This
level-4 vital article is rated B-class on Wikipedia's
content assessment scale. It is of interest to the following WikiProjects: | ||||||||||||||||||||||||||||
|
|
Is there any article on Audible spectrum? Anwar ( talk) 19:35, 3 May 2008 (UTC)
Isn't the exact range of the visible spectrum disputed?--Luke Elms 20:33, 17 May 2008 (UTC)
Human eyes do not see wavelengths, they see frequencies. The color of light perceived through water, for example, is not massively affected by the drastic decrease in wavelength because the frequency is about the same. — Preceding unsigned comment added by 129.210.115.18 ( talk) 21:50, 18 June 2013 (UTC)
Other sources (many, I think) indicate not the round numbers "400 to 700," as in this article, but 380 nm to 750 nm or more. as in the electromagnetic spectrum article (the citation of Biology: Concepts and Applications looks tertiary at best). Yellowing on of the lens of the eye due to aging causes loss of sensitivity at the short wavelength end. I recommend that someone correct the numbers in the article and propagate the changes throughout. Eric Drexler ( talk) 12:33, 17 March 2013 (UTC)
Here are authoritative sources that present quantitative data re. human eye sensitivity vs. wavelength (and age of subject, at the violet end):
The article should be corrected, and perhaps other articles as well. There’s a lot of misinformation in circulation. Eric Drexler ( talk) 12:58, 17 March 2013 (UTC)
Saw a short article a while ago in a science news magazine that Birds see more of the spectrum than we do. They have additional cells in their retina that also have a tiny bubble of oil over them.
The speculation was that the non-bird animals or proto-mammals that survived the extinction of the dinosaurs (the great K-T Extinction) were almost exclusively nocturnal types and did not need the extra part of the spectrum anyway. I got the impression that this part of the extension was not at either end of our visible spectrum, but somewhere nearer the center. I am not sure this makes sense, but maybe some can look into it. WonderWheeler ( talk) 23:20, 20 August 2008 (UTC)
Birds are tetrachromats and can see down to 320 nanometers in the ultraviolet, whereas since we are trichromats, our violet limit is 380 nanometers. Pigeons are pentachromats and besides seeing into the ultraviolet, can also see into the infrared beyond 750 nanometers. Keraunos ( talk) 05:20, 22 March 2011 (UTC)
While all the notes about other species are interesting, I think that the tendency (especially of new editors) to simply chime in on the article with unsourced factoids about this or that species' abilities needs some kind of control. After all, the article begins with the true statement that the article is about what the human eye sees. The introductory paragraph that diverges onto the topic of other species is not only slightly off topic, it also begins to look like it was designed by committee. Which of course it was! Ehusman ( talk) 21:00, 24 March 2013 (UTC)
I've made an image of the visible spectrum in Inkscape (a rectangle with a linear gradient having various transition points), which got pulled out of this article rather quickly because it looked odd. My aim is/was to try to make it represent the spectrum using transition points that are as precise as possible and fit the RGB colours represented on the respective colour pages ( red, orange, yellow, green, blue, violet). Spectral violet is obviously a tricky one, not actually being in the RGB Gamut. I've already reduced the luminosity of a few colours to make it blend better -- any ideas on how to improve it? Perhaps a logarithmic representation would be better? gringer ( talk) 03:14, 28 August 2008 (UTC)
Here is a logarithmic representation, with wavelengths at the boundaries marked. gringer ( talk) 04:11, 28 August 2008 (UTC)
I uploaded an inverse spectrum (linear by wavelength). Mach bands seem to be a bit more noticeable in this one. I've since realised that doing the whole "make it look like a prism" thing is silly, because dispersion is probably not going to be a nice function of wavelength, and is material-dependent. gringer ( talk) 03:54, 30 August 2008 (UTC)
I believe CIE publishes tables of 2° and 10° standard observer values for the spectrum, as XYZ or xyY coordinates. I’m not sure how to track such a table down though. Dicklyon, do any of your color science books have something like that at the back? — jacobolus (t) 06:14, 30 August 2008 (UTC)
This seems pretty reasonable, and ends up with several renderings to choose from. — jacobolus (t) 12:30, 30 August 2008 (UTC)
Skoch3 ( talk) 07:50, 3 November 2008 (UTC): I changed the text for spectral colors (and included a link). I took out the reference to rainbows, because rainbows have significant overlap of colors with each other, and thus the colors are not spectral.
Skoch3 (
talk) 07:56, 3 November 2008 (UTC): While I can see that a significant amount of work has been put into this image of the spectrum, I think as is it is very misleading. As far as simulations go, it is quite different from the image on the
spectral color page (the spectral colors are on the outside horseshoe...and are quite different from the current image on this section). However, I think much better than a simulation would be a good photo of a white light spectrum, spread out by a diffraction grating, along with wavelength calibration. I am thinking of switching to add the image to the right, above. The most striking thing missing to me is the cyan color, which is lacking in the current simulated image. I agree that the NASA image is a bit dark.
The actual data spectrum from NASA represents a particular source, whereas the brighter illustration is about the general concept of visible sprectrum. The latter works better as an illustration, I think. It's hard to make good representations of pure spectral colors in RGB; the chromaticity plot has other goals, being normalized by x+y+z such that it doesn't go black at the ends of the spectrum, and also shows purples. The existing image is also not great, having poor yellow and cyan. Dicklyon ( talk) 20:54, 3 November 2008 (UTC)
"In the illustration, the narrow red, green and blue bars show the relative mixture of these three colors used to produce the color directly above.
what illustration? 12.125.134.114 ( talk) 16:22, 8 October 2008 (UTC)
I changed the name from visble spectrum to visible light spectrum -- Chicagobears94 ( talk) 21:21, 10 January 2009 (UTC)
Actually, you moved it to Visible Light Spectrum, which anyone who has been editting wikipedia should know is improper capitalization. I moved it back. If you think that Visible light spectrum would be better than Visible spectrum, make a proposal here and we can talk about it. You're not the only one likely to have an opinion on this. Dicklyon ( talk) 06:39, 11 January 2009 (UTC)
Why do we have this article and also have the Light article whose first line says: "Light, or visible light, is electromagnetic radiation of a wavelength that is visible". Also, we have Visible Light redirect here. -- MarsRover ( talk) 19:54, 21 March 2009 (UTC)
I had previously attempted to merge light with electromagnetic radiation, which was rejected, but the discussion appeared to suggest that the best way of consolidating the information was to perform a merge or partial merge between visible spectrum and light, called either light or visible light. I was wondering how this might be achieved. Serendi pod ous 15:27, 5 October 2009 (UTC)
Re: removal of indigo from the table. I approve. If I had noticed that the table was drawn from a source I would have removed it too.
The idea that there is a distinct colour "indigo" in the spectrum between blue and violet has no basis in objective fact nor even unbiased qualitative observation. It was added because of Newton's bias that the number of colours should be the same as the number of notes in the musical scale.-- Srleffler ( talk) 23:29, 4 December 2009 (UTC)
Capabilities of one's vision seems to be dependent on their growth environment, culture and perhaps even language. Therefore talking about distinct colors is highly subjective matter. See "BBC Horizon: Do you see what I see?" or this YouTube clip: http://www.youtube.com/watch?v=4b71rT9fU-I . — Preceding unsigned comment added by 80.75.107.170 ( talk) 09:09, 16 May 2012 (UTC)
The image at the top of the page, showing a narrow beam of white light passing through a prism, is quite inaccurate and needs replacing. The problem is that the beam as it is shown inside the prism, after having passed through the left-side wall, is still a narrow beam of white light when it should be already diverging into color separations as it has been refracted once. Other images on the web get this part correct.
