Is it yellow, green or yellow green color to which human eye is most sensitive?

[answers]

Hello first post here!

I can't find an answer to what hue of green/yellow exactly is the human eye most senstivite to (in RGB).

 

These are the most informative articles I could find:

http://en.wikipedia....n_the_human_eye

http://en.wikipedia.org/wiki/Cone_cell

 

This could be useful for stuff like enemy color in FPS games that allow brightskins for enemies, etc.

 

 

#edit:

http://www.powertech...sensitivity.pdf

 

 

Sensitivity in daylight vs. darkness

Our eyes are most sensitive in daylight to green light around 550nm and slightly less sensitive to yellow light.

They are approximately half as sensitive to orange light and only about a tenth as sensitive to red and violet

light. Therefore, for example, we can see a 100-Watt green light bulb three times farther away than we can a

red or violet bulb of the same power. A 532nm Nd:YAG laser diode module will emit highly visible green

output and is, therefore, a popular choice for outdoor pointing and alignment applications. In darkness,

however, we are most sensitive to blues and least sensitive to reds.

 

 

This is interesting. Does it mean that if I play quake in dark settings then I should set my enemies to blue color. But again -- which hue of blue?

Edited April 5, 2011 by answers

[swansont]

Luminous flux/intensity are defined at ~555 nm (540 THz), because that's the peak sensitivity of photopic scotopic vision. Photopic Scotopic vision shifts to the blue about 50 nm.

http://en.wikipedia.org/wiki/File:Luminosity.png

Edited April 5, 2011 by swansont
fix error

[SMF]

Swansont. I think you have reversed scotopic (rod, low light) with photopic (cone, bright) vision. Photopic is 555nm and scotopic is 498nm. SM

[CaptainPanic]

Although I risk derailing the actual topic now: isn't contrast more important than the actual sensitivity of the eye? Since we're talking about gaming, shouldn't this be about how enemies can stand out the most in different settings?

 

My humble apologies if I take this discussion in the wrong direction.

[answers]

Luminous flux/intensity are defined at ~555 nm (540 THz), because that's the peak sensitivity of scotopic vision. Photopic vision shifts to the blue about 50 nm.

http://en.wikipedia....:Luminosity.png

 

 

What is the value in RGB? 0 255 0 ?

 

Although I risk derailing the actual topic now: isn't contrast more important than the actual sensitivity of the eye? Since we're talking about gaming, shouldn't this be about how enemies can stand out the most in different settings?

 

My humble apologies if I take this discussion in the wrong direction.

 

Could the answer be "optic yellow" that was concluded to be most visible tennis ball color for TV viewers in one scientific research?

 

typical environments in the game:

 

 

 

[swansont]

Swansont. I think you have reversed scotopic (rod, low light) with photopic (cone, bright) vision. Photopic is 555nm and scotopic is 498nm. SM

 

Yes, I did. Thanks.

[thinker_jeff]

CaptainPanic is in the correct diretion for this topic. OP needs a better contrast setting in your game design, not better sensitivity to the visual object.

Generally there are two kinds of contrast: the contrast of brightness and the contrast of color.

The contrast of brightness depends on the ratio of brightness between visual object and background. For a light source with pulse wavelength, ~555 nm is the most sensitive to your eyes in brightness, which means that the object with this wavelength in dark background should have the highest contrast subjectively.

The contrast of color depends on the wavelength difference between the object and background. The human eyes are more sensitive for the difference between RED and GREEN.

If you have a mixed color on the background, the best contrast for the object should be some color between ~555 nm and red. That's why "optical yellow" is good color for tennis ball on TV.

[John Cuthber]

My computer monitor has three phosphors which emit red, green and blue.

I can do what I like with the software, but it will never emit yellow light.

What you need is not data on the sensitivity of the eye but data on visible contrast on PC screens. That will probably vary with make, condition, settings etc.

Good luck.

[thinker_jeff]

My computer monitor has three phosphors which emit red, green and blue.

I can do what I like with the software, but it will never emit yellow light.

 

God bless your monitor!

