PDA

View Full Version : Blue and Yellow Don't Make Green


Fmalo
02-08-2002, 06:24 PM
I am only now beginning to realize what a color dummy I have been. I have resisted using the proven formulas(ae?). Sure, I know the basic color wheel & have done work using only the 'big 3' colors, but the subtle nuances have escaped me until now.

I will follow the suggestion of one of your experts here who PM'd me suggesting that I read the posts carefully & learn from them (you know who you are & yer brogue gives you away).

Anyway, I was at a Demo today & the painter used a circular palette which came directly from the author of 'blue 7 yellow don't make green' & he explained briefly the layout with cool to warm hues & demo'd the results of 'close colors' opposed to 'distant colors'

I think I'll try to get a hold of this book & study the principals so my foreground differs from mid & back.

Keep painting,

Frank

islandwoman
02-09-2002, 06:21 AM
Hi Fmalo,

Einion put me (and many others, I'm sure) onto "Blue & Yellow Don't Make Green. Waiting anxiously for it's arrival. New paper edition being released & sold on Amazon. They gave me a ship date of 2/4/02. $18.95, I think.

djstar
02-11-2002, 01:44 AM
I get totally overwhelmed by all the numbers and formulae but I think I got the principle in the first few pages.
I am anti-tecnical. I think what it did for me was comfort me.
I understand why I simply don't get the same purples when I mix cad red and ultramarine as cobalt and alizarine.
BUT knowing why makes me more mindful of what a great choice that would be for a very gray suggestion of a violet.
A realization hit me this evening. Most of my work is from the colors left on my palette from my last paining. WHICH indicates the opposite of my palette. It would be the colors I DON'T use in the painting that are left behind.
It also seems to explain why my color sense is all over the map.
I think I will clean out my paint box and see if I can come up with a couple of standard palettes so I make more picture that look alike as I go.
In a week or two, I may have a serious answer here...
dj*

Mario
02-11-2002, 10:13 AM
OK, DJ I'll remind yu in case yu get involved in too many projects meanwhile..:cool:

cjorgensen
01-30-2009, 12:25 PM
I have not read the book, so I don't know exactly what you are talking about, but from my own experience, there are "warms" and "cools" of each primary.

For example: making a green with a, a redish-blue and as opposed to a yellowish-blue. Of course, you are going to get different greens with these blues, because ONE already has some red it in. If you mix blue with a redish-yellow or a greenish-yellow, you will also get more different greens. So if you want a pure, bright green (sun shining through new leaves), mixed the yellow-blue (cerulean) with the blue-yellow (cad yellow light). If you mix red-blue (ultramarine) with red-yellow (yellow ochre), you will get a dull green (eucalyptus leaves). The same for mixing oranges and purples too.

Is this what you guys are all talking about? :)

lenepveu
01-30-2009, 06:37 PM
Well sort of. Blue and Yellow make green of course. The farther apart a hue is on the color wheel, the duller the mixture. The duller the principle colors, the duller the mixture. The more divergent value-wise a color is, in general, the duller the mixture. This is not true for dark blues and purples, since, when mixed with white, they gain chroma.

Don't be worried about formulas. The ones that work do so for realistic reasons. The ones that don't are in error. Most greens don't need elaborate additions but require a few pigments mixed to advantage. Yellow Ochre, Viridian, Burnt Umber, black and white will mix almost all that is needed. The addition of Cadmium Yellow and Phthalo Green will enliven the chroma. C'est tout.

wal_t
01-31-2009, 04:06 PM
I read - and actually studied - that book quite some years ago and found it extremely helpfull.

The basic idea for the title is based on the following "imagine that there were real primary pigments like primary blue and primary yellow, mixing these would result in black, not green .... "

.... and then an explanation followed why you will get actually a greenish result when you mix these two. Clearer greens when you mix the yellow and blues that contain the highest traces of green (the cooler ones) and duller when you mix the yellow and blues that contain the least traces of green (the warmer ones). It worked for me, I loved that book and ever since I paint mostly with 6 colors (warm and cool red, yellow and blue) and white ....

Regards, Walter

lenepveu
01-31-2009, 06:40 PM
I am glad it worked for you, but the theory is flawed, or at best, incomplete.

What do the terms, for example, clear, trace, cool and warm mean. And which green? People can't even agree on a primary, unique green. These are all things one has to consider if color is to be mastered

mrking
01-31-2009, 07:12 PM
I am glad it worked for you, but the theory is flawed, or at best, incomplete.

What do the terms, for example, clear, trace, cool and warm mean. And which green? People can't even agree on a primary, unique green. These are all things one has to consider if color is to be mastered

Flawed, incomplete? Hmmm. If you read the book you'd know the author doesn't use the terms 'warm or cool'. It has to be the most scientific theory I've read to date.

He defines colour by the spectrum the colour so there is no doubt what a specific pigment carries in terms of its colours.

He gives recommendations to what to use for each colour, but of course you are free to find your own 'green yellow' or orange red' etc.

What this does give you is an absolute way to determine what colour you want.