TimProof ( talk) 20:51, 18 March 2010 (UTC)
Hi everyone. There are currently a ton of spectrum images spread through English wikipedia & the rest of wikimedia projects. I’ve stuck the ones I could find over to the right (this doesn’t include the vertically oriented ones, or the couple that were flipped right-to-left, or scaled logarithmically, but otherwise identical to one shown here). None of them seems to me to have an especially legitimate methodology behind its construction, and most don’t even explain how they were generated. This is understandable enough – there’s no clear right approach to rendering something that falls so far outside the sRGB gamut. Still, I’ve been playing with it, and the best rendering I can come up with results from the steps: (1) Use the CIE 1931 standard observer data to define XYZ at each wavelength; (2) Convert each XYZ triplet to CIECAM02 space, assuming D65 white point and “average” surround; (3) Apply a bit of a gamma curve to the resulting J values, to brighten the darkest parts so they’re a bit easier to see; (4) Compress this gamma-adjusted J to fall between 8 (the sRGB black point) and 75 – i.e. map [0, 100] onto [8, 75] – so that we can get a decent amount of chroma at every point; (5) For each (new) J value, and each h, take the point within the sRGB gamut with maximum possible chroma. Some of these steps are somewhat arbitrary, but the key parts of this method are (A) perceived hues are preserved, and (B) The relative lightness is preserved, with just enough fudging to make things look somewhat reasonable.
Anyway, I’m going to keep fiddling a bit, and then upload the result I get. It might be worth even ending up with two different renders by different methods (in particular, it might be worth ignoring or partially ignoring relative lightness, and just taking the most colorful possible color for each hue, so we can have a bright yellow, a bright cyan, etc.), so that readers can compare them. Having 6 or 10 though, as we do now, seems sort of absurd. We should try to get some kind of consensus about color content, and then about what labels/etc. are necessary, and then just put up a couple of these, in SVG format, and recommend (on all the image description pages) that articles use those.
Cheers,
jacobolus
(t) 20:10, 6 April 2010 (UTC)
if (abs(sin(hr)) >= abs(cos(hr))) { p4 = p1 / sin(hr); cb = (p2 * (2.0 + p3) * (460.0 / 1403.0)) / (p4 + (2.0 + p3) * (220.0 / 1403.0) * (cos(hr) / sin(hr)) - (27.0 / 1403.0) + p3 * (6300.0 / 1403.0)); ca = cb * (cos(hr) / sin(hr)); } else { p5 = p1 / cos(hr); ca = (p2 * (2.0 + p3) * (460.0 / 1403.0)) / (p5 + (2.0 + p3) * (220.0 / 1403.0) - ((27.0 / 1403.0) - p3 * (6300.0 / 1403.0)) * (sin(hr) / cos(hr))); cb = ca * (sin(hr) / cos(hr)); }
=if((H94=0),0,if(abs(sin(H83*pi()/180))>=abs(cos(H83*pi()/180)),(((H90/H48)+0.305)*(2+(21/20))*(460/1403)) /(((H89/H94)/sin(H83*pi()/180))+(2+(21/20))*(220/1403)*(cos(H83*pi()/180)/sin(H83*pi()/180))-(27/1403)+ (21/20)*(6300/1403))*(cos(H83*pi()/180)/sin(H83*pi()/180)),(((H90/H48)+0.305)*(2+(21/20))*(460/1403))/(((H89 /H94)/cos(H83*pi()/180))+(2+(21/20))*(220/1403)-((27/1403)-(21/20)*(6300/1403))*(sin(H83*pi() /180)/cos(H83*pi()/180)))))
I have taken an independent approach to render a visible spectrum. Currently PNG only since
gnuplot has difficulties to create proper SVG images (or I didn't find out the right way). The method is similar to that of
gringer, i.e. a binary search along a constant-hue line in CIECAM02, with some modifications: I have used 1 nm instead of 10 nm spacing, no reduction of brightness towards the grey axis and I used a simple Gaussian blur as smoothing algorithm (this might be enhanced later). Furthermore, I made a gamma correction with power 1.2 to enhance the brightness at the violet and red end. In addition, the spectral intensity had to be weighted with a quite high color temperature to compensate the suppression of blue due to the projection onto the sRGB gamut border. I chose a 2500010000 K blackbody spectrum as a good compromise. The result seems to be promising; I get slightly higher saturation compared to the original image. I see here that the color representation in the thumbnail is not identical to that in the local image viewer, and there are more prominent Mach bands (btw what are the reasons for that? How are PNGs scaled here?). Soo please view the full-scale image on the media viewer to see it correcly. What are the opinions on this result? What can be done to improve this (especially to get an identical color representation)?--
SiriusB (
talk) 17:30, 17 December 2016 (UTC)
Update: I have finally succeeded to create an SVG version (I needed to first create a PDF which I transformed into SVG via Inkscape). Direct SVG output from gnuplot was found corrupted by the Wikimedia upload server.-- SiriusB ( talk) 19:27, 17 December 2016 (UTC)
One more update: I found that different browsers display the colors quite differently; my image and that by gringer look best in e.g. Firefox while the display under Safari (OS-X) is not optimal (somewhat "bumpy" in saturation and brightness). It seems that Safari uses a different RGB space or so for image display. The same is true for the standard document viewer under OS-X (Preview).-- SiriusB ( talk) 20:00, 17 December 2016 (UTC)
Human beings are actually capable of distinguishing UV light that they cannot directly perceive due to fluorescence of material within the eye itself. Because human beings are intelligent enough to deduce the presence of UV, it can be said that people can see UV because when looking at a source of near-UV; it is distinguishable from other sources of light because of the fuzzy/blurry area that will appear to surround it. This is a direct indicator that one is seeing fluorescence in the eye itself, and thus, light that is capable of inducing this fluorescence. Additionally, it seems likely that many other mammals could eventually be trained to make this distinction as well, although other mammals would likely never have any idea of what was actually going on with the photons themselves, etc ... so why is it said that human beings are "blind" to this frequency of light, when we so clearly are not?
