🚀 go-pugleaf

In article <162300002@uiucdcsb>, kadie@uiucdcsb.cs.uiuc.edu writes:
> I've got two miscellaneous science questions. 
> 
> 1) On TV's and computers screens, why is it RGB (red, green, blue)
> instead of RYB (red, yellow, blue) the primary colors?

Color perception in humans is achieved by the cone receptor cells (opposed to
rod receptor cells, which respond to dim light). There are three types of
cones: B, G, and R, having each a response curve, which indicates their
sensitivity to photon energy as a function of wavelength (for monochromatic
light). These response curves have peaks at roughly 430nm (somewhat blue),
530nm (somewhat green), and 560nm (somewhat yellow). But these peaks do not
correspond to the colour percept associated to each type. Indeed, a light of a
given colour will stimulate a response from each type of cones, and the
perceived colour depends thus on the ratio of responses between the three
types. Thus the colour percept associated to a type of cones corresponds to
the light wavelength where the ratio of response between this type of cones
and the two other types is highest. This gives:

indigo-blue (around 430nm) for the B cones.
green (around 530nm) for the G cones.
red (around 600nm) for the R cones.

Reference: 
J.L. Schnapf, D.A. Baylor: How Photoreceptor Cells respond to Light.
Scientific American Vol. 256 no. 4, April 1987, pp. 32-39

This is why B,G,R are the primary colours in TV. The combinations B+G gives
the same colour perception as cyan-blue (around 480nm), G+R gives the same
colour perception as yellow, R+B gives the colour compound magenta (which
corresponds to no monochromatic colour in the light spectrum), and B+G+R gives
the same colour perception as white, that is a light having uniform wavelength
distribution in the visible spectrum.

Reference: Any book on colour perception, starting from H. von Helmholtz.

These colour mixtures are called additive. There are also subtractive colour
mixtures, as in painting or printing, where each colour absorbs certain
wavelengths, and a compound of colours absorbs wavelengths absorbed by any of
them. Here the primary colours are cyan (absorbs red), yellow (absorbs
indigo-blue), and magenta (absorbs green). Of course, this is a very
simplified situation, and in practice the primary colours in painting will be
a cyanish blue, yellow, and a reddish magenta.

This issue was discussed in sci.physics a few months ago.

> 2) Some light wave length produces the color green. A mixture of
> the wave lengths of blue and yellow also produces green.
> Even though these two greens are indistinguishable to our eyes, are there
> (could there be) instruments that distinguish them?

Yes. This is called a spectrometer. It is used in astronomy to analyse the
light coming from stars. Or you can refract the light through a prism, as
Newton did.

Christian Ronse		maldoror@prlb2.UUCP
{uunet|philabs|mcvax|...}!prlb2!{maldoror|ronse}

	STAT ROSA PRISTINA NOMINE, NOMINA NUDA TENEMUS

In article <162300002@uiucdcsb>, kadie@uiucdcsb.cs.uiuc.edu writes:
> 1) On TV's and computers screens, why is it RGB (red, green, blue)
> instead of RYB (red, yellow, blue) the primary colors?

Green, not yellow, is the primary color.  The three types of cones in our
eyes have peak sensitivities to red, green, and blue.  Red+green "add" to
appear "yellow", because yellow light stimulates both green and red cones
much more than blue cones. When you mix paints, you are using "subtraction",
which is more complicated. Yellow paint absorbs blue light well, green and
red less well.  Blue paint absorbs red and yellow well, green and violet
less well.  When you mix the paints, you get therefore get a paint which
absorbs red very well, yellow fairly well, blue fairly well, but green only
somewhat, so it appears green.

> 
> 2) Some light wave length produces the color green. A mixture of
> the wave lengths of blue and yellow also produces green.
Not true (see above).  Recasting the question, substituting "yellow" for
"green"...
> Even though these two greens are indistinguishable to our eyes, are there
> (could there be) instruments that distinguish them?

Yes. Absolutely.  They exist.


-- 
David Loewenstern                     Bangpath: {rutgers...}!moss!whuts!davel
The above are not my opinions. 
My lawyer has advised me not to tell you what my opinions really are.