I can easily find coloured greys, mid greens, bright greens, coloured blacks, etc. All from reading the book and making swatches over the last three weeks.

wal_t
02-01-2009, 04:30 AM
Yes the terms "traces" and so I used myself so I didn't give a scientific accurate reply I guess :-). There are also some nice notes from Wittgenstein on color but that doesn't help me either in painting ....

What I liked about the Wilcox book is the practical possibilities. I believe that after studying it I am better able to judge in advance if a get a bright, not-so-bright or more sparkling green when I use certain blues and yellows (regardless of my experience with these colors) and quite a few more practical advices are covered in that book.

I just liked it but also realize that there is much more to say on color then described in that practical guide. Where is that color anyway, on the object, in the light rays, in the eye, in the brain or in none of these :-)

Regards, Walter

lenepveu
02-01-2009, 11:48 AM
I have read the book and found it only somewhat useful. The reason it is not so good is because it is imprecise, just like the language you are using: sparkling greens, mid greens, colored greys, colored blacks, bright and no-bright greens. I only have a general idea of what you mean. All color theories have terms such as hue, value and chroma that are used specifically.
Any mixing system must address these elements. One has a low value green, a certain hue, such yellow-green or a high chroma green; sparkling suggests a flavor. Warm and cool are impressions and relative.

mrking
02-01-2009, 12:12 PM
I disagree. You have to be able to describe a colour to be able to mix it.

I have issue with what you are saying as you act like your full of knowledge but only repeat yourself over and over with no explanation of how to address colour. Step up and show us.

How do you describe colour?

Oh and nowhere did I say 'sparkling greens'. A mid green is what it says so I have no idea why you have issue with that.

Of course it is all relative, it is colour. What is a bright green when mixed with cad yellow and cerulean blue might look dull compared to cad yellow and phthalo blue. It will all depend on the pigments you use. If you understand what pigments do what, as this book tells you, then you can get what you want. Easy.

mrking
02-01-2009, 03:04 PM
oops, I meant to say HANSA Yellow and Cerulean Blue and HANSA Yellow and Phthalo blue.

dcorc
02-01-2009, 03:17 PM
You have to be able to describe a colour to be able to mix it.

Indeed - but there are better, more precise ways of describing it than "mid-green" or "warm yellow".

How do you describe colour?

Munsell notation. Read the numerous threads on the site that have discussed this.

Dave

lenepveu
02-01-2009, 05:26 PM
not-so-bright or more sparkling green

Sparkling Green. No? I am just not sure what you mean. There are countless color notation systems that are precise. Blue and Yellow is a good book for a beginning, but there are some flaws and the approach is less than precise.

dcorc
02-02-2009, 01:16 AM
I believe that after studying it I am better able to judge in advance if a get a bright, not-so-bright or more sparkling green

wal_t, not mrking. From post #10, above.

Just to prevent confusion as to who said what :)

Einion
02-02-2009, 08:13 AM
I have not read the book, so I don't know exactly what you are talking about, but from my own experience, there are "warms" and "cools" of each primary.
The idea of 'warm' and 'cool' primaries is just a way of referring to the hues of each - green-yellow ('cool' yellow), orange-yellow ('warm' yellow) and so forth. But there are so many traps for the unwary in the idea of 'colour temperature' that I think it's something that should never be used as the bedrock for a general understanding of and use of colour. There are lots of prior threads, here and in other forums on the site, that touch on this and give specific examples to highlight the problems.

For example: making a green with a, a redish-blue and as opposed to a yellowish-blue. Of course, you are going to get different greens with these blues, because ONE already has some red it in.
This is sorta the point although the idea of the colours reflected by certain paints is only a beginning. Apart from the simplification of the idea (not a bad thing per se) an important thing overlooked in the book is the part that specific pigments play in colour-mixing interactions. Very important for people to learn that two paints that appear to be identical in colour can mix quite differently with one other paint. Sometimes the difference is slight, other times much more pronounced; no writing about colour mixing should overlook this or downplay it.


I read - and actually studied - that book quite some years ago and found it extremely helpfull.
Me too.

I'm less enamoured of it now because even after updating it doesn't cover certain important things but it's a great simple way of improving the understanding of colour-mixing outcomes for people who are having trouble in this area, which frankly a great many books before this (and since!) do not explain in a useful, wide-ranging way.


Flawed, incomplete? Hmmm. If you read the book you'd know the author doesn't use the terms 'warm or cool'. It has to be the most scientific theory I've read to date.
There are many much more science-based things to be read on colour interaction right in this forum for a start :)

He defines colour by the spectrum the colour so there is no doubt what a specific pigment carries in terms of its colours.
Yes, but wrongly. Simplification is fine, but there's no getting around the fact that in a few cases it's just an easily-digested idea of the reflected colours, nothing like the genuine spectrum (which in all fairness are sometimes very hard to interpret usefully).

I can point you to one or two prior threads on the book and the underlying principle if you'd like to see more on this Michael.

I disagree. You have to be able to describe a colour to be able to mix it.
I agree it helps, but it's not necessary. This is an interesting related point because many artists are not good at talking/writing about colour while still being very good at using it; we've seen that here over the years and there's plenty of historical precedent for it too.