Violet needs some discussion as what is a tertiary colour, blue + blue + red link This is violet seen on our pictures, and article pictures. The violet you see through a spectrometer has fluoresent background glow about it. To recreate this effect you could put some fluorent matterial in a camera. The reason you see it as violet, that is some mixing of red, arises from the lift in the spectral tail of the rho cone cell in the eye. helpful diagram. So to repeat: the colour violet shown is not the colour from a spectrometer even though they stimulate the eye in the same way firstly by two, secondly by one wavelength of light - a hue and a spectral colour. The second issue arising from the problematic violet, is our ability to add an unrealistic amount of red to it making it purple and even magenta. Rainbows have these colours, but this is due the addition of red by the underlying Airy Rings. I will try to find an excellent link for everyone from the US NavalLabs. Here is a picture for you. rainbow What I would like to discuss is how to intelligently add this information into this page. I think it is important as links the colour wheel cycle with the linear spectra, and allows the decerning reader to identify unrealistically retouched photos. I will ask how to include pictures and referencs, but in the meantime thanks in advance for any thoughts. Drhillteach ( talk) 13:02, 9 May 2010 (UTC)
No doubt this was discussed and settled ages ago, but what is the reason why the article use the word "cyan" (which to me is a subtractive, printer's color, not a spectral color) and not the time-honored, hallowed, Newton-approved "indigo?" Is there a citation, in the "citation required" sense, to some authority that has decided it is better or more accurate or a recommended practice to use the name "cyan" for the spectral color between "blue" and "violet?" IMHO the article explains the traditional use of the name "indigo" adequately, but does not explain the article's own use of the word "cyan;" I think it should, if only in a footnote. Dpbsmith (talk) 22:31, 28 August 2010 (UTC)
Cyan is in no way an alternative to indigo. Indigo in on the violet side of blue. Cyan is on the green side of blue. Dicklyon ( talk) 04:30, 22 January 2014 (UTC)
The introductory paragraph of this article currently remarks: "The spectrum does not, however, contain all the colors that the human eyes and brain can distinguish. Unsaturated colors such as pink, or purple variations such as magenta, are absent, for example, because they can be made only by a mix of multiple wavelengths."
Although the general idea is correct, I think the wording should be revised to avoid misinformation. Magenta (50% blue, 50% red, 0% green) is a fully saturated color; see http://en.wikipedia.org/wiki/Magenta. Pink can also be a fully saturated color depending what type of pink it is. White might be a better color choice here, although even white/gray could be said to be within the spectrum, namely manifested as black, depending on what kind of spectrum is being referred to. To avoid any kind of confusion, it might be best to reference a color such as 40% blue, 40% red, 20% green. 67.53.36.194 ( talk) 22:46, 16 March 2012 (UTC)
In the lede it says:
I don't think we should define the word "visible" here. Even though the definition provided is a perfectly valid definition, it is not the only possible definition of "visible." Many people who are legally blind have eyes that can detect light. Since "visible" is a common word, it should not require a definition. Also, "Visible spectrum" is a noun phrase that cannot be analyzed entirely in terms of its constituent parts. The temperate zone of the Earth, for example, is actually defined astronomically, and only coincides very roughly with the region in which temperate climate occurs.
Next, the range 390nm - 700nm seems odd. Even though it's referenced (to a textbook) I don't think it's one of the five most commonly cited ranges, or even in the top ten. The 390 at one end suggests a degree of precision that is not matched by the 700 at the other end. I suggest we go back to 400nm - 700nm, but qualify it with words such as "conventional," "nominal," or "customary." For example: "The conventional range of the visible spectrum is 400nm - 700nm. This range is, of course, purely nominal."
After that we could mention a few alternative ranges. For example, the CIE range for photopic (color) vision, 360nm - 830nm, and the CIE range for scotopic (night) vision, 380nm - 780nm. There's also a much narrower range, 420nm - 680nm, that comes up occasionally in technical literature, especially biochemistry. Even though that may seem exceedingly narrow, most people rarely see anything below 420nm or above 680nm, not because we are not capable of seeing it, but because our sensitivity at those wavelengths is so weak that they are drowned out by more visible wavelengths.
And finally, two subjects near and dear to my heart.
1) Nowadays many people have access to exotic light sources such as ultraviolet and infrared lasers, diodes, etc. As a result, many people are seeing light at wavelengths far beyond the more widely published ranges. For example, many people who work with uv lasers in the 320s and 330s can see the Rayleigh scattering. At the opposite end, a $15 5mW 980nm laser pointer shined directly into the eye (reasonably safe for up to 0.25 sec.) is clearly visible. In other words, people are "seeing the impossible."
2) Over the years there has been quite a bit of fundamental research into the extremes of human vision. The upshot is that the actual of range of human vision is about 300nm - 1100nm, but with qualifications. For example, children can usually see down to about 300nm, teens and young adults to 315nm, the elderly usually not below 400nm, the middle aged highly variable. In the low 300s color perception and visual acuity are seriously compromised. The more extreme the wavelength (both in uv and ir) the brighter the light needs to be in order to be visible. Eye damage is a serious concern for wavelengths below 320nm (because of the inherent dangers of uv at those wavelengths) and for brightness reasons above 950nm. Zyxwv99 ( talk) 02:33, 22 September 2013 (UTC)
References
400-700nm
http://www.britannica.com/EBchecked/topic/340440/light
G.K. Pal, Pal, G.K., Orient Blackswan, 2001 - Physiology - 530 pages http://books.google.com/books?id=CcJvIiesqp8C&pg=PA387
Pierre A. Buser, Michel Imbert, MIT Press, 1992 - Medical - 559 pages http://books.google.com/books?id=NSZvt8Ld2-8C&pg=PA50
CIE
http://books.google.com/books?id=DdzBQsqPbzcC&pg=PA5
LIMITS
"In young adults, wavelengths as high as 1000 nm or down to 300 nm may be seen, but the standard range for human vision is typically given as 400-700 nm." Scott E. Umbaugh
http://books.google.com/books?id=UQTMw5uoGHgC&pg=PA405
"Limits of the eye's overall range of sensitivity extends from about 310 to 1050 nanometers, but strong illumination is necessary for sensation at these wavelength extremes."
David K. Lynch, William Charles Livingston, Cambridge University Press, 2001 http://books.google.com/books?id=4Abp5FdhskAC&pg=PA231
ULTRAVIOLET
"Wave length 334 mμ was described as highly unsaturated blue, bluish gray and silver; 313 mμ was given as light without color, almost colorless, gray with a trace of blue..."
Albert Bachem The American Journal of Psychology, Vol. 66, No. 2 (Apr., 1953), pp. 251-260 http://www.jstor.org/discover/10.2307/1418730?uid=3739256&uid=2129&uid=2&uid=70&uid=4&sid=21102663451727
"According to different authors, under appropriate conditions seeing is possible in the ultra-violet down to a wave-length as small as 3100 Å. This fact has been confirmed on 21 persons (age 25–50 years)..." W. de GROOT
Nature 134, 494-494 (29 September 1934) | doi:10.1038/134494a0 http://www.nature.com/nature/journal/v134/n3387/abs/134494a0.html
INFRARED Infrared color reversal
edited by Karl R. Gegenfurtner, Lindsay T. Sharpe Cambridge University Press, May 28, 2001 - Medical - 492 pages http://books.google.com/books?id=4zQMQLLVkFYC&pg=PA93
"The foveal sensitivity to several near-infrared laser wavelengths was measured. It was found that the eye could respond to radiation at wavelengths at least as far as 1064 nm. A continuous 1064 nm laser source appeared red, but a 1060 nm pulsed laser source appeared green, which suggests the presence of second harmonic generation in the retina."
