Pointer gamut anyone?

[davidedric]

I was doing a bit of Sunday afternoon browsing and came across the Pointer gamut

http://www.tftcentral.co.uk/articles/pointers_gamut.htm

It is said to be the measured gamut that the human eye can actually perceive. One interesting aspect is when common colour spaces are mapped onto it. sRGB falls well within it. AdobeRGB does a lot better, though goes unnecessarily far into the green. ProphotoRGB certainly does cover it all but also includes a lot of colours we can't perceive.

I don't think any of this is of cosmic significance, but I found it interesting.

Dave

[NorthernFocus]

Thanks for sharing. Color is something that I haven't paid nearly enough attention to in the past. I'll give it a read.

[Richard Lundberg]

I believe the Pointer gamut is all the reflective colors.

[fenix]

I was doing a bit of Sunday afternoon browsing and came across the Pointer gamut

http://www.tftcentral.co.uk/articles/pointers_gamut.htm

It is said to be the measured gamut that the human eye can actually perceive.

Dave

It actually says:

The Pointer’s gamut is (an approximation of) the gamut of real surface colors as can be seen by the human eye, based on the research by Michael R. Pointer (1980). What this means is that every color that can be reflected by the surface of an object of any material is inside the Pointer’s gamut.

In other words, a gamut of Real World colors reflected by Real World surfaces (the ' as can be seen by the human eye' being both redundant and confusing). Colors 'as can be seen by the human eye' continue to be represented by the CIE horseshoe style diagrams.

Here's another Real World gamut:



cheers.

[Glenn NK]

In other words, a gamut of Real World colors reflected by Real World surfaces (the ' as can be seen by the human eye' being both redundant and confusing). Colors 'as can be seen by the human eye' continue to be represented by the CIE horseshoe style diagrams.

cheers.

Which raises the question; "whose eyes"? There is a large variation in colour sensitivity amongst humans.

G

[fenix]

Which raises the question; "whose eyes"? There is a large variation in colour sensitivity amongst humans.

G

Hi Glenn,

The implication is the average eyeball, perhaps that of the CIE 1931 2-deg Observer?

Reckon?

T

[Glenn NK]

Of course which raises another question: "who is average"?

Some degree of colour blindness is quite common among human males, so perhaps the average human female.

I always thought my colour sensitivity was very good (and accurate). When I worked for an organization that had medium blue vehicles (close to what we call cyan), I noted that I could see them more easily at greater distances than could my co-workers (at least I thought I could). Perhaps I have more sensitivity to blue than what is "normal". But here we go again - what's normal? And do we compensate?

[Mike Buckley]

cosmic significance

It is worth reading your post just to come across that phrase. I love it!

[davidedric]

It actually says:


The Pointer’s gamut is (an approximation of) the gamut of real surface colors as can be seen by the human eye, based on the research by Michael R. Pointer (1980). What this means is that every color that can be reflected by the surface of an object of any material is inside the Pointer’s gamut.


In other words, a gamut of Real World colors reflected by Real World surfaces (the ' as can be seen by the human eye' being both redundant and confusing). Colors 'as can be seen by the human eye' continue to be represented by the CIE horseshoe style diagrams.

Don't really understand your post, could you please expand.

It does seem to me that "as can be seen by the human eye" is important. "Any colour that can be reflected...." How do you interpret colour? All that I was musing on is that a gamut that contains colours (by which I mean combinations of electromagnetic wavelengths) which cannot be perceived by the human eye (but quite possibly by insect eyes, for example) has some redundancy. Like I said, no "cosmic significance" (Glad you liked the phrase Mike )

Dave

[ajohnw]

There is a need for a bit of care when reading about gamuts. This page explains why with one example using sRGB but it can apply to all.

http://brucelindbloom.com/index.html...ityGamuts.html

I haven't read all of the posts so may be incorrect in my assumption that Pointers gamut isn't being "SEEN" in the right context. The CIE gamuts cover human vision. Pointer's covers subtractive colour mixing. I see it as the limitations of a CMYK colour space but don't quote me.

What the article is getting at relates to additive colour mixing. Usually this is a 3 colour system. R, G and B. The actual wavelengths of these that are used limit how far each can go in a pure form. A red in other words can't be made more redder than it is. Same for green so mixing various intensities red and green can only produce colours in between the 2. He shows that 3 colours can't cover Pointer's gamut as far as displays are concerned. Try and cover it with 3 straight lines.

It all makes things more complicated when light sources are used to illuminate things. There are RGGB Led's about aimed at improving this. The 2 shades of green are needed to make certain colours reflect correctly - the reflection is subtractive colour mixing unless it's a mirror. He mentions a red car travelling through a forest and the effect on the redness of the car.

Don't know if this will help. Truth is trying to gauge the span of human colour vision in the first place is rather difficult. Even more so when the 3D aspect is included.

