Primary colors

David Lyga · Jul 21, 2014

For artists who mix paint, the primaries are RED, YELLOW, BLUE. In photography, the primaries are RED, GREEN, BLUE. Is there a simple explanation as to why there is a disparity? And, which set of three shows the most even and rational wave length separation?

I print color with plastic filters. To me, the 'red' seems less red and a bit orange. So, is 'red', RED?

This might be an interesting and informative discussion. - David Lyga

DREW WILEY · Jul 21, 2014

Just get a basic primer on color theory. The three primary additive colors are red, green, and blue. Only kindergarten teachers teaching kinds to use finger paints thought otherwise. This corresponds to the three specific sensitivities of human color vision, which are red, green, and blue. True red is pure red. In photography you'd have to acquire a set of very deep color separation filters to be reasonably on target. Most people don't use filters that intense because they'd to hard to focus with, and slow to shoot through. The relevant red would be a 29, not a 25 or 23.

DREW WILEY · Jul 21, 2014

... but I missed on point in your query, David. You were talking about printing, not shooting. Yes, you could do sequential additive color printing
with a set of glass color separation filters like a 29 red, 58 green, and 47B blue. Printing would be a pain it the butt this way. Yellow is of course paired with magenta and cyan for ordinary subtractive printing. Making a simultaneous additive colorhead is quite an engineering chore. I've
built and routinely use both a 5x7 additive unit and a huge 8x10 one. The quality of the printing light is superb, but the nature of the brains,
the electronics, is pretty tricky.

Rick A · Jul 21, 2014

David, the additive colors for printing are cyan, magenta, yellow (and black). These are also the filters I used to use for color separation work in a print shop whilst attending college. I have to add here, letter press printing (very little offset), I'm old.

Steve Smith · Jul 21, 2014

David Lyga said:
For artists who mix paint, the primaries are RED, YELLOW, BLUE.

Not strictly true. The primaries for ink/paint mixing are cyan, magenta and yellow (secondary colours of light). The magenta and cyan are often refered to as red and blue.

http://www.johnmuirlaws.com/art-and-drawing/color-theory

Steve.

removed account4 · Jul 21, 2014

david there is some information about this here
(there was a url link here which no longer exists)

have fun because some colors just dont exist.

Prof_Pixel · Jul 21, 2014

I remember in the late 1950's someone (maybe Unicolor) sold a color printing kit that included the three additive filters on a notched strip with a holder that went under the enlarging lens making it easy to switch between the R, G and B filters in the dark. Also included was an under the lens diffuser that acted to 'integrate the light' from the negative and a sort of step wedge device that went on the color paper in your easel. The assumption was that the diffuser 'integrated' the scene exposure to Gray (like the color printers of that day did). You used a standard exposure time through the filters and the exposed and processed paper gave you recommended R, G and B printing times. The process worked fairly well. The nice thing was it didn't require you to have a big set of C, M and Y color correction filters.

David Lyga · Jul 21, 2014

OK, mea culpa (of sorts). When I print I use the subractive filters, NOT the primaries (one by one). But I do have a 'red' filter of about '30' to substitute for 30Y + 30M. Professor Pixel, printing with primaries seems to be the "purist's" way of printing color, but, of course, cumbersome, and that is perhaps why it was not as popular as using the subtractive method which handily requires only one exposure in the enlarger. However, I have often wondered if making three exposures through each of the primaries, separately, would bring about better color. (Dodging would be, thus, impossible.)

But for the life of me, when I talk to art teachers (high school) they all claim that RED, YELLOW, and BLUE are the primaries.

But, in essence, I AM (indirectly) talking about the three PRIMARIES because I am using their respective complements: CYAN, MAGENTA, YELLOW for primaries RED, GREEN, BLUE. I am NOT using the complements for the 'artist' colors which would be: CYAN, BLUE, YELLOW for primaries RED, YELLOW, BLUE. It is really confusing and because this seems to be counter to logic (maybe it is not) that is why I am confused. - David Lyga

cliveh · Jul 21, 2014

Mixing paint is totally different to mixing coloured light.

David Lyga · Jul 21, 2014

cliveh said:
Mixing paint is totally different to mixing coloured light.

I guess the substance of what I am asking is 'WHY SHOULD IT BE DIFFERENT'? - David Lyga

DREW WILEY · Jul 21, 2014

Ultimately, this is all based upon either the physics of light or the physiological nature of normal human color vision. Yellow is NEVER a primary
in either. Most printing ink systems use CMY, and sometimes K (black - for when the respective inks or dyes are too weak to achieve sufficient density by themselves). Mixing paint is something I'm fairly expert at, and it abides by the same rules, as do complex inkjet systems which also
use multiple pigments. The nature of vision is still the same, even if you throw in dirty water from the kitchen sink. If you want to argue fine.
Just try printing color paper using an RYB colorhead instead of RGB or CMY. What you end up with will look like dirty dishwater.