                                     The Answers

          I've gotten answers to my questions about color. Here is a summary.

Q1.       On TV's and computers screens, why is it RGB (red, green, blue)
          instead of RYB (red, yellow, blue) the primary colors?

A.        There are two methods of mixing color, one the RYB for pigments,
          and one the RGB for light. If I remember the explanation correct-
          ly, the light combinations are truly combinations of wavelengths;
          the pigments combinations are really combinations of light wave-
          lengths filtered out by selective reflection and absorption. Thus
          the terms "additive" primaries for the light RGB group and "sub-
          tractive" primaries for the pigment RYB group.

          This whole business of "primary" colors is as much a function of
          how the human eye works as it is of the physics.  We have four
          kinds of light-gathering cells in our retinas.  One kind (rods?
          cones?) is sensitive to low levels of illumination and does not
          distinguish colors.  That's why, at night, everything looks grey.
          The other three kinds have pigments in them, so that one bunch
          responds most to long waves (red), another to intermediate wave-
          lengths (yellow, green), and the third to shorter waves (blue,
          violet).  The sense of vision somehow puts these three signals
          together to make the subjective experience of color.

Q2.       Some light wave length produces the color green. A mixture of 
          the wave lengths of blue and yellow also produces green.  Even
          though these two greens are indistinguishable to our eyes, are 
 	  there (could there be) instruments that distinguish them?

A.        Yes, I think a spectroscope would do it.  It pulls apart the
          different frequencies; the mixed light would show up as a line of
          yellow and a line of blue.

          It is actually, quite hard to get a system to recognize color 
          the same way as your eye (mixing, etc.) It is very easy, however, 
          to get a system to discern the difference between "real" yellow 
          light and green/red mix.

ALSO      Other Information

i.        For more information look up the Retinex theory of color
          vision. There was an article in Scientific American within the
          last seven or eight years.

ii.       There is "no such color" as purple! Mixing red and blue ink 
          causes your eye to react in a way which is not reproducible 
	  by any single wavelength of light.

THANKS to          
          Barry Hayes
          Joe Beckenbach
          Christopher J. Henrich
          John M. Pantone



Carl Kadie
Inductive Learning Group
University of Illinois at Urbana-Champaign
UUCP: {ihnp4,pur-ee,convex}!uiucdcs!kadie
CSNET: kadie@UIUC.CSNET
ARPA: kadie@M.CS.UIUC.EDU (kadie@UIUC.ARPA)

In article <162300002@uiucdcsb>, kadie@uiucdcsb.cs.uiuc.edu writes:
> 1) On TV's and computers screens, why is it RGB (red, green, blue)
> instead of RYB (red, yellow, blue) the primary colors?
> 
> 2) Some light wave length produces the color green. A mixture of
> the wave lengths of blue and yellow also produces green.
> Even though these two greens are indistinguishable to our eyes, are there
> (could there be) instruments that distinguish them?

1) RGB are primary colors for mixing light (eg. red light plus green light
gives "yellow" light). When you are mixing paint, remember that paint
absorbs light, so mixing red & green paint makes this icky brownish grey,
because the mixture absorbs both red and green colors.

2) Simple instruments that will distinguish them are prisms, diffraction
gratings, and fog (rainbows).

-- 
Craig Hansen
Manager, Architecture Development
MIPS Computer Systems, Inc.
...{ames,decwrl,prls}!mips!hansen or hansen@mips.com

In article <162300002@uiucdcsb> kadie@uiucdcsb.cs.uiuc.edu writes:

>I've got two miscellaneous science questions. 

>1) On TV's and computers screens, why is it RGB (red, green, blue)
>instead of RYB (red, yellow, blue) the primary colors?

>2) Some light wave length produces the color green. A mixture of
>the wave lengths of blue and yellow also produces green.
>Even though these two greens are indistinguishable to our eyes, are there
>(could there be) instruments that distinguish them?