David H. Sliney, Robert T. Wangemann, James K. Franks, and Myron L. Wolbarsht Affiliations JOSA, Vol. 66, Issue 4, pp. 339-341 (1976) http://dx.doi.org/10.1364/JOSA.66.000339
Zyxwv99 ( talk) 23:38, 22 September 2013 (UTC)
This lede states:"A typical human eye will respond to wavelengths from about 390 to 700 nm.[1]" The lede for the article "Light" states:"Under ideal laboratory conditions, people can see infrared up to at least 1050 nm,[8] children and young adults ultraviolet down to about 310 to 313 nm.[9][10][11]" (reference [8] actually states a 1064nm laser was detectable). There are several problems with this lede. First, a "typical" human eye will not only "respond" to visible light, it will "respond" to x-rays and microwaves as well. I'm not sure how to phrase it: 'will see', 'can visualize', 'will visually respond to', 'can sense', but a bald statement about an undefined response is inaccurate. Second, the human eye's sensitivity, especially to UV, diminishes throughout life, hence "typical" isn't very well defined here (and its meaning is unclear, imho). If I recall CIE did define a color space - but that was based on experts in the field (which we should presume biases the result). I've been involved in color matching my entire professional career (35 years), and I've read in numerous texts that the visible spectrum is 430 to 690nm, (Halliday & Resnick (1967) claim that the eye's sensitivity is maximum at 555nm and that it drops to 1% at 430 and 690nm, for a "standard observer"). It seems obvious to me that "typical" eyesight is like "typical" skin color or height or weight. That is, there ain't one. Its a moving target; we all are getting older. Unless there are some useful (global) norms for color vision acuity (?), I think we need to say both that most of us can see colors between 430nm (say) and 680nm under 'typical' viewing conditions and that there are reports of light perception as high as 1064nm, in the lab, and as low as 313nm (I'd like to see another reference confirming that the 1060nm result isn't due to harmonics). Abitslow ( talk) 19:59, 7 June 2014 (UTC)
Color Wavelength Frequency Photon energy violet 380–450 nm 668–789 THz 2.75–3.26 eV blue 450–495 nm 606–668 THz 2.50–2.75 eV green 495–570 nm 526–606 THz 2.17–2.50 eV yellow 570–590 nm 508–526 THz 2.10–2.17 eV orange 590–620 nm 484–508 THz 2.00–2.10 eV red 620–750 nm 400–484 THz 1.65–2.00 eV
should be changed to
Color Wavelength Frequency Photon energy violet 380–450 nm 789-668 THz 2.75–3.26 eV blue 450–495 nm 668-606 THz 2.50–2.75 eV green 495–570 nm 606-526 THz 2.17–2.50 eV yellow 570–590 nm 526-508 THz 2.10–2.17 eV orange 590–620 nm 508-484 THz 2.00–2.10 eV red 620–750 nm 484-400 THz 1.65–2.00 eV — Preceding unsigned comment added by Tempedi ( talk • contribs) 15:57, 31 May 2017 (UTC)
The section "Animal color vision" makes this statement about snakes: "other snakes with the organ may detect warm bodies from a meter away." However, no "organ" is mentioned in this section, so it is unclear what is meant here by "the organ." Presumably there was something here previously about which organ allows a snake to perceive radiant heat, but no such reference is present now. This statement needs to be made clearer. — Preceding unsigned comment added by 2601:602:8480:3343:2145:CFCE:C4CA:A507 ( talk) 05:03, 4 February 2019 (UTC)
I don't know if it is already mentioned in this quite large page, but... I want to say that when the spectrum from a sunlit prism (corners 60°) is observed at a large distance (the distance prism - projection screen) then one shall notice the absence of the colors orange, yellow, yellowish green, cyan (greenish blue), and indigo. There's only the three colors red, green, and ultramarine blue (the same colors from the RGB color model). So... Sir Isaac Newton's color system (red, orange, yellow, green, blue, indigo, violet) is a bit childish or schoolish. Johann Wolfgang von Goethe also knew about this rather childish approach of Newton and noticed Newton's wrong observations after he (Goethe) performed his own experiments with sunlit prism. Moreover, there's no violet in the spectrum of white light! Sir Isaac Newton observed the Primary Rainbow but he didn't knew about the existence of the Supernumerary Arcs on the "inside" (on the blue part) of the bow. He noticed a color that he called violet, but... what he really observed was the "overlap" of the ultramarine blue from the Primary Rainbow with the red from the first Supernumerary Arc. This "overlap" is the color Magenta (deep pink). Newton must have been unaware of the existence of the "overlap-color" Magenta, thus... he thought he observed the color violet. DannyJ.Caes ( talk) 07:25, 5 August 2019 (UTC)
The chart is incorrect at the end of spectrum: the atmosphere is quite transparent to the long waves. — Preceding unsigned comment added by 2003:DF:2812:FF05:EEF4:BBFF:FE36:3F5C ( talk) 07:42, 25 September 2020 (UTC)
While reorganising the article Optical window, I noticed that although this atmospheric window is characterised as optical, its range also covers the UV and infrared spectra. After some research I found that many valid sources differentiate between the visible and optical spectra, defining the visible spectrum as the one the human eye can detect and the optical spectrum as the one including the UV, the visible and the infrared spectra (see for example Frank L. Pedrotti's "Introduction to optics", Cambridge University Press, 2017, pp. 7-8). It seems that at some point in scientific history the terms optical and visible coincided, something that might hold true for many contemporary scientific fields as well. However, the matter should be looked into and this article should probably not present the terms visible spectrum and optical spectrum as synonymous.-- L'OrfeoGreco ( talk) 04:07, 28 December 2021 (UTC)
The only reference for this section is a page that describes the purpose of gamma correction and uses the gray projection thing as an illustration of part of the reasoning behind it, and the section gets that wrong. I propose someone just throw a link to a relevant RGB color explanation page in place of the entire section and be done with it. I'm not doing it because this talk page appears to be the most insane bitchfest on Wikipedia over a tiny article I've seen to date. A Shortfall Of Gravitas ( talk) 18:57, 25 April 2022 (UTC)
For color-accurate reproduction, a spectrum can be projected onto a uniform gray field. The resulting mixed colors can have all their R, G, B coordinates non-negative, and so can be reproduced without distortion. This accurately simulates looking at a spectrum on a gray background. [1]
References
This article is difficult to disambiguate from light, especially when there is also electromagnetic radiation. There is also lots of overlap with spectral colors, but I think that is easy enough to separate. Separating from light is harder, but as I've gleaned from talk:light and previous discussions here, light should head more in a physics/physical direction and visible spectrum should focus more on the biology/evolution/visual side of things. I've gone with that and expanded greatly on the basis of the visible range while removing some of the physics based information that belong in light or the color information that belongs in one of a hundred color based articles like gamut. Discussions of color doesn't have much place in this article, imho, and maybe I'll even pare #history down to suit. Curran919 ( talk) 14:54, 29 August 2023 (UTC)
I don't know about this one. Having 2 similar articles and disambiguating them is hard enough so most people just merge them together, but this is a different case. Visible spectrum is just a fancier word for visible light, so they should be merged, right? Well, this one's gonna sound off, but hear me out.
"Light" should be an article about all types of light, not just visible light. Instead, visible light should either be its own article or merged with visible spectrum. Besides, both article's first sentence is basically the same wording but tweaked to suit their different names. This just makes it clear even further that visible light and visible spectrum is clearly almost identical in every aspect because their leads read nearly the same.