Of course it is all relative, it is colour.
That's a cornerstone issue - colour does not have to be relative at all. It is key when using one of the woolliest colour-thinking systems (not one you used) but any colour can be considered in isolation since each is unique.


I believe that after studying it I am better able to judge in advance if a get a bright, not-so-bright or more sparkling green
wal_t, not mrking. From post #10, above.

Just to prevent confusion as to who said what :)
http://www.wetcanvas.com/Community/images/20-Aug-2003/3842-thumbsup.gif

Hence the value of using the quote system :)

Einion

mrking
02-02-2009, 05:07 PM
Munsell notation. Read the numerous threads on the site that have discussed this.

Dave
My only issue with Munsell is the extensive use of black. We all know, or should know, that black doesn't mix well with many colours. Yellow for example. When adding black you do not get a greyed yellow, but a dark green instead. Now having said that I have ever mixed a black with an orange yellow to see what happens. Hmmmm

Maybe you can enlighten me on why the munsell system is as good as everyone says it is. Yes, I have read the numerous posts here on Munsell but I still remain unconvinced it's a good system for painting although many people are using it successfully.

Could be I prefer a more direct understanding of colour mixing and that is why my brain has latched onto what is say in the B&YDMG book.

mrking
02-02-2009, 05:09 PM
Indeed - but there are better, more precise ways of describing it than "mid-green" or "warm yellow".

Dave
Like what?

mrking
02-02-2009, 05:15 PM
I can point you to one or two prior threads on the book and the underlying principle if you'd like to see more on this Michael.

Einion
Einion, please do. I'd like to read more.



That's a cornerstone issue - colour does not have to be relative at all. It is key when using one of the woolliest colour-thinking systems (not one you used) but any colour can be considered in isolation since each is unique.

Einion I agree but disagree. Sure a colour can be isolated and given defined parameters (as when you get paint mixed to match a swatch on your couch) but I don't think an artists would take the time to analyze a particular green they are seeing as an isolated colour when painting plein air, then try to recreate it on their palette. I would think the artists would look at the green, compare it to the surrounding colour(s), decide if it is darker, lighter, bluer, yellower (funny word), duller or more chromatic, etc then mix and paint it in depending on that info.

wal_t
02-02-2009, 05:42 PM
Maybe a bit off topic, please forgive me if that is the case, but I once read a nice story from a great philosofer asking himself the question where the brown color is coming from when he observed a certain table with a certain form/shape under certain light conditions from a certain position and if he will see the same as someone else in the room .....

Even if you can answer that it will not help you much in color mixing :-), maybe it's like drawing, you can read all the books there are - and yes, some are helpful of course to get you started and offering practical advice - but the real experience only comes from doing it.

But maybe I am just lazy as well and should study color systems/notations a bit more.

What I also found a nice point from the book (wish I had that book now but I got it from the library) is the lack of color memory in people. The example when you try to buy additional wallpaper (assuming you don't know the serial/batch number) a in a matching color of what you already have was a recognizable one.

Regards, Walter

Einion
02-02-2009, 06:56 PM
My only issue with Munsell is the extensive use of black. We all know, or should know, that black doesn't mix well with many colours. Yellow for example.
Ah, "mixes well" is a qualitative judgement. There's nothing inherently bad about the dull greens you get when you mix yellow paint with black paint. Sure, you shouldn't use just black to recreate typical yellow halftones - the greyed yellows you refer to - but equally there are times when those dull greens might be the colour that's needed, something that too often gets brushed under the carpet by those of the "black kills all colour" persuasion.

As far as the use of black as a general-purpose colour goes, simply put if you add it to something and the result isn't what's needed or desired then you either tweak the mixture or go another route entirely. Neither of those is actually a condemnation of black when you come down to it :)

Now having said that I have ever mixed a black with an orange yellow to see what happens. Hmmmm
Usually you do still get green. You can even get green, or what appears to be green (very dull yellows), when mixing black with some orange paints.

Could be I prefer a more direct understanding of colour mixing and that is why my brain has latched onto what is say in the B&YDMG book.
Nothing wrong with that.

We can take a bit of this and a bit of that too. I'm much like this with regards to Munsell notation. I like what the system has provided to us in terms of pinpointing a colour, to accurately gauge relative properties of one paint v. another, and more basically the value and chroma side of things. But partly because I don't like the hue designations I don't think purely in Munsell terms.

BTW, I don't believe that thinking in Munsell terms is actually less direct - once you immerse yourself in it I'm sure it's not that difficult for it to become as automatic as the more-common "ah, slightly dull yellow-green" and things along those lines; like any new understanding, especially if it's complex, mostly you just need to put the time in.

On to the prior threads. Here is the major one, Split-primary v. secondary palettes (http://www.wetcanvas.com/forums/showthread.php?t=285690). I do have to mention, it jumps around a bit on focus but that's inevitable given there are 240 replies :eek: But Michael Wilcox is actually involved in the discussion so on that basis alone it's worth the effort. Also worth a peek is this thread (http://www.wetcanvas.com/forums/showthread.php?t=270630) on limited palettes that he also contributed to.