(1)
You're talking about two different media here.  A TV monitor produces color
by mixing red, green, and blue light together in varying proportions.  This
is called additive color because you start out with black (no light) and add
light to make color.  On the other hand, when you produce color by mixing
paints, it is called subtractive color because you start out with white (all
colors) and remove light to make color.  For example, red paint absorbes all
colors except red so that when you shine a white light on it, it reflects the
red component and absorbes the rest.  If you shine a blue light on it, it looks
black because there is no red component to reflect.  It simply absorbes all the
blue light and reflects nothing.  Since subtractive systems work differently,
they need different primary colors than those used in additive systems.  For
example, color ink jet printers use primary colors of magenta (reddish),
yellow, and cyan (bluish).

(2)
Be careful, you are talking about subtractive colors, so to say you are mixing
blue and yellow wavelengths to produce green is not correct (see (1)).  Let's
try a different example.  If you mix red light with green light (as in a CRT)
you get yellow.  If you look at this light spectrally (e.g., using a prism),
you will see a peak in the red wavelengths and another in the green.  But it
looks yellow to our eyes because our eyes are fooled (don't ask me why our
eyes do this).  In other words, the mixture of two light frequencies causes our
eyes to perceive a new frequency (yellow).  Consider us lucky our eyes do this,
because otherwise color TV would have been impossible (or at least terribly
difficult).
-- 
Bill McFadden    Tektronix, Inc.  P.O. Box 500  MS 58-639  Beaverton, OR  97077
UUCP: ...{hplabs,uw-beaver,decvax}!tektronix!videovax!bill
GTE: (503) 627-6920         "How can I prove I am not crazy to people who are?"

> kadie@uiucdcsb.cs.uiuc.edu (Carl Kadie)
> 1) On TV's and computers screens, why is it RGB (red, green, blue)
> instead of RYB (red, yellow, blue) the primary colors?

RYB are the so-called "subtractive" primary colors.  Red, green, and
blue are the so-called "additive" primary colors.  When a color
sensation is produced by removing selected components of white light, as
in light reflected from pigments for example, the process is
subtractive.  When a color sensation is produced by adding selected
components to "no light" (or, "blackness"), as in television for
example, the process is called additive.

To oversimplify, the human eye has sensors for red, green, and blue.
Thus, starting with nothing, and adding these three components, any
color perception can be reproduced.  When subtracting from white light,
the "primary subtractive colors" are those gotten by removing one
element of white light at a time.  For example, removing red from white
light gives a bluish color called "cyan".

In fact, the subtractive primary colors are often called red, yellow,
and blue, but to photographers and others who need to be a little more
precise, they are called magenta, yellow, and cyan.

To sum up, there are two "kinds" of primary colors.  In each case, there
are three of them, to a large extent because there are three types of
receptors in the human eye.  In an additive mixing process, these colors
are red, green and blue.  This corresponds to the human eye's receptors.
In a subtractive mixing process, these colors are magenta, yellow and cyan.

The two processes can be related by noting that the additive primaries
can be produced from the subtractive primaries like so:

        red = magenta - yellow
        green = yellow - cyan
        blue = magenta - cyan

And the subtractive primaries can be produced from the addative
primaries like so:

        magenta = red + blue
        yellow = red + green
        cyan = green + blue

Again note, this is an oversimplification.

    (The "green = yellow - cyan" is what corresponds to the over-cute
     "ye-low an' bloo make green" of the ziplock comercials)

> 2) Some light wave length produces the color green. A mixture of
> the wave lengths of blue and yellow also produces green.
> Even though these two greens are indistinguishable to our eyes, are there
> (could there be) instruments that distinguish them?

Yes and no.  Subtracting cyan and yellow gives green, which is not the
same as "mixing the wave lengths of blue and yellow".  In fact, if you
add blue and yellow *light* (as opposed to adding blue and yellow
*pigment* which subtracts light), you'll get white, not green.

With that slight nit out of the way, and translating the question into
additive terms from subtractive, the rest is correct.  One can add red
to green to get a yellow that the eye couldn't tell from a pure
monochromatic yellow.  But the difference would be obvious to a
spectrograph.  In short, the one could be decomposed by a prism, and the
other could not.