So what do you think? Is it ok? 2001:44C8:40B2:3790:8876:61FA:ACAC:B779 ( talk) 17:10, 24 January 2024 (UTC)
This
level-4 vital article is rated B-class on Wikipedia's
content assessment scale. It is of interest to the following WikiProjects: | ||||||||||||||||||||||||||||
|
|
Is there any article on Audible spectrum? Anwar ( talk) 19:35, 3 May 2008 (UTC)
Isn't the exact range of the visible spectrum disputed?--Luke Elms 20:33, 17 May 2008 (UTC)
Human eyes do not see wavelengths, they see frequencies. The color of light perceived through water, for example, is not massively affected by the drastic decrease in wavelength because the frequency is about the same. — Preceding unsigned comment added by 129.210.115.18 ( talk) 21:50, 18 June 2013 (UTC)
Other sources (many, I think) indicate not the round numbers "400 to 700," as in this article, but 380 nm to 750 nm or more. as in the electromagnetic spectrum article (the citation of Biology: Concepts and Applications looks tertiary at best). Yellowing on of the lens of the eye due to aging causes loss of sensitivity at the short wavelength end. I recommend that someone correct the numbers in the article and propagate the changes throughout. Eric Drexler ( talk) 12:33, 17 March 2013 (UTC)
Here are authoritative sources that present quantitative data re. human eye sensitivity vs. wavelength (and age of subject, at the violet end):
The article should be corrected, and perhaps other articles as well. There’s a lot of misinformation in circulation. Eric Drexler ( talk) 12:58, 17 March 2013 (UTC)
Saw a short article a while ago in a science news magazine that Birds see more of the spectrum than we do. They have additional cells in their retina that also have a tiny bubble of oil over them.
The speculation was that the non-bird animals or proto-mammals that survived the extinction of the dinosaurs (the great K-T Extinction) were almost exclusively nocturnal types and did not need the extra part of the spectrum anyway. I got the impression that this part of the extension was not at either end of our visible spectrum, but somewhere nearer the center. I am not sure this makes sense, but maybe some can look into it. WonderWheeler ( talk) 23:20, 20 August 2008 (UTC)
Birds are tetrachromats and can see down to 320 nanometers in the ultraviolet, whereas since we are trichromats, our violet limit is 380 nanometers. Pigeons are pentachromats and besides seeing into the ultraviolet, can also see into the infrared beyond 750 nanometers. Keraunos ( talk) 05:20, 22 March 2011 (UTC)
While all the notes about other species are interesting, I think that the tendency (especially of new editors) to simply chime in on the article with unsourced factoids about this or that species' abilities needs some kind of control. After all, the article begins with the true statement that the article is about what the human eye sees. The introductory paragraph that diverges onto the topic of other species is not only slightly off topic, it also begins to look like it was designed by committee. Which of course it was! Ehusman ( talk) 21:00, 24 March 2013 (UTC)
I've made an image of the visible spectrum in Inkscape (a rectangle with a linear gradient having various transition points), which got pulled out of this article rather quickly because it looked odd. My aim is/was to try to make it represent the spectrum using transition points that are as precise as possible and fit the RGB colours represented on the respective colour pages ( red, orange, yellow, green, blue, violet). Spectral violet is obviously a tricky one, not actually being in the RGB Gamut. I've already reduced the luminosity of a few colours to make it blend better -- any ideas on how to improve it? Perhaps a logarithmic representation would be better? gringer ( talk) 03:14, 28 August 2008 (UTC)
Here is a logarithmic representation, with wavelengths at the boundaries marked. gringer ( talk) 04:11, 28 August 2008 (UTC)
I uploaded an inverse spectrum (linear by wavelength). Mach bands seem to be a bit more noticeable in this one. I've since realised that doing the whole "make it look like a prism" thing is silly, because dispersion is probably not going to be a nice function of wavelength, and is material-dependent. gringer ( talk) 03:54, 30 August 2008 (UTC)
I believe CIE publishes tables of 2° and 10° standard observer values for the spectrum, as XYZ or xyY coordinates. I’m not sure how to track such a table down though. Dicklyon, do any of your color science books have something like that at the back? — jacobolus (t) 06:14, 30 August 2008 (UTC)
This seems pretty reasonable, and ends up with several renderings to choose from. — jacobolus (t) 12:30, 30 August 2008 (UTC)
Skoch3 ( talk) 07:50, 3 November 2008 (UTC): I changed the text for spectral colors (and included a link). I took out the reference to rainbows, because rainbows have significant overlap of colors with each other, and thus the colors are not spectral.
Skoch3 (
talk) 07:56, 3 November 2008 (UTC): While I can see that a significant amount of work has been put into this image of the spectrum, I think as is it is very misleading. As far as simulations go, it is quite different from the image on the
spectral color page (the spectral colors are on the outside horseshoe...and are quite different from the current image on this section). However, I think much better than a simulation would be a good photo of a white light spectrum, spread out by a diffraction grating, along with wavelength calibration. I am thinking of switching to add the image to the right, above. The most striking thing missing to me is the cyan color, which is lacking in the current simulated image. I agree that the NASA image is a bit dark.
The actual data spectrum from NASA represents a particular source, whereas the brighter illustration is about the general concept of visible sprectrum. The latter works better as an illustration, I think. It's hard to make good representations of pure spectral colors in RGB; the chromaticity plot has other goals, being normalized by x+y+z such that it doesn't go black at the ends of the spectrum, and also shows purples. The existing image is also not great, having poor yellow and cyan. Dicklyon ( talk) 20:54, 3 November 2008 (UTC)
"In the illustration, the narrow red, green and blue bars show the relative mixture of these three colors used to produce the color directly above.
what illustration? 12.125.134.114 ( talk) 16:22, 8 October 2008 (UTC)
I changed the name from visble spectrum to visible light spectrum -- Chicagobears94 ( talk) 21:21, 10 January 2009 (UTC)
Actually, you moved it to Visible Light Spectrum, which anyone who has been editting wikipedia should know is improper capitalization. I moved it back. If you think that Visible light spectrum would be better than Visible spectrum, make a proposal here and we can talk about it. You're not the only one likely to have an opinion on this. Dicklyon ( talk) 06:39, 11 January 2009 (UTC)
Why do we have this article and also have the Light article whose first line says: "Light, or visible light, is electromagnetic radiation of a wavelength that is visible". Also, we have Visible Light redirect here. -- MarsRover ( talk) 19:54, 21 March 2009 (UTC)
I had previously attempted to merge light with electromagnetic radiation, which was rejected, but the discussion appeared to suggest that the best way of consolidating the information was to perform a merge or partial merge between visible spectrum and light, called either light or visible light. I was wondering how this might be achieved. Serendi pod ous 15:27, 5 October 2009 (UTC)
Re: removal of indigo from the table. I approve. If I had noticed that the table was drawn from a source I would have removed it too.
The idea that there is a distinct colour "indigo" in the spectrum between blue and violet has no basis in objective fact nor even unbiased qualitative observation. It was added because of Newton's bias that the number of colours should be the same as the number of notes in the musical scale.-- Srleffler ( talk) 23:29, 4 December 2009 (UTC)
Capabilities of one's vision seems to be dependent on their growth environment, culture and perhaps even language. Therefore talking about distinct colors is highly subjective matter. See "BBC Horizon: Do you see what I see?" or this YouTube clip: http://www.youtube.com/watch?v=4b71rT9fU-I . — Preceding unsigned comment added by 80.75.107.170 ( talk) 09:09, 16 May 2012 (UTC)
The image at the top of the page, showing a narrow beam of white light passing through a prism, is quite inaccurate and needs replacing. The problem is that the beam as it is shown inside the prism, after having passed through the left-side wall, is still a narrow beam of white light when it should be already diverging into color separations as it has been refracted once. Other images on the web get this part correct.
