rood and others. The difference is most pronounced near the yellow - blue axis, where for example modern ultramarine and lemon yellow paint when mixed optically (that is, by additive-averaging) make a neutral, but the exact same paints when mixed physically make a distinct green. So you can use the colours placed opposite each other in this region as a diagnostic for whether a hue circle is basically showing additive or paint-mixing complements. Despite theoretical objections that are sometimes overemphasized, a hue circle based on tested paint-mixing pairs would have to show complements more or less like the following:

ultramarine blue: cadmium orange
ultramarine violet (bluish kinds such as Blockx): cadmium yellow
dioxazine violet: sap/hooker's green

Many traditional (rot-gelb-blau) colour wheels place opposite yellow a colour more purplish than the bluish kinds of ultramarine violet; real paints of these purple hues would not be an exact paint-mixing or additive complement of a middle yellow paint.

I didn't know that Caran d'Ache has a colour wheel based on Ostwald. Is it available online, or would you be able to post a picture of it? If it's based on Ostwald it should show additive complements, but the other four wheels you list may well be different. Is the wheel from http.englisch-verlag.de that one you posted that had two of the twelve colours the same hue?



02-10-2014, 12:48 AM
Actually, pencil strokes do not intermix all together to a homogeneous mixture as watercolor or oil paint can do.
Artists most often do not mix, they prefers to use pencil which has an exact color they want.
Excluding watercolor pencils.(Lyra, Koh-i-noor Mondeluz watercolor pencils) Adding some water you can intermix pencil strokes to a homogeneous color using soft watercolor brush.

This underline statement is not entirely accurate. Sure, colored pencil artist do not mix colors like in other media, but they can layer and blend two or more colors to achieve a desired color. Using my preferred Prismacolor brand of colored pencils, I can layer and blend Lemon Yellow and Grass Green to achieve a high saturated mid green of the CMY color wheel or any intermediate green hue between those two colored pencils dependent on how lightly or heavily I layer one color unto the other. Here is a link to a WIP I did in the Colored Pencil Forum that demonstrates this layering and blending to achieve the desired color:


As to the color wheel, I had constructed a 24 hue CMY color wheel using pigments currently available in the acrylic medium using as my reference the Color Company's CMY color wheel and a virtual wheel I had constructed in the CMYK color space in the computer program DrawPlus. I use this wheel not intermix the various hues physically but to visually create color chords to use in my pictorial designs. Though this wheel in acrylic paint may not be 100% scientifically accurate, visually it is so for me. The images below show the constructed CMY color wheel and a couple designs based on it. My point & shoot digital camera didn't reproduce the colors 100% accurately, but you'll get the idea.

CMY Color Wheel

Radiance3 42''x42''

Color On Three Planes 39''x45''

02-14-2014, 06:48 PM
There is really no such thing as a truly accurate color wheel for mixing. Not only can this not be accomplished through guesswork or intuition, it also can't be accomplished through any kind of scientific measurement or analysis!

In the end, "color" is determined in the human brain, and it doesn't really correspond perfectly to directly measurable properties of light. Moreover, real world pigments behave inconsistently; seldom do two colors directly opposite on a color wheel produce a perfect neutral, more often there is some bend in one direction or another due to real world properties of pigments.

Still it is useful to start with a conceptual color wheel that ignores these problems. The one that makes the most sense to me is to use CYM as the "ideal" primary colors for pigments, with RGB (the primary colors our eyes see in light) as their compliments. It is important to understand here that there is no difference between a CYMK and an RBG color wheel. The wheel is the same, it's just a question of which group is considered primary and which secondary (depending on whether you are dealing with mixing pigment or light). There is some difference in traditional RBY wheels, but they should be considered only slightly in error, mostly due to limitations on available pigments (only fairly recently has a good magenta been available for instance).

On such an "ideal" wheel:

An ideal Yellow pigment would be one which absorbs only "blue" light, and thus is complimentary (and directly opposite) to blue.
An ideal Cyan pigment would be one which absorbs only "red" light and thus is complimentary (and directly opposite) to red.
An ideal Magenta pigment would be one which absorbs only "green" light and thus is complimentary (and directly opposite) to green.