--
Another interesting facet of ninja was the use of magic. They had
a reputation as sorcerors and wizards who could fly and hypnotize, and
walk through walls, and get away with huge deductions on their taxes.
    --- Bruce Israel martial-arts-request@brillig
-- 
Wayne Throop      <the-known-world>!mcnc!rti!xyzzy!throopw

In article <162300002@uiucdcsb>, kadie@uiucdcsb.cs.uiuc.edu writes:
> I've got two miscellaneous science questions. 
> 1) On TV's and computers screens, why is it RGB (red, green, blue)
> instead of RYB (red, yellow, blue) the primary colors?
Red, yellow, and blue are the primary *pigments*, which work by absorbing
the contrasting primary light colors.

> 2) Some light wave length produces the color green. A mixture of
> the wave lengths of blue and yellow also produces green.
> Even though these two greens are indistinguishable to our eyes, are there
> (could there be) instruments that distinguish them?

A prism.  Real green light in, green light out.  Apparent green light in,
blue and yellow lines out.

--
John Woods, Charles River Data Systems, Framingham MA, (617) 626-1101
...!decvax!frog!john, ...!mit-eddie!jfw, jfw@eddie.mit.edu

"Cutting the space budget really restores my faith in humanity.  It
eliminates dreams, goals, and ideals and lets us get straight to the
business of hate, debauchery, and self-annihilation."
		-- Johnny Hart

> I've got two miscellaneous science questions. 

>1) On TV's and computers screens, why is it RGB (red, green, blue)
>instead of RYB (red, yellow, blue) the primary colors?

>2) Some light wave length produces the color green. A mixture of
>the wave lengths of blue and yellow also produces green.
>Even though these two greens are indistinguishable to our eyes, are there
>(could there be) instruments that distinguish them?
----------

Your two questions are actually one...red, blue, and yellow are the primary
PIGMENTS, but red, blue, and green are the primary LIGHT COLORS. Red pigment
absorbs blue and green light; yellow absorbs blue and red; blue absorbs red
and green. Mix blue and yellow pigment and everything is absorbed except
green wavelengths, which are reflected. 

Green light does not come about from mixing blue and yellow light, though; 
its wavelengths are intermediate in frequency between blue and yellow. Shine 
a white light through a blue filter and a red filter in sequence, and only 
the green light will make it through. (No, I'm not talking about a blue- or 
red-COLORED filter - I'm talking about one that filters out blue or red light.)

So, in pigments, blue plus yellow makes green. In filters, white minus blue 
minus red makes green.


Don Coolidge

Did I actually write this? That should teach me to not post a response 
late on a Friday afternoon...:-)

> Your two questions are actually one...red, blue, and yellow are the primary
>PIGMENTS, but red, blue, and green are the primary LIGHT COLORS. Red pigment
>absorbs blue and green light; yellow absorbs blue and red; blue absorbs red
>and green. Mix blue and yellow pigment and everything is absorbed except
>green wavelengths, which are reflected. 

Ok, if the green is absorbed, it's clearly not reflected. I can't remember
exactly how to get there from here, though...help, someone!

In article <162300002@uiucdcsb> kadie@uiucdcsb.cs.uiuc.edu writes:
>
>
>I've got two miscellaneous science questions. 
>
>1) On TV's and computers screens, why is it RGB (red, green, blue)
>instead of RYB (red, yellow, blue) the primary colors?

While RYB are commonly called primary colors, there is no real basis
for this distinction. In fact, conceptually, any three distinct colors
could be used for the purpose of creating color images, if we could
subtract light.  In practice, we can only add light sources, thus the
RGB combination is used because it is practical from a cost, benefit,
coverage point of view.  This combination creates one of the larger
gamuts of colors (but not all the chroma that the eye is capable of
perceiving).

>
>2) Some light wave length produces the color green. A mixture of
>the wave lengths of blue and yellow also produces green.
>Even though these two greens are indistinguishable to our eyes, are there
>(could there be) instruments that distinguish them?
>

Yes - but you might want to consider what you mean by "green."  What
is a "green" light? 