TimProof ( talk) 20:51, 18 March 2010 (UTC)
Hi everyone. There are currently a ton of spectrum images spread through English wikipedia & the rest of wikimedia projects. I’ve stuck the ones I could find over to the right (this doesn’t include the vertically oriented ones, or the couple that were flipped right-to-left, or scaled logarithmically, but otherwise identical to one shown here). None of them seems to me to have an especially legitimate methodology behind its construction, and most don’t even explain how they were generated. This is understandable enough – there’s no clear right approach to rendering something that falls so far outside the sRGB gamut. Still, I’ve been playing with it, and the best rendering I can come up with results from the steps: (1) Use the CIE 1931 standard observer data to define XYZ at each wavelength; (2) Convert each XYZ triplet to CIECAM02 space, assuming D65 white point and “average” surround; (3) Apply a bit of a gamma curve to the resulting J values, to brighten the darkest parts so they’re a bit easier to see; (4) Compress this gamma-adjusted J to fall between 8 (the sRGB black point) and 75 – i.e. map [0, 100] onto [8, 75] – so that we can get a decent amount of chroma at every point; (5) For each (new) J value, and each h, take the point within the sRGB gamut with maximum possible chroma. Some of these steps are somewhat arbitrary, but the key parts of this method are (A) perceived hues are preserved, and (B) The relative lightness is preserved, with just enough fudging to make things look somewhat reasonable.
Anyway, I’m going to keep fiddling a bit, and then upload the result I get. It might be worth even ending up with two different renders by different methods (in particular, it might be worth ignoring or partially ignoring relative lightness, and just taking the most colorful possible color for each hue, so we can have a bright yellow, a bright cyan, etc.), so that readers can compare them. Having 6 or 10 though, as we do now, seems sort of absurd. We should try to get some kind of consensus about color content, and then about what labels/etc. are necessary, and then just put up a couple of these, in SVG format, and recommend (on all the image description pages) that articles use those.
Cheers,
jacobolus
(t) 20:10, 6 April 2010 (UTC)
if (abs(sin(hr)) >= abs(cos(hr))) { p4 = p1 / sin(hr); cb = (p2 * (2.0 + p3) * (460.0 / 1403.0)) / (p4 + (2.0 + p3) * (220.0 / 1403.0) * (cos(hr) / sin(hr)) - (27.0 / 1403.0) + p3 * (6300.0 / 1403.0)); ca = cb * (cos(hr) / sin(hr)); } else { p5 = p1 / cos(hr); ca = (p2 * (2.0 + p3) * (460.0 / 1403.0)) / (p5 + (2.0 + p3) * (220.0 / 1403.0) - ((27.0 / 1403.0) - p3 * (6300.0 / 1403.0)) * (sin(hr) / cos(hr))); cb = ca * (sin(hr) / cos(hr)); }
=if((H94=0),0,if(abs(sin(H83*pi()/180))>=abs(cos(H83*pi()/180)),(((H90/H48)+0.305)*(2+(21/20))*(460/1403)) /(((H89/H94)/sin(H83*pi()/180))+(2+(21/20))*(220/1403)*(cos(H83*pi()/180)/sin(H83*pi()/180))-(27/1403)+ (21/20)*(6300/1403))*(cos(H83*pi()/180)/sin(H83*pi()/180)),(((H90/H48)+0.305)*(2+(21/20))*(460/1403))/(((H89 /H94)/cos(H83*pi()/180))+(2+(21/20))*(220/1403)-((27/1403)-(21/20)*(6300/1403))*(sin(H83*pi() /180)/cos(H83*pi()/180)))))
I have taken an independent approach to render a visible spectrum. Currently PNG only since
gnuplot has difficulties to create proper SVG images (or I didn't find out the right way). The method is similar to that of
gringer, i.e. a binary search along a constant-hue line in CIECAM02, with some modifications: I have used 1 nm instead of 10 nm spacing, no reduction of brightness towards the grey axis and I used a simple Gaussian blur as smoothing algorithm (this might be enhanced later). Furthermore, I made a gamma correction with power 1.2 to enhance the brightness at the violet and red end. In addition, the spectral intensity had to be weighted with a quite high color temperature to compensate the suppression of blue due to the projection onto the sRGB gamut border. I chose a 2500010000 K blackbody spectrum as a good compromise. The result seems to be promising; I get slightly higher saturation compared to the original image. I see here that the color representation in the thumbnail is not identical to that in the local image viewer, and there are more prominent Mach bands (btw what are the reasons for that? How are PNGs scaled here?). Soo please view the full-scale image on the media viewer to see it correcly. What are the opinions on this result? What can be done to improve this (especially to get an identical color representation)?--
SiriusB (
talk) 17:30, 17 December 2016 (UTC)
Update: I have finally succeeded to create an SVG version (I needed to first create a PDF which I transformed into SVG via Inkscape). Direct SVG output from gnuplot was found corrupted by the Wikimedia upload server.-- SiriusB ( talk) 19:27, 17 December 2016 (UTC)
One more update: I found that different browsers display the colors quite differently; my image and that by gringer look best in e.g. Firefox while the display under Safari (OS-X) is not optimal (somewhat "bumpy" in saturation and brightness). It seems that Safari uses a different RGB space or so for image display. The same is true for the standard document viewer under OS-X (Preview).-- SiriusB ( talk) 20:00, 17 December 2016 (UTC)
Human beings are actually capable of distinguishing UV light that they cannot directly perceive due to fluorescence of material within the eye itself. Because human beings are intelligent enough to deduce the presence of UV, it can be said that people can see UV because when looking at a source of near-UV; it is distinguishable from other sources of light because of the fuzzy/blurry area that will appear to surround it. This is a direct indicator that one is seeing fluorescence in the eye itself, and thus, light that is capable of inducing this fluorescence. Additionally, it seems likely that many other mammals could eventually be trained to make this distinction as well, although other mammals would likely never have any idea of what was actually going on with the photons themselves, etc ... so why is it said that human beings are "blind" to this frequency of light, when we so clearly are not?
