3 ms·

I'm flogging a dead horse here, I apologize. But doesn't a color simply consist of levels of Red, Green, and Blue and the Pantone is essentially directing the u

by cheath 10y ago

I'm flogging a dead horse here, I apologize. But doesn't a color simply consist of levels of Red, Green, and Blue and the Pantone is essentially directing the user to achieve those levels. And therefore, couldn't that be easily achieved through something like RGB? I understand that the answer is a "no" and that I'm wrong. I just don't know why I'm wrong.

wmeredith 10y ago
Because mixing those RGB/CMYK?else levels are not the only factors affecting the actual color produced. You have environmental factors, human error, local temperature, media temperature, curing environment humidity, and on and on (and on). But if YOU have a pantone swatch and I have the same pantone swatch. We now have a reference that is the SAME color. I can tell you to keep adjusting your local parameters until the thing you're coloring (paper, fabric, plastic, rubber, metal) matches the swatch.
andreareina 10y ago
In theory, a given value -- say xff0000 -- and a color profile -- sRGB is the standard for most "consumer" uses e.g. monitors, cameras, etc -- specify an absolute point in color space. In practice, my monitor's 0xff0000 is going to look different from yours due to differences in manufacture. In LCD monitors it would be the backlight (color and intensity) and filters; in CRT monitors it would be the phosphors and beam strength. It's possible to bring them in line with calibration, but they'll drift again since phosphors and filters change their properties with age and use. It's the same with printing, except that calibration is more effort and expense. Enter Pantone. Pantone inks are pre-mixed so the color is always consistent. What printers do is they print most of the image using the normal four-color process (CYMK) on a "best-effort" basis and then do a separate pass on critical areas using specific Pantone inks that are guaranteed to yield a specific color. The extra process is expensive so spot colors are typically limited to brand-specific colors (e.g. Coca-Cola Red) and other things where precise color reproduction is important.
mindslight 10y ago
> But doesn't a color simply consist of levels of Red, Green, and Blue No, not at all. Pantone seems arduous because you're unaware of all the simplifying assumptions that you're otherwise making. Your perception of a color for a single illumination stimulus is indeed a simple three-dimensional quantity (as your eyes have three color channels). But the stimulus itself is defined by an intensity at every wavelength. For an extreme example, something that fluoresces takes in say UV light and outputs lower wavelengths. I think the Pantone model simplifies away things like fluorescence as well [0], but I've thrown out the example to show how many dimensions we're actually dealing with. For a purely diffuse example, first take a look at https://en.wikipedia.org/wiki/File:Cones_SMJ2_E.svg https://en.wikipedia.org/wiki/File:Cones_SMJ2_E.svg . To my approximate reading, 530nm and 575nm have equivalent responses for your "green" color receptors. If we have one material that does not bounce 530nm and one that does not bounce 575nm, they can look the same if illuminated with equal amounts of each [1] but then look different if illuminated with a "same colored" source that is lacking 530nm. This is why people have problems with fluorescent and LED lights - colors look weird because their emissions are "peaky" and not the smooth "black body" curve we're used to. [0] Although maybe not, with the prevalence of whiter-than-white papers, laundry detergents, etc. [1] With each material having the appropriate amount of red/blue bounce to equalize those channels as well.

by cheath 10y ago

wmeredith 10y ago

Because mixing those RGB/CMYK?else levels are not the only factors affecting the actual color produced. You have environmental factors, human error, local temperature, media temperature, curing environment humidity, and on and on (and on). But if YOU have a pantone swatch and I have the same pantone swatch. We now have a reference that is the SAME color. I can tell you to keep adjusting your local parameters until the thing you're coloring (paper, fabric, plastic, rubber, metal) matches the swatch.

andreareina 10y ago

In theory, a given value -- say xff0000 -- and a color profile -- sRGB is the standard for most "consumer" uses e.g. monitors, cameras, etc -- specify an absolute point in color space. In practice, my monitor's 0xff0000 is going to look different from yours due to differences in manufacture. In LCD monitors it would be the backlight (color and intensity) and filters; in CRT monitors it would be the phosphors and beam strength. It's possible to bring them in line with calibration, but they'll drift again since phosphors and filters change their properties with age and use. It's the same with printing, except that calibration is more effort and expense. Enter Pantone. Pantone inks are pre-mixed so the color is always consistent. What printers do is they print most of the image using the normal four-color process (CYMK) on a "best-effort" basis and then do a separate pass on critical areas using specific Pantone inks that are guaranteed to yield a specific color. The extra process is expensive so spot colors are typically limited to brand-specific colors (e.g. Coca-Cola Red) and other things where precise color reproduction is important.

mindslight 10y ago

> But doesn't a color simply consist of levels of Red, Green, and Blue No, not at all. Pantone seems arduous because you're unaware of all the simplifying assumptions that you're otherwise making. Your perception of a color for a single illumination stimulus is indeed a simple three-dimensional quantity (as your eyes have three color channels). But the stimulus itself is defined by an intensity at every wavelength. For an extreme example, something that fluoresces takes in say UV light and outputs lower wavelengths. I think the Pantone model simplifies away things like fluorescence as well [0], but I've thrown out the example to show how many dimensions we're actually dealing with. For a purely diffuse example, first take a look at https://en.wikipedia.org/wiki/File:Cones_SMJ2_E.svg https://en.wikipedia.org/wiki/File:Cones_SMJ2_E.svg . To my approximate reading, 530nm and 575nm have equivalent responses for your "green" color receptors. If we have one material that does not bounce 530nm and one that does not bounce 575nm, they can look the same if illuminated with equal amounts of each [1] but then look different if illuminated with a "same colored" source that is lacking 530nm. This is why people have problems with fluorescent and LED lights - colors look weird because their emissions are "peaky" and not the smooth "black body" curve we're used to. [0] Although maybe not, with the prevalence of whiter-than-white papers, laundry detergents, etc. [1] With each material having the appropriate amount of red/blue bounce to equalize those channels as well.