CMYK is a very limited colorspace; not nearly as broad as RGB. It’s the subtractive color model and was developed for color reproduction processes like lithography and other ink/dye/pigment based techniques and mediums. It’s so limited because of the physical nature of how inks mix in an impure way.*
Whereas RGB, being an additive color model, which can mix its primaries in a much purer way, since it’s dealing almost directly with light itself.
*Which is why black(k) is added into the mix, since CMY alone produce a muddy black at best. Black ink compensates for this. See CMYK vs. RGB color gamut here.
If I’m working with images or graphics for print, I’ll usually work in 8 or 16 bit sRGB, and virtually proof how the colors will collapse in the CMYK space as I work (and the results are always comparatively dingy, losing its vibrance—especially in the blues). I’ll also convert all my images to CMYK myself, keeping RGB copies as masters, before I send off to print.
Don’t get me wrong. If the work was going to four color, it would eventually have to be made into CMYK separations. It’s just that the production department usually handled it or, in the case of my vendors, they prefer to deal with RGB and do their own separations. Of course, a single CMYK file can be separated a number of ways, and you can often fine tune to he separations for optimum reprintability if you do the CMYK seps yourself. So, if one really knows what they’re doing (like you)they can achieve better results from doing their own separation than trusting the folks on the other end to do it optimally for their content.
But stuff like home printers–a reasonable amount these days have more than four inks, so a CMYK separation wouldn’t be appropriate, anyway, and you’d lose the wider gamut afforded to you by the extra inks if you went into CMYK. That’s all.
This debate is all a confirmation of something I came to understand over the years: designers and printers see things, from technical standpoints, in almost mirror-image of each other. The way a designer sees a color definition is an inverse of the way a printer does, and that applies to many aspects of getting from a layout to a printed item.
And even though designers are usually working in RGB, I think for most people CMYK is the more intuitive model, as subtractive color theory reflects the way most of us learned about mixing color in art class. Even though I almost never venture into CMYK these days,I still look at CMYK values to check things like skin tones.
Of course, but my experience goes back to the rubylith days, when printers (who spent years sucking down ink and solvent fumes) did it one way and everyone else was just stupid.
I don’t think most people know CMYK from ROFL. Certainly the vast majority of graphics tools use only RGB and don’t even offer CMYK as a working model. (Photoshop Elements = Photoshop with everything CMYK removed.) You might be right in that people think in additive colors, though.
Heh… I’d almost go so far as to say that RGB isn’t intuitive for anything other than populating color registers
Pretty much. I think outside of Gimp and Photoshop CMYK isn’t an option. Maybe there are some outliers. I’m guilty of ignoring it as well in my software. I wrote a pixel editor some years ago that handles RGB, HSL, HSV, and uses Lab for a few internal calculations.
I think you’re missing my point. Most people know “yellow plus blue make green” which is a subtractive process. When they learn that “red plus green makes yellow” in RGB, that’s not intuitive to most.
Subtractive colors. I work almost exclusively in RGB and I still have to stop and think sometimes about additive color theory as it’s hard to overcome my formative years being exposed to subtractive theory. Whether it’s CMYK or red yellow blue (as we were taught were “primaries”) is kind of immaterial. It’s just the idea of mixing pigments vs mixing light.
Missed the edit: Although I could be wrong. Maybe art theory as taught today puts more emphasis on additive theory, where red+green=yellow. I don’t have any kids yet, so I don’t know what the current curriculum teaches. I just know that I wasn’t exposed to red+green=yellow until physics class in high school, but that was the early 90s and computer art wasn’t that big a thing. So it’s possible kids today are way more savvy about this. We just mainly screwed around with crayons and watercolors and pigments in general, where subtractive theory was appropriate.
Agreed about CMYK not being well known (but more intuitively understood per pulykamell’s point), which I find interesting due to the almost ubiquitous uprising of inkjet printers in the home and office.
The colors/letters are right there every time you have to change those damn cartridges… and still, people think my username is pronounced “smick”!
I have kids, and unfortunately no. Still the errant RYB subtraction model persists. Go up to any kid today and ask them what colors to mix to make green, you’ll never hear “Yellow and cyan! Duhh!”
The additive color model doesn’t seem to even come into play until you get into newtonian optics and the EM spectrum.
I’d say most people think in an RYB subtractive model. People don’t see red and think it’s a combination of magenta and yellow. They see magenta and think it’s a light red with blue mixed in.
And I think that, as long as you remember that red and green make yellow and that purple has a lot more blue in it than red, thinking in an RYB model translates pretty well to RGB. Sure, adding colors brightens rather than darkens, but that’s not too hard to wrap your head around.
It’s still something that causes me to stutter a bit when thinking. I just remember that RGB pair up with CMY (R with C, G with M, B with Y). In the additive model, cyan, magenta, and yellow are made up of the sum of the other two channels’ complements. So, cyan is G+B, magenta is R+B, yellow is R+G.
Still, even though I’ve been working in RGB almost daily for 20 years, I still find it easier to think in subtractive color much of the time because of how I was exposed and taught to think about color for the first 18 years of my life (minus the brief section of physics class where we were taught some additive theory). I suspect if I were taught both models simultaneously, it would be as instinctual for me.