Violet needs some discussion as what is a tertiary colour, blue + blue + red link This is violet seen on our pictures, and article pictures. The violet you see through a spectrometer has fluoresent background glow about it. To recreate this effect you could put some fluorent matterial in a camera. The reason you see it as violet, that is some mixing of red, arises from the lift in the spectral tail of the rho cone cell in the eye. helpful diagram. So to repeat: the colour violet shown is not the colour from a spectrometer even though they stimulate the eye in the same way firstly by two, secondly by one wavelength of light - a hue and a spectral colour. The second issue arising from the problematic violet, is our ability to add an unrealistic amount of red to it making it purple and even magenta. Rainbows have these colours, but this is due the addition of red by the underlying Airy Rings. I will try to find an excellent link for everyone from the US NavalLabs. Here is a picture for you. rainbow What I would like to discuss is how to intelligently add this information into this page. I think it is important as links the colour wheel cycle with the linear spectra, and allows the decerning reader to identify unrealistically retouched photos. I will ask how to include pictures and referencs, but in the meantime thanks in advance for any thoughts. Drhillteach ( talk) 13:02, 9 May 2010 (UTC)
No doubt this was discussed and settled ages ago, but what is the reason why the article use the word "cyan" (which to me is a subtractive, printer's color, not a spectral color) and not the time-honored, hallowed, Newton-approved "indigo?" Is there a citation, in the "citation required" sense, to some authority that has decided it is better or more accurate or a recommended practice to use the name "cyan" for the spectral color between "blue" and "violet?" IMHO the article explains the traditional use of the name "indigo" adequately, but does not explain the article's own use of the word "cyan;" I think it should, if only in a footnote. Dpbsmith (talk) 22:31, 28 August 2010 (UTC)
Cyan is in no way an alternative to indigo. Indigo in on the violet side of blue. Cyan is on the green side of blue. Dicklyon ( talk) 04:30, 22 January 2014 (UTC)
The introductory paragraph of this article currently remarks: "The spectrum does not, however, contain all the colors that the human eyes and brain can distinguish. Unsaturated colors such as pink, or purple variations such as magenta, are absent, for example, because they can be made only by a mix of multiple wavelengths."
Although the general idea is correct, I think the wording should be revised to avoid misinformation. Magenta (50% blue, 50% red, 0% green) is a fully saturated color; see http://en.wikipedia.org/wiki/Magenta. Pink can also be a fully saturated color depending what type of pink it is. White might be a better color choice here, although even white/gray could be said to be within the spectrum, namely manifested as black, depending on what kind of spectrum is being referred to. To avoid any kind of confusion, it might be best to reference a color such as 40% blue, 40% red, 20% green. 67.53.36.194 ( talk) 22:46, 16 March 2012 (UTC)
In the lede it says:
I don't think we should define the word "visible" here. Even though the definition provided is a perfectly valid definition, it is not the only possible definition of "visible." Many people who are legally blind have eyes that can detect light. Since "visible" is a common word, it should not require a definition. Also, "Visible spectrum" is a noun phrase that cannot be analyzed entirely in terms of its constituent parts. The temperate zone of the Earth, for example, is actually defined astronomically, and only coincides very roughly with the region in which temperate climate occurs.
Next, the range 390nm - 700nm seems odd. Even though it's referenced (to a textbook) I don't think it's one of the five most commonly cited ranges, or even in the top ten. The 390 at one end suggests a degree of precision that is not matched by the 700 at the other end. I suggest we go back to 400nm - 700nm, but qualify it with words such as "conventional," "nominal," or "customary." For example: "The conventional range of the visible spectrum is 400nm - 700nm. This range is, of course, purely nominal."
After that we could mention a few alternative ranges. For example, the CIE range for photopic (color) vision, 360nm - 830nm, and the CIE range for scotopic (night) vision, 380nm - 780nm. There's also a much narrower range, 420nm - 680nm, that comes up occasionally in technical literature, especially biochemistry. Even though that may seem exceedingly narrow, most people rarely see anything below 420nm or above 680nm, not because we are not capable of seeing it, but because our sensitivity at those wavelengths is so weak that they are drowned out by more visible wavelengths.
And finally, two subjects near and dear to my heart.
1) Nowadays many people have access to exotic light sources such as ultraviolet and infrared lasers, diodes, etc. As a result, many people are seeing light at wavelengths far beyond the more widely published ranges. For example, many people who work with uv lasers in the 320s and 330s can see the Rayleigh scattering. At the opposite end, a $15 5mW 980nm laser pointer shined directly into the eye (reasonably safe for up to 0.25 sec.) is clearly visible. In other words, people are "seeing the impossible."
2) Over the years there has been quite a bit of fundamental research into the extremes of human vision. The upshot is that the actual of range of human vision is about 300nm - 1100nm, but with qualifications. For example, children can usually see down to about 300nm, teens and young adults to 315nm, the elderly usually not below 400nm, the middle aged highly variable. In the low 300s color perception and visual acuity are seriously compromised. The more extreme the wavelength (both in uv and ir) the brighter the light needs to be in order to be visible. Eye damage is a serious concern for wavelengths below 320nm (because of the inherent dangers of uv at those wavelengths) and for brightness reasons above 950nm. Zyxwv99 ( talk) 02:33, 22 September 2013 (UTC)
References
400-700nm
http://www.britannica.com/EBchecked/topic/340440/light
G.K. Pal, Pal, G.K., Orient Blackswan, 2001 - Physiology - 530 pages http://books.google.com/books?id=CcJvIiesqp8C&pg=PA387
Pierre A. Buser, Michel Imbert, MIT Press, 1992 - Medical - 559 pages http://books.google.com/books?id=NSZvt8Ld2-8C&pg=PA50
CIE
http://books.google.com/books?id=DdzBQsqPbzcC&pg=PA5
LIMITS
"In young adults, wavelengths as high as 1000 nm or down to 300 nm may be seen, but the standard range for human vision is typically given as 400-700 nm." Scott E. Umbaugh
http://books.google.com/books?id=UQTMw5uoGHgC&pg=PA405
"Limits of the eye's overall range of sensitivity extends from about 310 to 1050 nanometers, but strong illumination is necessary for sensation at these wavelength extremes."
David K. Lynch, William Charles Livingston, Cambridge University Press, 2001 http://books.google.com/books?id=4Abp5FdhskAC&pg=PA231
ULTRAVIOLET
"Wave length 334 mμ was described as highly unsaturated blue, bluish gray and silver; 313 mμ was given as light without color, almost colorless, gray with a trace of blue..."
Albert Bachem The American Journal of Psychology, Vol. 66, No. 2 (Apr., 1953), pp. 251-260 http://www.jstor.org/discover/10.2307/1418730?uid=3739256&uid=2129&uid=2&uid=70&uid=4&sid=21102663451727
"According to different authors, under appropriate conditions seeing is possible in the ultra-violet down to a wave-length as small as 3100 Å. This fact has been confirmed on 21 persons (age 25–50 years)..." W. de GROOT
Nature 134, 494-494 (29 September 1934) | doi:10.1038/134494a0 http://www.nature.com/nature/journal/v134/n3387/abs/134494a0.html
INFRARED Infrared color reversal
edited by Karl R. Gegenfurtner, Lindsay T. Sharpe Cambridge University Press, May 28, 2001 - Medical - 492 pages http://books.google.com/books?id=4zQMQLLVkFYC&pg=PA93
"The foveal sensitivity to several near-infrared laser wavelengths was measured. It was found that the eye could respond to radiation at wavelengths at least as far as 1064 nm. A continuous 1064 nm laser source appeared red, but a 1060 nm pulsed laser source appeared green, which suggests the presence of second harmonic generation in the retina."