In relation to a point I raised above, wanted to repeat this part of a post from another thread:
in my view, "prediction" means you can describe in advance an event you have not experienced before and with a useful degree of accuracy. "color theory" simply will not allow you to predict what color will occur from the mixture of two paints you have never used before, or used in a specific proportion...
I just wanted to highlight this part because it's so important. This is something I've stressed a number of times recently in different contexts: a colour's masstone appearance tells you very little about how it will mix (except in very general terms) showing the futility of mapping paint colours to a wheel and saying "this is how they will mix".

I also want to take the opportunity to link to some general observations I made regarding Munsell's strengths and weaknesses in this post (http://www.wetcanvas.com/forums/showpost.php?p=5479808&postcount=6) (also relates to predictive ability).

Einion

lenepveu
02-03-2009, 02:12 AM
does knowledge of a three-dimensional model help you paint? It can, but not knowing about it doesn't exactly handicap the artist.
Do the Munsell numbers help us to locate, ahead of time, mixing complements? No.
Can they help us to predict the value of a mix of paint A and paint B? No.
Can they help us to predict the hue of a mix of paint A and paint B? No.
Can they help us to predict the chroma of a mix of paint A and paint B? No.
Do they show you which paints are strong mixers and which are not? No.

You might revisit these points. I would also suggest that you are equivocating, at least a bit. Painting is a complex art of which color is only one aspect, albeit usually an important one. Predictions, moreover, are just that, guesses, based on the likelihood of something. If one has the Munsell notation of two paints, it is highly likely that the combination will be between them. This is certainly the case with colors near each other on the color wheel. The closer the colors, the more accurate the prediction. It is harder to guess the outcome of paints far apart on the color wheel, but not impossible. A red mixed with white will not be a blue, for example.

The only statement which is true, unequivocally, is the last. Tinting strength cannot be predicted with Munsell.

sidbledsoe
02-03-2009, 06:49 AM
Only one question about this, does munsell predict that blue and yellow don't make green. If yes :clap: if not then :(
(hint: sometimes I ask questions that I know the answer to make a point)

dcorc
02-03-2009, 06:56 AM
I bet munsell says that blue and yellow make green.

Munsell isn't about paint-mixing. Munsell is about accurate colour identification.

(hint: most people think the problem is paint-mixing. Its not - the real problem is accurately identifying what colour you ought to be mixing. :p )

sidbledsoe
02-03-2009, 07:11 AM
Good answer Dave, as I understand it from you guys, munsell makes no predictions. I am guessing that it would be like looking at a ruler to try to determine the square root of a number like 13, attempting to use it for that is up to the operator. The absence of measuring the tinting strength alone would make it fully unpredictable. Then try on three or four colors mixed together, and the ratios of each, and one is terre verte and one is pthalo blue. In my painting, subtle differences in colors are fundamental, I use the most sophisticated color measuring instrument ever devised, my eye. Good luck my friend!

Einion
02-03-2009, 08:25 AM
You might revisit these points.
I'm happy with them.

The only statement which is true, unequivocally, is the last. Tinting strength cannot be predicted with Munsell.
If only that were true. Wiggling and jostling with specifics aside, one other statement is also absolute.

I would also suggest that you are equivocating, at least a bit.
I was not being ambiguous and I was certainly not avoiding taking a position or intending to mislead so the 'suggestion' that I was is specious.

Painting is a complex art of which color is only one aspect, albeit usually an important one.
Of course. I don't see that it's relevant though given the issue is about concrete mixing issues and predictive value.

If one has the Munsell notation of two paints, it is highly likely that the combination will be between them.
In most cases that's not correct, not if you mean between in the way I think is the natural inference. The small mixing booklet published by Liquitex covers this and a few related issues in their practical examples.

This is certainly the case with colors near each other on the color wheel. The closer the colors, the more accurate the prediction.
Obviously. Most people with a reasonable amount of experience in this area would not expect - correctly - a mixing result between two paints that are close together* that deviates strongly from the straight-line path between them; the important point is that mixes can deviate when they're further apart and by how much.

The reason it's important is that most have palettes made up of colours that are reasonably far apart in almost all cases, plus a lot of people don't realise that a mixture can be off in value, chroma or hue, or a combination of all three. Now the Book Of Color would of course be an excellent tool for identifying in exactly what way a mix is off from the hoped-for outcome, but it doesn't predict it**.

*Either mixes or single-pigment colours.

It is harder to guess the outcome of paints far apart on the color wheel, but not impossible. A red mixed with white will not be a blue, for example.
Who's equivocating now? There is NO predictive value in the Munsell model for tints of two different reds of the 'same colour', sometimes even when made from the same pigment!** It would only take about five seconds to prove this as there only needs to be a single example.

**This is of course a detail thing, sometimes it'll be off by only a single step vertically or horizontally, but when we're using a model of this type then its own subdivisions are the minimum standard :cool:

Einion

Einion
02-03-2009, 08:26 AM
I bet munsell says that blue and yellow make green.
This isn't quite as silly a question as it looks on the face of it. The bottom line is no, it doesn't really; but then it's not intended for that purpose. It's worth mentioning that no colour wheel/model actually fulfils this purpose, since it's technically impossible (despite any claims to the contrary from their creators).


