Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

RGB, CYMK, and cones

  1. Jan 14, 2004 #1

    Nereid

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    In the Theory of Everything board in the MKaku Forum, 8LPF16 asked about colours (see "question" thread), and the discussion got onto the relationship between colour perception, spectral intensities, and the colour systems used in printing, TVs and computer monitors, etc.

    8LPF16 asked some questions I couldn't answer, and am myself curious about. Can anyone here help?

    1) Violet, or purple (English colour names are far from unambiguous!) is a small range of wavelengths around ~400 nm. However, the 'blue' cones have only limited sensitivity to these wavelengths, and the 'green' and 'red' ones virtually none. On my computer screen I get a purple colour with ~60% blue and ~40% red (using the RGB system). The intensity of the colour doesn't seem to matter much, until it become so faint that all colours get washed out (i.e. only the rods are stimulated) What's happening, in terms of rod and cone activity, when you look at the violet part of the rainbow?

    2) Magenta (as in the M in the CYMK system) is clearly not 'a small range of wavelengths around x nm', which is how any colour in the rainbow could be described. In this case, intensity does seem to matter; as it gets faint (or dark), magenta appears to me to become more purplish. What is the spectrum of a magenta-coloured object?
    (This may be a question better directed to someone in a technology board)

    LPF also asks some other interesting questions; perhaps she'll post them here.
     
  2. jcsd
  3. Jan 14, 2004 #2

    Monique

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

  4. Jan 14, 2004 #3

    Nereid

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

  5. Jan 14, 2004 #4

    Monique

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    They are not at all that different, the only difference is that blue has a higher absorption in one.
     
  6. Jan 15, 2004 #5
    Monique,

    The main question was to define the wavelength of magenta - any help?

    I am a Generalist, apparently that led Nereid to assume I would be female. (No)

    I had also mentioned to Nereid that the "red" cone is actually yellow. 579nm peak is 7nm from spectral yellow @572nm. Orange just starts around 614nm, why the insistance on calling that cone red?

    LPF

    P.S. That is a lovely picture of you, if the threads on this forum included photos, I'm sure your "view/reply" ratio would be 100%.
    :wink:
     
  7. Jan 16, 2004 #6

    Nereid

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    colour perception in humans is complicated!

    With the help of Ms Google, I found a couple or three webpages which go a long way to helping me understand this topic:
    http://www.ntsc-tv.com/ntsc-main-05.htm
    http://www.cs.sfu.ca/undergrad/CourseMaterials/CMPT365/material/notes/Chap3/Chap3.3/Chap3.3.html
    http://www.fourmilab.ch/documents/specrend/

    I certainly hadn't appreciated how complicated the topic of human colour perception is!

    Some comments on LPF's questions:
    - the names 'blue', 'green', and 'red' for the cones are shorthand; the actual spectral responses cover ~>100nm
    - for example, a person with non-functioning blue and green cones (and defective rods) could perceive almost the entire spectrum (apparently the red cones do respond to light around 420nm, though the sensitivity is <10% that at the peak, ~590nm); however, if only the blue cones were functional, the person would be blind to all colours longer than ~520nm (what their subjective perception would be is a different question)
    - the intensity-wavelength curves for the RGB phosphors in TVs and computer monitors do not match the response curves of the human cones. This has all kinds of interesting implications.
    - the spectrum of a magenta object will have significant blue and red components. There is no unique 'magenta' intensity-wavelength curve; in this respect, magenta is no different from any other colour (as perceived by humans), including blue, green, and red.
     
  8. Jan 19, 2004 #7
    magenta

    via photons...

    not perception,

    LIGHT !



    LPF
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: RGB, CYMK, and cones
Loading...