In the end though, any such wheel will have different minor flaws when it comes to real world mixing. But how often do you really need to mix a perfect grey anyway?

More useful for the artist though might be a visual wheel, which is more directly based on our visual perceptions. And a wheel such as the above also works quite well as a visual color wheel. Thus it is useful for looking at which colors are optimal for producing the most pleasing contrasts in your work, or for using as complimentary colors in shadow areas, etc.

For mixing though, you can really use ANY of these as a rough guide, and then work out the details based on the results you get with whatever materials you are actually using. However, some prefer a "mixing wheel" which makes some modifications to the above which reflect the behaiour of a least some of the most common and traditional pigments.

In a mixing wheel, normally:

The "blue" opposite yellow leans a little more toward blue violet (such as an ultramarine violet).
The "red" color opposite cyan leans a little more toward orange (such as pyrrole orange, or even burnt sienna).
The "green" opposite magenta leans more toward yellow green, while a bluer-green ends up more opposite to red.

But maybe the best thing to do is take the materials you plan on using and make your own color wheel. In the process you might find out what works best for the pigments you are actually using.

02-24-2014, 12:54 PM
First, and most of all, I have to apologize for my absence. My heart was 'acting up' and an inflamed wisdom tooth thought it was time to make its appearance. As you might expect, no fun whatsoever. Only as an explanation and no whining.

Dr. Briggs, many thanks for your kind help. Unfortunately, Caran d'Ache soes not seem to post it on their website, but I have copied it out of a brochure of theirs. I hope it can be read. For my personal need it seems to be rather useful. If it stands up to any scientific probing, I don't know. Sorry that the explanation is only in French (I have eliminated German to be able to make the image larger and more readable).

What I like additionally with Caran d'Ache (Pablos, Supracolour and Luminance) they share the pigments used in these pencils making it easier to understand where the colours might fit in the colour wheel. Their numbering makes for another 'goodie'.

DeBro, thanks so much for your kind explanation and the posting of your most helpful colour wheel. I have been following your link and am most impressed.

I think both of you, Gigalot and you are correct with the aspect of 'mixing'. I think that mixing in its real meaning is rather difficult in coloured pencils. We in coloured pencils have to rely more on layering and hence the impression of a new colour. And this is where I have my problems, especially when looking for a complementary colour to neutralize. If I am not absolutely lucky in finding serendipisously the correct complement, the results are more than iffy. Added to that is the problem that coloured pencils rarely disclose the pigment mixture and then strange things can happen.

acerimusdux, so many thanks. Seems that I really have to do my own colour wheel with what I have. I was afraid it came down to that. But sometimes one has to bite the bullet.

To help me in that quest (finding the holy grail seems to be child's play in comparison), I have purchased the colour wheel as published by Joen Wolfrom and her 'Ultimate 3-in-1 Color Tool' to help me. I am convinced that neither the wheel nor the colour cards stand up to any kind of scientific probing. What I find most helpful is that, the wheel and the cards are based on CMYK, they both have tints, shades and tones making it easier to put the colours I have into a meaningful order. At least this is what I am hoping for.

03-04-2014, 09:55 AM
Thanks very much for the scan, Reinhard, I hope you're feeling better.


03-04-2014, 11:11 AM
My pleasure. Any thoughts on this wheel? As much as one can see from a scan that is.

03-07-2014, 07:46 AM
If it accurately gives positions in the Ostwald system then it should show true additive (or "visual") complements. And you have the Quiller book for subtractive complements.

As I explain on my site, mixing colourants is a combination of subtractive and additive averaging mixing. It's likely that mixing of coloured pencils is well towards additive averaging. Have you tried overlaying pencils that are opposite on the Caran d'Ache wheel?

03-07-2014, 01:48 PM
Dr. Briggs, thanks for the info. As of yet I have not really tested, but will do. Unfortunately I'll be off for 2 weeks. Will do then.