(hint: most people think the problem is paint-mixing. Its not - the real problem is accurately identifying what colour you ought to be mixing. :p )
Well, I've long argued it's both in concert. Which is the harder one, that's an interesting question :)

Einion

dcorc
02-03-2009, 08:44 AM
In my painting, subtle differences in colors are fundamental, I use the most sophisticated color measuring instrument ever devised, my eye. Good luck my friend!

And what do you think we are using if we have the Munsell book?

The trouble with your eye is that its not a sophisticated measuring instrument. Its a sophisticated comparator, which is not quite the same thing. Furthermore, it automatically adjusts for lighting intensity and for white balance in order to try to optimise and normalise the image for whatever you are currently focussing on. These dynamic adjustments are one of the things that make accurate colour identification difficult.

Munsell provides a calibration set.

dcorc
02-03-2009, 08:49 AM
My only issue with Munsell is the extensive use of black.
What "extensive use of black"?

Munsell is a colour notation system, a way of describing colour accurately.

We all know, or should know, that black doesn't mix well with many colours. Yellow for example. When adding black you do not get a greyed yellow, but a dark green instead.
No, what you know is that mixing yellow with black doesn't fit your preconception of how it should mix. Have you considered the possibility that it might be your preconception that's wrong?

http://www.wetcanvas.com/Community/images/03-Feb-2009/30792-olivegreen.jpg

Look here, nothing to do with paint: select yellow in photoshop, slide the value down - what colour do you get? "Olive green"


there are better, more precise ways of describing it than "mid-green" or "warm yellow".Like what?
Like 2.5G 5/10 or 10YR 8/12 - both of which would allow you to match pretty accurately the colours I have in front of me here.

dcorc
02-03-2009, 08:58 AM
(hint: most people think the problem is paint-mixing. Its not - the real problem is accurately identifying what colour you ought to be mixing. )
Well, I've long argued it's both in concert. Which is the harder one, that's an interesting question

Wetcanvas has enormous numbers of images posted aspiring to realism where the works are less successful than they could be because colours are off. I doubt many of these people were putting down colours different from the ones they intended to mix. The problem is more that their intended colours turned out to be wrong. This is an insight gained from using Munsell (and indeed also from approaches such as spot-sampling digitally and in photorealism).

sidbledsoe
02-03-2009, 09:25 AM
Dave I am aware of the sensitivities of munsell proponents/opponents and I kinda expected to have to do some damage control after I wrote that, people do seem to hinge on my every word here. I was talking about color mixing, I didn't mean to imply that munsell was in any way a flawed or inferior measuring tool. I have no problem with munsell at all and value it for what it is. Yes, I think we use our eyes to look at a munsell chart but I think a better chart for predicting color mixing is made from the pigments you use in real life ala Richard Schmid;s advice.
A comparator is what I was talking about. By the eye I mean to include the brain which it is connected to and with which it works with and not without. By sophisticated I mean a complexity that no other device can achieve, such as, but not limited to, perceiving something as nebulous as warmth or coolness. That is in fact a sophisticated measuring instrument that measures parameters and dimensions of color as nothing else can, though "measuring" was probably not the best choice of words there, perception device may have worked better. I preceded that comment with the caveat "in my painting" . I am moving away from classic realism, more towards impressionism but recently doing some expressionist type painting. I am fully aware that there are instruments that can measure color precisely. Using munsell charts for the purpose of predicting color mixing is what I found to be flawed as both Einion and myself have clearly stated.

Einion
02-03-2009, 09:30 AM
Wetcanvas has enormous numbers of images posted aspiring to realism where the works are less successful than they could be because colours are off.
I agree this is a common problem, one of the cornerstone issues that separates amateurish/unskilled/less-good/pretty good work from better output (the other two key issues being good drawing and composition I think).

I doubt many of these people were putting down colours different from the ones they intended to mix.
Honestly, I think they are in a lot of cases.

The problem is more that their intended colours turned out to be wrong. This is an insight gained from using Munsell (and indeed also from approaches such as spot-sampling digitally and in photorealism).
I know what you mean and I agree basically. At the very least the ability to get past the commonplace simultaneous-contrast illusions is a huge boon; and we know just how off some of those make colour judgments!

But IME lots of people also have the problem of not being able to mix on-target, as a secondary issue - being unable to identify how a mix is off, to compensate appropriately. I remember exactly what this was like myself and I come up against the problem repeatedly when trying to help. But colour mixing, as you know, isn't really that hard when you get down to it despite how it seems when one isn't good at it :) I think we've talked about this before, that often mixes are settled on, they're "good enough", because they can't seem to get closer.

Einion

dcorc
02-03-2009, 10:05 AM
Sid, the point that I am making, the one I would really like you and others to take away from this discussion - is that correctly identifying a colour in a scene before you is much more difficult than most people believe.

While being able to mix a desired colour is also an issue, because paints do not generally mix in straight-lines through colourspace, I am suggesting that most people worry about this, while largely neglecting the issue which in fact is more contributary to producing paintings which fail the "realism" test.