The following are a set of references into color vision, the
reporduction of color, and color standards.  If you are interested,
I would suggest that you start with chapter 35 (I think - I am not
certain, but it is somewhere near there) of the Feynman physics text.
It is the most readable of the descriptions of the visual process that
I have found.  I particular, it will answer the question of what is a
"green" light.

@article{kn:faugeras,
	author = "Faugeras, Oliver D.",
	title = "Digital Color Image Processing Within the Framework of a
		 Human Visual Model",
	institution = "IEEE",
	journal = "IEEE Transactions on Acoustics, Speech, and Signal 
		   Processing",
	volume = "ASSP-27",
	number = 4,
	pages = "380--393",
	month = "August",
	year = 1979
	}

@article{kn:horn,
	author = "Horn, Berthold K. P.",
	title = "Exact Reproduction of Colored Images",
	journal = "Computer Vision, Graphics, and Image Processing",
	volume = 26,
	pages = "135--167",
	year = 1984
	}


@article{kn:joblove,
	author = "Joblove, George H. and Greenberg, Donald",
	title = "Color Spaces for Computer Graphics",
	institution = "ACM",          
	journal = "Computer Graphics",
	volume = 12,
	number = 3,
	pages = "20--25",
	month = "August",
	year = 1978
	}

@article{kn:pritchard,
	author = "Pritchard, D. H.",
	title = "{US} Color Television Fundamentals-A Review",
	institution = "IEEE",
	journal = "IEEE Transactions on Consumer Electronics",
	volume = "CE-23",
	pages = "467--478",
	month = "November",
	year = 1977
	}

@article{kn:smith,
	author = "Smith, Alvy Ray",
	title = "Color Gamut Transform Pairs",
	institution = "ACM",
	journal = "Computer Graphics",
	volume = 12,
	number = 3,
	pages = "12--19",
	month = "August",
	year = 1978
	}

@book{kn:boynton,
	author = "Boynton, Robert M.",
	title = "Human Color Vision",
	publisher = "Holt, Rinehart and Winston",
	address = "New York",
	year = 1979
	}

@book{kn:graphics,
	editor = "Beatty, John C. and Booth Kellogg S.",
	title = "Tutorial: Computer Graphics",
	institution = "IEEE Computer Society",
	publisher = "IEEE Computer Society Press",
	address = "Maryland",
	year = 1982,
	edition = 2
	}

*************************************************************
START HERE!!!!!!!!!
@book{kn:feynman,
	author = "Feynman, Richard P. and Leighton, Robert B. and Sands,
		  Matthew",
	title = "The Feynman Lectures on Physics",
	volume = 1,
	publisher = "Addison-Welsey Publishing Company",
	address = "Massachusetts",
	month = "July",
	year = 1964
	}
*************************************************************

@book{kn:foley,
	author = "Foley, J. D. and van Dam, A.",
	title = "Fundamentals of Interactive Computer Graphics",
	publisher = "Addison-Wesley Publishing Company",
	address = "Massachussetts",
	month = "March",
	year = 1983
	}

@book{kn:hunt,
	author = "Hunt, Robert William Gainer",
	title = "The Reproduction of Colour",
	publisher = "John Wiley \& Sons",
	address = "New York",
	year = 1975,
	edition = 3
	}

@book{kn:colortv,
	editor = "Rzeszewski, Ted",
	title = "Color Television",
	institution = "IEEE",
	publisher = "IEEE Press",
	address = "New York",
	year = 1983
	}

@techreport{kn:cie1,
	author = "{CIE International Commission on Illumination}",
	title = "Recommendations on Uniform Color Spaces, Color Difference
		Equations, Psychometric Color Terms",
	institution = "CIE International Commission on Illumination",
	year = 1978
	}

@techreport{kn:cie2,
	author = "{CIE  International Commission on Illumination}",
	title = "Colorimetry-Official Recommendations of the International
		 Commission on Illumination",
	institution = "CIE International Commission on Illumination",
	year = 1971
	}

@techreport{kn:cie3,
	author = "{CIE  International Commission on Illumination}",
	title = "International Lighting Vocabulary",
	institution = "CIE International Commission on Illumination",
	volume = 1,
	year = 1957
	}