[SMF]

Because color is based on subjective experience, and has only a passing relationship to light frequencies, my monitor produces quite a nice yellow from a mixture of red and green phosphors. Similarly my eye is capable of seeing a whole spectrum of colors with only three cone pigments. Color is manufactured by the retina. SM

 

 

[StringJunky]

Because color is based on subjective experience, and has only a passing relationship to light frequencies, my monitor produces quite a nice yellow from a mixture of red and green phosphors. Similarly my eye is capable of seeing a whole spectrum of colors with only three cone pigments. Color is manufactured by the retina. SM

 

Colour processing starts in the retina and the brain does the rest.

 

http://en.wikipedia.org/wiki/Color_vision

[Iggy]

I can't find an answer to what hue of green/yellow exactly is the human eye most senstivite to (in RGB).

 

Luminous flux/intensity are defined at ~555 nm (540 THz), because that's the peak sensitivity of photopic scotopic vision.

 

Lime green (#BFFF00) corresponds roughly to 564 nm. Like the wiki page says "Lime (color#BFFF00) is a pure spectral color at approximately 564 nanometers on the visible spectrum when plotted on the CIE chromaticity diagram... [it] is approximately the most visible color on the spectrum because it corresponds to a wavelength near the peak of the photopic luminosity function."

 

Here is a splotch on a screen cap of one of the games you linked:

 

 

because of jpg compression that probably won't be exactly #BFFF00, but it will be close.

[thinker_jeff]

Lime green (#BFFF00) corresponds roughly to 564 nm. Like the wiki page says "Lime (color#BFFF00) is a pure spectral color at approximately 564 nanometers on the visible spectrum when plotted on the CIE chromaticity diagram... [it] is approximately the most visible color on the spectrum because it corresponds to a wavelength near the peak of the photopic luminosity function."

 

This color is good for the background of the picture. But remember, it won't be always good for different backgrounds.

555 nm is the best color for perfect dark background, which has been written in the textbooks.

If the background is a sunny beach, guess what? The color giving you the best contrast is RED!

[Iggy]

This color is good for the background of the picture. But remember, it won't be always good for different backgrounds.

 

Respectfully, I think you might be missing the point.

 

550 nm yellow-green is the most visible color. This means that the human eye can distinguish more shades of that color than any other. We can, in other words, spot an enemy wearing some shade of yellow-green on a different shade yellow-green background better than we can spot an enemy wearing some shade of red on a different shade red background.

 

This is why the military makes use of this color in a heads up display:

 

 

it is the best color when the background could be anything.

Edited April 7, 2011 by Iggy

[thinker_jeff]

550 nm yellow-green is the most visible color. This means that the human eye can distinguish more shades of that color than any other.

it is the best color when the background could be anything.

 

It will be the worst color for the background color at 550 nm!

 

We need scientific understanding, not just searching.

[Iggy]

It will be the worst color for the background color at 550 nm!

 

We need scientific understanding, not just searching.

 

Ok, I'm assuming that you're being sincere. The question in the OP is worded correctly,

 

I can't find an answer to what hue of green/yellow exactly is the human eye most senstivite to (in RGB)... This could be useful for stuff like enemy color in FPS games

 

He is looking for the best color of uniform in which to dress his enemy and he wants that color to be most sensitive to his sense of vision. The color of his opponent does not change as the background changes. It is constant. For this reason alone, it is pointless to say "well, for some backgrounds the color X provides the best contrast and for other backgrounds the color Y provides the best contrast".

 

The question is rather: "what color is best in any background?" -- in other words, in a random background. Scientists have studied this question. For example, they have recommended that fire engines and other emergency vehicles should be colored lime green. They found that tennis balls are most visible when colored lime green.

 

More than that, I looked at all three videos that the OP linked. They have predominantly dark-colored backgrounds.

 

If you think some other specific color offers the OP a better answer then by all means... otherwise, I don't see the point of your objection or your accusation that I'm somehow avoiding "scientific understanding".

Edited April 7, 2011 by Iggy

[thinker_jeff]

Ok, I'm assuming that you're being sincere.

 

I am sincere as far as in this topic.