Namely, that much of the time, the colour that people are trying to mix isn't even the correct one in the first place.

Now you can either try hitting colours exactly (which can be done in certain well-controlled studio-based situations, which is what's done with the Carder approach), or you can use transpositions, where you maintain the relationships between colours (for example, you may want to compress the range of values, to within what paint will handle). But a common problem (aside from the other drawing and composition issues Einion mentions), is that the colours and the relationships between them are not hit correctly. People learn about the need to get value relationships right, but we also need to get chroma relationships right and hue relationships right (and of these three, hue is, somewhat paradoxically - since hue is the attribute of colour most people think of first - hue is the least important).

I'll say it again - concentrate efforts on correctly identifying a colour in a scene before you - it's much more difficult than most people believe, and getting it right first is the necessary prerequisite before starting to mix coloured paint.

Dave

sidbledsoe
02-03-2009, 10:41 AM
Dave, I agree with that and it involves a lot more than merely predicting how colors will mix. It addresses arriving at the right color with regard to painting and I agree that using munsell or carder methods are valid tools to employ.

Trivia: It is estimated that the human eye/brain can detect about 7-10 million different colors. Different individuals to some degree, detect them differently, thus using a reference can involve variations between other individuals perceptions also.

mrking
02-03-2009, 11:08 AM
What "extensive use of black"?

Munsell is a colour notation system, a way of describing colour accurately.



No, what you know is that mixing yellow with black doesn't fit your preconception of how it should mix. Have you considered the possibility that it might be your preconception that's wrong?

http://www.wetcanvas.com/Community/images/03-Feb-2009/30792-olivegreen.jpg

Look here, nothing to do with paint: select yellow in photoshop, slide the value down - what colour do you get? "Olive green"

Exactly, its an olive green, not a dull or grey yellow


Like 2.5G 5/10 or 10YR 8/12
You'll need to expand on that further please. Is it 2.5 parts G then 5/10 of what? Just need ot understand what that greek means.

dcorc
02-03-2009, 11:34 AM
Exactly, its an olive green, not a dull or grey yellow
Which is what Munsell gives you, what you get in photoshop, and what you get when you mix paint. The only one around here who seems to be expecting a "dull or grey yellow" is you.

Right now painting for me is freedom, which means being able to achieve what I want not what is defined by a numeric system. Later on it may change and I may have more interest in it. I'd rather learn earlier on than later.

You'll need to expand on that further please. Is it 2.5 parts G then 5/10 of what? Just need ot understand what that greek means.
Those sufficiently interested can search using the term "Munsell" here and in the Oils forum, and by doing so will turn up all the relevant threads.

Dave

sidbledsoe
02-03-2009, 12:38 PM
Mike, it is no crime to not be familiar with munsell notation and I am sure with further reading you will become well acquainted with it since I know you recently began using oils and have been reading books such as Blue and yellow DMG. Munsell notation measures and assigns a number to the Hue Value and Chroma. Those are the numbers for that particular color green closest to it's hue value and chroma. There is a lot of info out there regarding color. Relatively speaking, I am low on the learning curve but am slowly creeping along. I also just read this book and am looking for opinions of several points that I have questions about, however from a practical color mixing standpoint, I think it works very well.

mrking
02-03-2009, 12:57 PM
Mike, it is no crime to not be familiar with munsell notation and I am sure with further reading you will become well acquainted with it since I know you recently began using oils and have been reading books such as Blue and yellow DMG. Munsell notation measures and assigns a number to the Hue Value and Chroma. Those are the numbers for that particular color green closest to it's hue value and chroma. There is a lot of info out there regarding color. Relatively speaking, I am low on the learning curve but am slowly creeping along. I also just read this book and am looking for opinions of several points that I have questions about, however from a practical color mixing standpoint, I think it works very well.
That is what I was looking for. Thanks!!!!!! :clap::clap:

Einion
02-03-2009, 01:50 PM
Sid, the point that I am making, the one I would really like you and others to take away from this discussion - is that correctly identifying a colour in a scene before you is much more difficult than most people believe.

While being able to mix a desired colour is also an issue, because paints do not generally mix in straight-lines through colourspace, I am suggesting that most people worry about this, while largely neglecting the issue which in fact is more contributary to producing paintings which fail the "realism" test.

Namely, that much of the time, the colour that people are trying to mix isn't even the correct one in the first place.

Now you can either try hitting colours exactly (which can be done in certain well-controlled studio-based situations, which is what's done with the Carder approach), or you can use transpositions, where you maintain the relationships between colours (for example, you may want to compress the range of values, to within what paint will handle). But a common problem (aside from the other drawing and composition issues Einion mentions), is that the colours and the relationships between them are not hit correctly. People learn about the need to get value relationships right, but we also need to get chroma relationships right and hue relationships right (and of these three, hue is, somewhat paradoxically - since hue is the attribute of colour most people think of first - hue is the least important).

I'll say it again - concentrate efforts on correctly identifying a colour in a scene before you - it's much more difficult than most people believe, and getting it right first is the necessary prerequisite before starting to mix coloured paint.