>
>Carl Kadie
>UUCP: {ihnp4,pur-ee,convex}!uiucdcs!kadie
>CSNET: kadie@UIUC.CSNET
>ARPA: kadie@M.CS.UIUC.EDU (kadie@UIUC.ARPA)


			Eric Hildum

				dehildum@ucdavis.edu	(Internet)
				dehildum@ucdavis.bitnet	(BITNET)
				ucbvax!ucdavis!dehildum	(uucp)

In article <3290002@hpindda.HP.COM>, dfc@hpindda.HP.COM (Don Coolidge) writes:
> Did I actually write this? That should teach me to not post a response 
> late on a Friday afternoon...:-)
> 
> > Your two questions are actually one...red, blue, and yellow are the primary
> >PIGMENTS, but red, blue, and green are the primary LIGHT COLORS. Red pigment
> >absorbs blue and green light; yellow absorbs blue and red; blue absorbs red
> >and green. Mix blue and yellow pigment and everything is absorbed except
> >green wavelengths, which are reflected. 
> 
> Ok, if the green is absorbed, it's clearly not reflected. I can't remember
> exactly how to get there from here, though...help, someone!

	O.K.  The conventional ``primaries'' for light are red, green, and blue.
A pigment that absorbs red will reflect green and blue (what actually happens is
more complex, involving integrals of energy per wavelength and visual pigment
sensitivities, but it winds up like this), the combination of which is usually
known as cyan.  A green-absorber reflects red and blue, so appears magenta, and
a blue-absorber reflects red and green, and thus is yellow.  Cyan is a greenish
blue and magenta a bluish red, so the red, yellow, and blue pigments you used in
kindergarten are not far off, and in fact can be mixed to form a decent set of
colors.  Notice that they add together into murky brown, not true black, since
they don't absorb all of the light incident on them; in color printing, even
with cyan, magenta, and yellow, it is customary to use black ink for the darkest
areas.
	In fact, the concept of a primary color is meaningless; any three colors
(excepting some degenerate cases) will serve to create a more-or-less wide range
of colors.  One standard color chart (the C.I.E diagram) looks like this, in a
crude rendition:

    |---- Green
    v

   /-\
  |   |
 |     | 
 |      |
|    .  |     . = white
|        |
|        |
----------  <---- Blue

^
| Red

All the colors of a given brightness fall within a horseshoe-shaped boundary,
where the monochromatic colors (colors formed by a single wavelength of light)
make up the boundary.  This diagram has the nice property that a mix of X%
of color A and (100-X)% of color B falls X% along the line between A and B.
Thus, arbitrary mixes of two colors span a line, and mixes of three non-
collinear colors (the degenerate case mentioned above) span a triangle.
You can see from the diagram that it's advantageous to span the triangle with
vertices red, green, blue, since that triangle has the greatest area, and also
is the only triangle that includes the purples (the bottom line between blue
and red).  In reality, though, there are no reproduction processes that even
come close to this ideal triangle; instead they span smaller triangles in the
middle.  The moral is that everyone pretends that their color system is based
completely on logic and necessity, whereas in fact all are compromises, and none
truly create real (spectral) colors.  A photograph can never create the
sensation of a rainbow, for instance.

	Well, enough lecture.  I hope this clears up a few things.

-- 
- Dan Rice, Swarthmore College, Swarthmore PA 19081
``I hear you're mad about Brubeck... I like your eyes, I like him too...''
 UUCP: ...!seismo!bpa!swatsun!rice, ...!sun!liberty!swatsun!rice
CSNET: rice%swatsun.swarthmore.edu@relay.cs.net

Carl Kadie
Inductive Learning Group
University of Illinois at Urbana-Champaign
writes:
>ii.       There is "no such color" as purple! Mixing red and blue ink 
>          causes your eye to react in a way which is not reproducible 
>	  by any single wavelength of light.

The eye can see colours (for example, in afterimages) that cannot be
reproduced by any combination of wavelengths of light !
There was an article in Scientific American c.1970 entitled "Phosphenes"
that went into this.