Steve Smith · Jul 21, 2014

David Lyga said:
But for the life of me, when I talk to art teachers (high school) they all claim that RED, YELLOW, and BLUE are the primaries.

Then they are all wrong! However, in the printing industry (offset and screen) process magenta and cyan are often referred to as red and blue (but not by me or anyone I work with!).

Steve.

DREW WILEY · Jul 21, 2014

Cliveh - in paint mixing, matching, whatever you call it, whenever we do throw in a mismatch, the official term most of us give it is, in fact,
"dirt". No difference doing it by eye or an advanced spectophotometer (the eye is just more accurate, but fatigues more easily). If you want
a beige, a greige, then you mix a primary with a tertiary, or some less than pure colorant in such a manner to achieve a complex color. If you
hypothetically started with pure red, blue, and green pigments, and mixed them in equal strength and proportion, you'd end up with black. Same if you trying looking through a stack of RGB colored filters, or even any two out of the three. Nothing gets thru. That's why they're called color separation filters to begin with - they don't allow thru colors on the opposite points of the color wheel. ... But if you prefer to eat Elmer's glue and finger paint, then there is the alternative Kindergarten method. With paints or things like inkjet inks things get more complicated in real life simply because pure colorants don't exist, especially if your priority mandates that you can only choose colorants from
among those that will pass thru tiny inkjet nozzles and be cooperative with certain computer programs. Same with architectural pigments.
You have factors of price, lightfastness, physical properties - all kinds of things other than just hypothetical color response. It's an interesting game, and when I was younger I made some good money at it. Too old and lazy now. Just care for my darkroom toys at this point
in life (and getting my own house painted before the dry season ends!)

Steve Smith · Jul 21, 2014

DREW WILEY said:
No difference doing it by eye or an advanced spectophotometer (the eye is just more accurate, but fatigues more easily).)

Over the last twenty five years, we have had a few computerised colour matching systems for screen printing ink - however, they are all sitting on a shelf in a storage room as we have a real person who can do it better!

Steve.

DREW WILEY · Jul 21, 2014

David... I think those "art teachers" were describing the colors they saw after they inhaled a lot of solvent, or maybe smoked something they
shouldn't be. In this town everyone who has green hair thinks they're an "artiste". Any decent text on the subject will tell you exactly the same thing. ... I've never heard of anyone in the printing industry referring to process magenta as red myself, but if they're anything like
the kids who work in the backroom of industrial paint stores, they're probably breathing toluene more often than oxygen, and that might explain
it.

DREW WILEY · Jul 21, 2014

Missed your last post because of mine going in simultaneously, Steve. How I teach it is to do the rough color matching steps by either spectro/
computer or slightly shy of a formula color chip, then fine tune it by eye at the end. That way, by the end of the day, the eye won't be so
fatigued (which leads to inaccuracy). Or finish off the matches in the morning when the eye if fresh. Quality of light is crucial, both for its own
spectral quality and in relation to eye fatigue. But it takes a long time to get good at it. In terms of photographic terminology, I can actually
see within 1cc of color difference, well beyond the average person. But my wife still gets the upper hand when it comes to picking house paint!

cliveh · Jul 21, 2014

DREW WILEY said:
Cliveh - in paint mixing, matching, whatever you call it, whenever we do throw in a mismatch, the official term most of us give it is, in fact,
"dirt". No difference doing it by eye or an advanced spectophotometer (the eye is just more accurate, but fatigues more easily). If you want
a beige, a greige, then you mix a primary with a tertiary, or some less than pure colorant in such a manner to achieve a complex color. If you
hypothetically started with pure red, blue, and green pigments, and mixed them in equal strength and proportion, you'd end up with black. Same if you trying looking through a stack of RGB colored filters, or even any two out of the three. Nothing gets thru. That's why they're called color separation filters to begin with - they don't allow thru colors on the opposite points of the color wheel. ... But if you prefer to eat Elmer's glue and finger paint, then there is the alternative Kindergarten method. With paints or things like inkjet inks things get more complicated in real life simply because pure colorants don't exist, especially if your priority mandates that you can only choose colorants from
among those that will pass thru tiny inkjet nozzles and be cooperative with certain computer programs. Same with architectural pigments.
You have factors of price, lightfastness, physical properties - all kinds of things other than just hypothetical color response. It's an interesting game, and when I was younger I made some good money at it. Too old and lazy now. Just care for my darkroom toys at this point
in life (and getting my own house painted before the dry season ends!)

Drew, I don't dispute this. Why do you think my post mentions anything to the contrary?

Steve Smith · Jul 21, 2014

DREW WILEY said:
Missed your last post because of mine going in simultaneously, Steve. How I teach it is to do the rough color matching steps by either spectro/
computer or slightly shy of a formula color chip, then fine tune it by eye at the end.

I think that's what our colour matching guy does. He has been doing it for thirty years.