Dave
QFE


Exactly, its an olive green, not a dull or grey yellow
Er, no. It's an 'olive green', not olive green :) This is a very important distinction, despite how nitpicky it may seem - dull yellows can tend to appear to be greens but are not, as is perhaps most easy to see in a colour-picker example like the one posted by Dave.

When mixing black paint with yellow paint however we do actually get greens.

Like 2.5G 5/10 or 10YR 8/12
You'll need to expand on that further please. Is it 2.5 parts G then 5/10 of what? Just need ot understand what that greek means.
You can look up Munsell notation in various places online for more detail but the hue position is given first, then value, then chroma.

Einion

wal_t
02-03-2009, 04:51 PM
Almost everything in life and nature seems to be relative and there are hardly any absolutes other then the ones we choose for convenient purposes. Maybe the same applies to color as a 10YR 8/12 next to a green will be perceived differently by the human eye when it is placed next to an orange.

So what good does it do when I can describe it as 10YR 8/12 but I can perceive this differently all depending on the surrounding environment (or the mood I am in). Maybe the more loose descriptions are not so bad in practical painting after all ?

Regards Walter

dcorc
02-03-2009, 05:14 PM
http://www.artandartistry.com/style_emoticons/default/nohope.gif

If you can identify it as 10YR 8/12 as viewed from your viewing position when painting, that's the colour you need to mix to represent it accurately, irrespective of what surrounds it.

Dave

wal_t
02-03-2009, 05:44 PM
Dave, I am afraid that I can never judge if something is 10YR 8/12 from my viewing position. And even if I could I would not be able to mix it on my palette with accuracy. Besides that, my viewing position changes all the time and so does the environment like the amount of light particles bombarding my canvas. Believe it will be impossible to try and mix the colors one sees in nature. I am pretty satisfied when I can get it about right with the many simplifications that are neccessary.

I disagree with your statement as color (how it is perceived by a human being that is) is influenced heavily - among other things - the surrounding colors and a holistic view is needed for the painting as a whole, not just isolated patches of color.

Regards, Walter.

gunzorro
02-03-2009, 05:58 PM
"You can disagree as much as you like, and believe whatever you choose. However, you are incorrect."

Ha-ha! LOL!!! :)

Painting with faith instead of technique! The last bastion of independence! ;)

Perhaps WC! will open a religious forum?

wal_t
02-03-2009, 06:27 PM
Gunzorro, I have a complete different understanding of the matter and have to live with my eyes and brain. There is no color reality without these as we always need an object, light conditions, an eye and a brain (even if it's only a small one like mine) to see a color. The color I see will always be different then the color you see since we can never be in the same location and see this color in the same light conditions etc. and maybe our brains also operate differently as well :-)

Dave, as I mentioned before I disagree with your statement :-). But maybe, just maybe, it's not all rubbish and incorrect what I am writing. Well I hope not. I am certainly not a specialist in the field of color but believe we can shake hands there. Just following this discussion with interest.

Greetings, Walter

gunzorro
02-03-2009, 07:15 PM
Walter -- It is true that we all see things somewhat differently, and that only increases the need for some sort "control set" to judge by. We could say the same about computer monitors, where serious users in the field calibrate their monitors and printers for consistent output.

We all do the best we can at whatever level our talent and training. The point Dave and I are making is "reach for the stars" and acquire the most knowledge and skill as possible.

I am totally behind innovation and individuality, but frequently the resistence to change and an insistence on independence is grounded in a limited view or sometimes misunderstandings about the concepts that keep one from buckling down and doing the hard work needed.

The light at the end of the tunnel is that with broad experience comes improved discrimination and judgement, and better results.

When it comes to Color Theory and Mixing forum, we are looking for absolutes (or close to it!) to eliminate anecdotal experience and provide rules or methods of seeing and mixing colors, as well as solutions to mixing problems. We've encountered many books on color theory and mixing that were devoted mainly to opinion or gimmicks and not necessarily repeatable or teachable.

wal_t
02-04-2009, 08:22 AM
Gunzorro, thanks, and you have some good points. Fact remains that there are highly skilled "color mixers" without much knowledge on color theories while the opposite is also true; a person that knows almost everyting of one or several color theories can still be a lousy "color mixer". Knowledge and skill don't always go hand in hand I am afraid. I am not saying off course that study of color is a useless thing to do and certainly knowledge can help a lot to improve the skill but I certainly wouldn't overrate the knowledge aspect.

And yes I don't know too much on color theory and found the Wilcox book very helpfull in getting me started in my "mixing tryouts". The theory in his book may be incorrect at places, but I don't care and I certainly think the concepts in this book are teachable and can help at least some people to be able to predict in advance (before the actual "mixing") what the result would be if only approximately.

p.s. I am not a professional artist and painting is (unfortunately) only a hobby for me so not enough time to do all the studying.


