Can CMY cover all colors ? If not what does it ?

[snapte]

Hello , I am after to create perfect autochrome dye colors one by one with cmy filters ????

Hey Mustafa, did you end up having much luck with this? I am currently trying to do something similar, and have good acetate screens but are yet to make my own panchromatic emulsion

 

[tcolgate]

Colour gamuts are complicated as colour is a largely neurological thing (not the simple physics of wave lengths the "magenta does not exist" crowd would like you to believe).
If you imagine a typical colour wheel (YRMBCG, the middle being a neutral grey), you can take three points on that wheel (we call that a triad), and from the colours you chose, you can "mix" any of the colours inside that triangle, and none on the outside of it. So if you pick a pure saturated CMY, you can mix all the things inside the triangle you would draw between those colours. The red, green and blue will not be the purest saturated possible colours, but as it happens, we almost never actually need purely saturated colours in printing or painting. When we do need them, we can just reach for a pigment different pigment. So fancy printers will let you do exactly that, you can keep to CMYK, but add in additional inks (this is one of the main uses for pantone colour names).
Our brain also does lots of interesting stuff with colour in close proximity. An extreme example of that is a Zorn palette in painting: vermilion (red), ivory black (a cool, very faintly blue leaning black), white and yellow ochre. There are lots of colours you can't reach with this palette, but you can get the brain to interpret some colours as leaning a bit more blue, or even green, by what colours they are next to.
If you take the colour wheel mentioned about, there are a couple of extra aspects that are worth visualising. Image that wheel in the middle, then take a point and stretch it up, so you get a cone, and imagine that is pure white, then stretch a second point down int he opposite direction, imagine that is pure black. That's the 3d "shape" of the typical colour gamut we think of, but we can get more accurate. For instance, not all hues (colours), are equally saturated at what we perceive as maximum saturation. The bluest blue, is darker, than the yellowest yellow. So if we imagine that that middle circle isn't actually a circle, but each hue is proportionally as far out as the brighest it can get (hard to describe), we end up with this wobbly shape. If you do the double cone thing but with that shape instead, that's more or less all the colours and tones we can see).
Then it gets even weirder because we can perceive different granularity of tones in different hues (we can distinguish a much larger range of greens than any other hue).
That before you even get to colour blindness or even tetrachromacy (some women can see more colours than the rest of us).

 

[tcolgate]

I learned that postcard printers printed with RYB inks , not rgb or cmyk.

Traditional painting typically uses RYB as the basic triad. This is probably just down to the availability of different pigments at different times. A lot of green paints are annoying to mix with. I think it's just hard to mix a clean yellow than a decent green.

 

[DREW WILEY]

Kindergartners with finger paints were taught RYB. But that is not how either our own sight works, nor color filtration for photographic purposes. In the real world, when color filters are needed for printing color separations,
typical CMY filters also pass through a fair amount of residual white light as well, making them less precise than more dense RGB color separation filters. Almost all commercial printing was based on RGB, plus a K black separation due to printing ink limitations.

Basic color theory should always be studied first. But achieving a usable gamut afterwards can get awfully complicated at times. Just look at how many different inks are involved in modern inkjet printing, and still there are distinct flaws in the quality of reproduction. I analogously worked with architectural coatings and pigments, doing color consultation; there are many pertinent variables. Every process has its idiosyncrasies, and every dye and pigment has its limitations. The whole point is to recognize those, and make them work in our favor, technically and esthetically.

 

[tcolgate]

Kindergartners with finger paints were taught RYB.

RYB is the traditional limited colour triad for painting with actual paint. It's not just taught in "kindergarten", it is taught in art schools and ateliers. With the commonly used pigments, it works well enough. It is not "correct" by some measures, but it is pragmatic when it comes to crushing up rocks and smearing stuff on a canvas.

 

[Samu]

Yes, CMY can create all colors of the spectrum, if the amount of light is also regulated by shutter or aperture. and the light source is neutral white, containing all the shades of visible light. This would be an ideal system for subtractive color. In reality, no system is ideal. Also, this is true only for the spectrum visible to human eye.

 

[DREW WILEY]

The relevant point is, get ahold of the deepest yellow filter you can find, and you can still see other colors through it, on the opposite side of the color wheel. But do the same with a blue color separation filter like a 47B, and you can't see any yellow or green at all, nor red. Same applies to magnenta and cyan - no matter how deep you get them, other colors still get through. I have Wratten examples of the deepest magenta you can buy (or could have formerly gotten), and it still isn't as strong as additive RGB filtration.

I work with both conventional CMY colorheads for enlarging, and true narrow-band RGB additive ones, so am quite aware of the fundamentals. There's no need to repeat what I've described on other threads already. But practitioners learned way back, around 150 yrs ago, that trying to reproduce color starting out on the premise of CMY is a far more complicated route than RGB. We normally separate colors RGB, then print them with CMY dyes or pigments afterwards, and not the other way around.

 

[alanrockwood]

No, there are colors than can't be reproduced by any single set of CMY dyes. The same applies to RGB light sources. This is well known in color theory.