Our process is a bit more complex as we print colours onto the rear of polyester and it's usually backed up with an all over white ink to give it density and often a dark colour will be backed up with one or more of the lighter colours and then the white.

Our guy is very good at predicting the effect of the back up colours and can quickly get the right mix.

Steve.

gone · Jul 21, 2014

There is a difference between opaque, pigmented colors, and colors that are designed for light to go through them (like a lens filter). Red, yellow and blue are paint primary colors. If you are working in printing, those are translucent and will be different. If you're talking about transmitting light (filters), it's something different again. So when people speak of red, yellow and blue being the primary colors, they are of course correct....for paints. If printing, it's subtractive, and CMYK. We could talk about the secondary and complimentary colors, but we'd better not.

https://www.google.com/search?q=wha...heel-artist.com%2Fprimary-colors.html;288;252

pentaxuser · Jul 21, 2014

What David Lyga has stated about what is taught as primary colours theory in painting is the case. I remember being told that the primaries were red, yellow and blue whereas in colour photography and colour printing I read that the primaries are red, green and blue.

If both are right then can someone explain in straightforward, relatively simple terms how these can be reconciled and if the paint colour primaries are wrong then what is the flaw and why is it that red, yellow and blue are wrong?

I don't think that David has been given a simple explanation to date and if he believes he has then it will have to be me that you pity. So please will someone try and explain this difference?

As far as I know painting class instruction hasn't changed and painting teachers still persist with their "theory"

Thanks in advance to anyone who both knows enough about both media and is patient enough to attempt to explain this.

Pretend that your are a teacher and your salary depends on how many of the students can give a reasonable explanation of the difference to a panel of examiners

Thank you for your indulgence

pentaxuser

DREW WILEY · Jul 21, 2014

Red, yellow, and blue have NEVER been classified as primaries, and never will be. Maybe art teachers taught such a thing because that comprised the colors they ate as little kids, and might be related to the fact that it is generally difficult to get a true green pigment (pthalo
blue is fairly blue, pthalo green is way off from green, for example, and chromium oxide is worse - at least in terms of well known traditional colorants). Any art teachers who teaches this is not in fact teaching actual color theory but hearsay. The opacity or translucency of colorants is a different subject, relative to the intended application. In color printing or shooting thru a lens, and opaque color would of course be useless. If you want process colors for something like a pigment print, it would be wonderful is they were relatively transparent like dyes (which they aren't). If you are mixing a house paint, ideal colorants are as opaque as possible (which they often aren't). Theory is
one thing, industrial and artistic ingredient compromises are something else.

DREW WILEY · Jul 21, 2014

Steve - in this area there are two distinctly different kinds of color matchers. The ones who do things like packaging on assembly lines get paid a lot of money, but their needed skill is the ability to toss in large quantity of ingredient very quickly in a volume assembly situation, and still come out with something acceptable. It doesn't need to be dead-on in terms of hue. On the other hand, someone who mixes paint for auto touchup or architectural applications is graded by the ability to get the color very accurate, and without metamerism. If you can also consult, then the money is quite good. Of course, the average monkey in the average home center can recognize and mix colors just about as well as a mole rat born without any eyes, but somewhat better than the average artiste with green hair around here.

Photo Engineer · Jul 21, 2014

Additive colors are R/G/B and Subtractive colors are C/M/Y. Additive "systems" work by emitting light and thus a TV is an additive system where R+G+B = white and the absence of any color is black. Additive systems can only be used by emitting light which we mix in our eyes to make the entire gamut of colors.

Subtractive systems work in prints, films and paintings by absorbing light striking them. Thus, C+M+Y = black and no color = white, or rather it equals the color of the illuminant.

Early color systems were additive. This included Autochrome and Dufay Color which used R/G/B type pigments to filter the film during exposure and then separate white light into a color image after processing. This was much like current color TV.

PE

DREW WILEY · Jul 21, 2014

Color printing is normally done subtractive because true additive filters are much more dense, and therefore require a much more powerful light
source, and therefore lead to a lot more heat in the system. There are ways to cope with this, but involve more engineering difficulty. You have to mix them after the fact, after filtering thru them - so that means multiple light sources, as opposed to CMY subtractive, where simply you can stack them to filter simple white light. Cool sources like LED simply don't exist in the right narrow-spectrum colors at the moment. Maybe some day.

Bill Burk · Jul 21, 2014

Suppose you have three bands of light you are interested in... Red Green and Blue

The Additive colors pass one band of the three, to be added as light. Shine three lights, one through each Red, Green and Blue filter and vary how bright the lights are, these add to get the color you want.

The Subtractive colors each pass two bands of the three colors. Mix them in varying proportions and where a couple Subtractive colors overlap, because they block one band each, a mixture of any pair blocks two and allows one band through.

Primary colors

Privacy & Transparency

Privacy & Transparency