Cheers and happy painting, Walter

sidbledsoe
02-04-2009, 09:41 AM
http://marciasandmeyerwilson.com/
Walt here is a wonderful artist I know who's work I think is just great. She most often paints with no reference, from her mind/memory in other words.
Does not use a palette, squeezes out gobs of paint into plastic parts bins, mixes paint on the canvas, (as she says, any color will do if I like it) No color theory knowledge, just experience. Her technique/method can be decribed with three words, she just paints.
This is something I thought would interest you. It is not directed at or meant to refute or argue with anyone elses opinions put forth here. I think there are different valid approaches to making art and that is a good thing.

Einion
02-04-2009, 10:56 AM
Almost everything in life and nature seems to be relative and there are hardly any absolutes other then the ones we choose for convenient purposes. Maybe the same applies to color as a 10YR 8/12 next to a green will be perceived differently by the human eye when it is placed next to an orange.
Yes that's true of course - hence the references above to simultaneous-contrast illusions. But the point I believe is, as was made above, that the first step to painting something as accurately as possible in colour terms is to know what the colours actually are.

So what good does it do when I can describe it as 10YR 8/12 but I can perceive this differently all depending on the surrounding environment (or the mood I am in). Maybe the more loose descriptions are not so bad in practical painting after all ?
That's fine, if the person wants to paint that way. Lots of people do.

But if you want/need greater accuracy then...

I am pretty satisfied when I can get it about right with the many simplifications that are neccessary.
If you're happy with the work that this method produces then that's fine. Different strokes for different folks.

I disagree with your statement as color (how it is perceived by a human being that is) is influenced heavily - among other things - the surrounding colors...
There's ample proof of this provided on many sites on human vision.

There's also ample evidence provided within paintings where certain kinds of colours, in context, are perceived wrongly because of the push-pull effect within the visual field. There are a number of good prior threads, here and in Oil Painting, which touch on this to one extent or another; just look for simultaneous and contrast and you should find many of them easily.

Fact remains that there are highly skilled "color mixers" without much knowledge on color theories while the opposite is also true; a person that knows almost everyting of one or several color theories can still be a lousy "color mixer".
While both of those are undeniably true this is not really relevant. In terms of realistic work the colour can be way off or not and still work as a painting. But the best representational realism is often much closer in colour that other work and that is one of perhaps two key elements, the other being draughtsmanship.

Einion

gunzorro
02-04-2009, 01:23 PM
Walter -- I agree very strongly with your last post. I am an advocate of color theory, but just a much in favor of practical experience. The two go hand in hand, and fortunately we have a number of painters here on this forum (not least of which is the Mod, Einion) who are proficient in both.

I'm glad this book that is the topic of the thread has given you an advance in color theory and mixing. I hope you don't think I am trying to minimize your effort to expand your color horizons -- I'm not. Kudos to you for taking the time to develop your skills.

Any progress is good progress, whether an amatuer or pro. Keeping an open mind and willingness to discuss ideas are a couple strong point in your favor as you explore the realm of color.

Einion -- Excellent post! :)

Einion
02-04-2009, 01:35 PM
The two go hand in hand, and fortunately we have a number of painters here on this forum (not least of which is the Mod, Einion) who are proficient in both.
Just for thoroughness, I'm not a painter. I paint of course, but that doesn't make me a painter :)

I'm of the opinion that unless you paint for a living it's a bit pretentious to use the painter label. Artist, sure; painter, nah. Just as cooking to a high standard in the home doesn't make you a chef.

Einion

wal_t
02-04-2009, 02:51 PM
sidbledsoe,thanks for the link and there is some great - and not so great - stuff there. I love some of the dog paintings :-)

Einion, gunzorro, thanks and will look-up some of the threads as the more I hear you talking the more interest I am getting in this subject. By the way I noticed that there is also a color theory of Goethe but that one will probably do me not much good :-)

Thanks, and happy painting, walter

gunzorro
02-04-2009, 06:55 PM
Einion -- Ha-ha! Got it! :)

mrking
02-04-2009, 07:12 PM
Walter -- As Dave just indicated, and Einion mentioned here or on a related type thread about isolating the targeted color -- the effects of simultaneous contrast, among other effects, will skew the color to look different than what it really is. For an accurate interpretation of reality, the artist must first be aware of these deceptions of the eye and brain, then have a method to override them. Otherwise, such "artifacts" help distinguish amatuerish paintings from fully realized ones.

In a way, it is all relative -- one has the choice to paint like a relative new-comer or advance to a more relative professional level! ;)

I have to comment on this one. So given your above statement Rembrandt, Van Gogh need to be considered amateurish as they didn't use the munsell system to distinguish colour?

I think this debate has gone on past the point of ridiculous. Can we agree to disagree?

gunzorro
02-04-2009, 08:00 PM
Walter -- I wasn't particularly referring to Munsell's system of color notation, or any other specific color method. Simply that the eye has to be trained or methods used to correctly identify color for either realism, or to produce specific effects.

I love both artists you mention, but neither can be cited as achieving colors matching reality. Rembrandt was a genius using an extremely limited color palette, and Van Gogh the opposite with and unrestrained interpretation of subjects. Regardless, both were not only geniuses, but were highly skilled professionals at their work.

Your point is well taken that exceptional artists, through rigorous training and inspiration, can achieve masterpieces.