David H. Sliney, Robert T. Wangemann, James K. Franks, and Myron L. Wolbarsht Affiliations JOSA, Vol. 66, Issue 4, pp. 339-341 (1976) http://dx.doi.org/10.1364/JOSA.66.000339
Zyxwv99 ( talk) 23:38, 22 September 2013 (UTC)
This lede states:"A typical human eye will respond to wavelengths from about 390 to 700 nm.[1]" The lede for the article "Light" states:"Under ideal laboratory conditions, people can see infrared up to at least 1050 nm,[8] children and young adults ultraviolet down to about 310 to 313 nm.[9][10][11]" (reference [8] actually states a 1064nm laser was detectable). There are several problems with this lede. First, a "typical" human eye will not only "respond" to visible light, it will "respond" to x-rays and microwaves as well. I'm not sure how to phrase it: 'will see', 'can visualize', 'will visually respond to', 'can sense', but a bald statement about an undefined response is inaccurate. Second, the human eye's sensitivity, especially to UV, diminishes throughout life, hence "typical" isn't very well defined here (and its meaning is unclear, imho). If I recall CIE did define a color space - but that was based on experts in the field (which we should presume biases the result). I've been involved in color matching my entire professional career (35 years), and I've read in numerous texts that the visible spectrum is 430 to 690nm, (Halliday & Resnick (1967) claim that the eye's sensitivity is maximum at 555nm and that it drops to 1% at 430 and 690nm, for a "standard observer"). It seems obvious to me that "typical" eyesight is like "typical" skin color or height or weight. That is, there ain't one. Its a moving target; we all are getting older. Unless there are some useful (global) norms for color vision acuity (?), I think we need to say both that most of us can see colors between 430nm (say) and 680nm under 'typical' viewing conditions and that there are reports of light perception as high as 1064nm, in the lab, and as low as 313nm (I'd like to see another reference confirming that the 1060nm result isn't due to harmonics). Abitslow ( talk) 19:59, 7 June 2014 (UTC)
Color Wavelength Frequency Photon energy violet 380–450 nm 668–789 THz 2.75–3.26 eV blue 450–495 nm 606–668 THz 2.50–2.75 eV green 495–570 nm 526–606 THz 2.17–2.50 eV yellow 570–590 nm 508–526 THz 2.10–2.17 eV orange 590–620 nm 484–508 THz 2.00–2.10 eV red 620–750 nm 400–484 THz 1.65–2.00 eV
should be changed to
Color Wavelength Frequency Photon energy violet 380–450 nm 789-668 THz 2.75–3.26 eV blue 450–495 nm 668-606 THz 2.50–2.75 eV green 495–570 nm 606-526 THz 2.17–2.50 eV yellow 570–590 nm 526-508 THz 2.10–2.17 eV orange 590–620 nm 508-484 THz 2.00–2.10 eV red 620–750 nm 484-400 THz 1.65–2.00 eV — Preceding unsigned comment added by Tempedi ( talk • contribs) 15:57, 31 May 2017 (UTC)
The section "Animal color vision" makes this statement about snakes: "other snakes with the organ may detect warm bodies from a meter away." However, no "organ" is mentioned in this section, so it is unclear what is meant here by "the organ." Presumably there was something here previously about which organ allows a snake to perceive radiant heat, but no such reference is present now. This statement needs to be made clearer. — Preceding unsigned comment added by 2601:602:8480:3343:2145:CFCE:C4CA:A507 ( talk) 05:03, 4 February 2019 (UTC)
I don't know if it is already mentioned in this quite large page, but... I want to say that when the spectrum from a sunlit prism (corners 60°) is observed at a large distance (the distance prism - projection screen) then one shall notice the absence of the colors orange, yellow, yellowish green, cyan (greenish blue), and indigo. There's only the three colors red, green, and ultramarine blue (the same colors from the RGB color model). So... Sir Isaac Newton's color system (red, orange, yellow, green, blue, indigo, violet) is a bit childish or schoolish. Johann Wolfgang von Goethe also knew about this rather childish approach of Newton and noticed Newton's wrong observations after he (Goethe) performed his own experiments with sunlit prism. Moreover, there's no violet in the spectrum of white light! Sir Isaac Newton observed the Primary Rainbow but he didn't knew about the existence of the Supernumerary Arcs on the "inside" (on the blue part) of the bow. He noticed a color that he called violet, but... what he really observed was the "overlap" of the ultramarine blue from the Primary Rainbow with the red from the first Supernumerary Arc. This "overlap" is the color Magenta (deep pink). Newton must have been unaware of the existence of the "overlap-color" Magenta, thus... he thought he observed the color violet. DannyJ.Caes ( talk) 07:25, 5 August 2019 (UTC)
The chart is incorrect at the end of spectrum: the atmosphere is quite transparent to the long waves. — Preceding unsigned comment added by 2003:DF:2812:FF05:EEF4:BBFF:FE36:3F5C ( talk) 07:42, 25 September 2020 (UTC)
While reorganising the article Optical window, I noticed that although this atmospheric window is characterised as optical, its range also covers the UV and infrared spectra. After some research I found that many valid sources differentiate between the visible and optical spectra, defining the visible spectrum as the one the human eye can detect and the optical spectrum as the one including the UV, the visible and the infrared spectra (see for example Frank L. Pedrotti's "Introduction to optics", Cambridge University Press, 2017, pp. 7-8). It seems that at some point in scientific history the terms optical and visible coincided, something that might hold true for many contemporary scientific fields as well. However, the matter should be looked into and this article should probably not present the terms visible spectrum and optical spectrum as synonymous.-- L'OrfeoGreco ( talk) 04:07, 28 December 2021 (UTC)
The only reference for this section is a page that describes the purpose of gamma correction and uses the gray projection thing as an illustration of part of the reasoning behind it, and the section gets that wrong. I propose someone just throw a link to a relevant RGB color explanation page in place of the entire section and be done with it. I'm not doing it because this talk page appears to be the most insane bitchfest on Wikipedia over a tiny article I've seen to date. A Shortfall Of Gravitas ( talk) 18:57, 25 April 2022 (UTC)
For color-accurate reproduction, a spectrum can be projected onto a uniform gray field. The resulting mixed colors can have all their R, G, B coordinates non-negative, and so can be reproduced without distortion. This accurately simulates looking at a spectrum on a gray background. [1]
References
This article is difficult to disambiguate from light, especially when there is also electromagnetic radiation. There is also lots of overlap with spectral colors, but I think that is easy enough to separate. Separating from light is harder, but as I've gleaned from talk:light and previous discussions here, light should head more in a physics/physical direction and visible spectrum should focus more on the biology/evolution/visual side of things. I've gone with that and expanded greatly on the basis of the visible range while removing some of the physics based information that belong in light or the color information that belongs in one of a hundred color based articles like gamut. Discussions of color doesn't have much place in this article, imho, and maybe I'll even pare #history down to suit. Curran919 ( talk) 14:54, 29 August 2023 (UTC)
I don't know about this one. Having 2 similar articles and disambiguating them is hard enough so most people just merge them together, but this is a different case. Visible spectrum is just a fancier word for visible light, so they should be merged, right? Well, this one's gonna sound off, but hear me out.
"Light" should be an article about all types of light, not just visible light. Instead, visible light should either be its own article or merged with visible spectrum. Besides, both article's first sentence is basically the same wording but tweaked to suit their different names. This just makes it clear even further that visible light and visible spectrum is clearly almost identical in every aspect because their leads read nearly the same.
So what do you think? Is it ok? 2001:44C8:40B2:3790:8876:61FA:ACAC:B779 ( talk) 17:10, 24 January 2024 (UTC)