At first I said that the color you picked is good for the background, which you specified at 564 nm. Then you left the question from OP and said the best color should be at 550 nm, that you claimed that "it is the best color when the background could be anything."

If you do understand the contrast of color, how can you have two different results? Why is the second statement incorrect?

Edited April 7, 2011 by thinker_jeff

[Iggy]

At first I said that the color you picked is good for the background, which you specified at 564 nm.

 

You are still missing the point. The question is not "what is the best color on a specific background?". The question is "what is the best color for many different backgrounds?".

 

Then you left the question from OP and said the best color should be at 550 nm

 

The OP itself said "550 nm".

 

, that you claimed that "it is the best color when the background could be anything."

 

Exactly. FPS games have a multitude of environments and backgrounds. The point is to find an RGB color that is most visible in all of those environments with all of those backgrounds -- taken as a whole.

 

how can you have two different results?

 

Most sources say "approximately 550 nm". The exact wavelength which is most visible will depend on the person. The RGB value that corresponds to that wavelength will vary widely between monitors. If a person were a serious gamer and wanted to fine tune the color of his opponent, I would recommend getting an "optic yellow" tennis ball (most people probably already have one). A person could then match the closest pure spectral color to match.

 

A pure spectral color is either a single RGB primary color or a mix of red and green or green and blue. The color we're looking for is a mix of red and green. So, Answers may put the color #BFFF00 (the one I quoted from wiki) on his game character and see if it matches his tennis ball.

 

If it looks too yellow then he would make the "BF" part smaller. BF is a hex number. In decimal it is 191. To drop it down to 180 he would change it to "B4" and the new color would be #B4FF00. To make the color more yellow he would make the red value larger... maybe #C8FF00.

 

Really, this would be the only way to find the exact RGB value that corresponds to the "approximately 550 nm" color that is being sought because monitors, depending on type, model, and age display very different colors.

 

edited to add --->

 

here is an interesting map of RGB values vs. spectrum wavelength:

 

 

http://www.isc.tamu.edu/~astro/color/spectra.html

Edited April 7, 2011 by Iggy

[thinker_jeff]

 

A pure spectral color is either a single RGB primary color or a mix of red and green or green and blue.

 

 

Wrong. Please look the following explanation.

 

 

 

 

The C.I.E. Color Space

The CIE system characterizes colors by a luminance parameter Y and two color coordinates x and y which specify the point on the chromaticity diagram. This system offers more precision in color measurement than do the Munsell and Ostwald systems because the parameters are based on the spectral power distribution (SPD) of the light emitted from a colored object and are factored by sensitivity curves which have been measured for the human eye.

 

Based on the fact that the human eye has three different types of color sensitive cones, the response of the eye is best described in terms of three "tristimulus values". However, once this is accomplished, it is found that any color can be expressed in terms of the two color coordinates x and y.

 

The colors which can be matched by combining a given set of three primary colors (such as the blue, green,a nd red of a color television screen) are represented on the chromaticity diagram by a triangle joining the coordinates for the three colors.

 

http://hyperphysics.phy-astr.gsu.edu/hbase/vision/cie.html

[Iggy]

A pure spectral color is either a single RGB primary color or a mix of red and green or green and blue.

Wrong. Please look the following explanation.

 

What is your point?

 

Are you objecting because I said "a pure spectral color..." rather than "RGB's closest approximation to a pure spectral color..."?

 

I want to think you are being sincere and not trolling, but either way -- I don't think our discussion is benefiting the thread.

[thinker_jeff]

I think that the third person can tell who is sincere or not. The thread is for everyone to see what should be the correct understanding.

 

You're still arguing like this:

 

Really, this would be the only way to find the exact RGB value that corresponds to the "approximately 550 nm" color that is being sought because monitors, depending on type, model, and age display very different colors.

 

Really? You figure out "approximately 550 nm" color is 564 nm because Answers' monitor (type, model and age).

Good story.

Edited April 7, 2011 by thinker_jeff

[Iggy]

Answers, If you're still around here is an excellent program:

 

http://www.efg2.com/Lab/ScienceAndEngineering/Spectra.htm

 

It converts 555 nm to r=179 g=255 b=0 or #B3FF00.