John
-

[fenix]

Don't really understand your post, could you please expand.

It does seem to me that "as can be seen by the human eye" is important.

Here's how the phrase itself is misleading: If I ask someone to show me all the colors "as can be seen by the human eye" they would, in all probability, come up with the 1931 CIE Chromaticity Chart, thinking that I meant "all the colors that can be seen by the human eye". I said that the phrase was redundant because if you can't see it, it is not a color (by definition, a color is visible).

"Any colour that can be reflected...." How do you interpret colour?

Color is interpreted in many ways, as I am sure you know. For example, going to the paint store and asking for L* = 30, a* = 110, b* = -56 paint will get you nowhere, but quoting Pantone or Munsell numbers might. In photography, XYZ is the most definitive because everything else gets obfuscated with gamuts, white points, illuminants, etc., ad naus.

All that I was musing on is that a gamut that contains colours (by which I mean combinations of electromagnetic wavelengths) which cannot be perceived by the human eye (but quite possibly by insect eyes, for example) has some redundancy. Like I said, no "cosmic significance"
Dave

Strictly speaking, combinations of electromagnetic wavelengths which cannot be perceived by the human eye are not colors. Where I'm from, we say "If yew cain't see it, eeyut ain't no color". However, in the world of photography (as opposed to the science of color) we are able to misuse any word we like and get it published on the 'net - as in the article you linked to.

cheers,

[fenix]

Of course which raises another question: "who is average"?

Some degree of colour blindness is quite common among human males, so perhaps the average human female.

I always thought my colour sensitivity was very good (and accurate). When I worked for an organization that had medium blue vehicles (close to what we call cyan), I noted that I could see them more easily at greater distances than could my co-workers (at least I thought I could). Perhaps I have more sensitivity to blue than what is "normal". But here we go again - what's normal? And do we compensate?

In the context of both the link and this thread, in which several CIE chromaticity diagrams have been presented, "normal" has to be the average result of those individuals who took part in a series of experiments done in the late 1920s by William David Wright and John Guild.

"Normal" is the standard but is not necessarily attainable - just as it is impossible to buy a lamp that exactly matches the CIE D65 specification.

cheers,

[Peeshan]

Check out this link: http://www.xrite.com/online-color-test-challenge .

I don't know the scientific accuracy behind it, but it's always interesting.

In photography, I've noticed AdobeRGB tend to give duller color than sRGB when there is one dominant color (a blue sky looks greyish) - which is somehow logical. But I guess it can be useful for scene with a lot of different shades.
So in any case, it's good you brought up the topic.

[ajohnw]

Here's how the phrase itself is misleading: If I ask someone to show me all the colors "as can be seen by the human eye" they would, in all probability, come up with the 1931 CIE Chromaticity Chart, thinking that I meant "all the colors that can be seen by the human eye". I said that the phrase was redundant because if you can't see it, it is not a color (by definition, a color is visible).



Color is interpreted in many ways, as I am sure you know. For example, going to the paint store and asking for L* = 30, a* = 110, b* = -56 paint will get you nowhere, but quoting Pantone or Munsell numbers might. In photography, XYZ is the most definitive because everything else gets obfuscated with gamuts, white points, illuminants, etc., ad naus.



Strictly speaking, combinations of electromagnetic wavelengths which cannot be perceived by the human eye are not colors. Where I'm from, we say "If yew cain't see it, eeyut ain't no color". However, in the world of photography (as opposed to the science of color) we are able to misuse any word we like and get it published on the 'net - as in the article you linked to.

cheers,

That's being a bit unfairly pedantic really. A full blown scientific write up probably wouldn't be of much use to most of the general public and as a whole the page is a lot lot better than many and to top it up their monitor reviews are extremely useful.

John
-

[Rex]

I have an interest in the Pointer's Gamut and am trying to see if it available anywhere as a .icc profile. If anyone can help I'd appreciate a link.

Our eyes are extremely good at comparison and pretty useless at absolute measurement (skin and grass might be the exceptions ). In the world of the amateur photographer the differences in sRGB, aRGB and ProPhoto workflows can only be spotted if the prints are held side by side. Separate them by a minute or 10 metres and you cannot see the difference.

I have just had to explain to someone that an aRGB work flow is pretty pointless when they print out on a textured art paper that doesn't even encompass all of sRGB. That said some colours between aRGB and sRGB were reproduced but then you'd never notice anyway.

[Shadowman]

Thanks for the link.

[Manfred M]

I have an interest in the Pointer's Gamut and am trying to see if it available anywhere as a .icc profile. If anyone can help I'd appreciate a link.

Our eyes are extremely good at comparison and pretty useless at absolute measurement (skin and grass might be the exceptions ). In the world of the amateur photographer the differences in sRGB, aRGB and ProPhoto workflows can only be spotted if the prints are held side by side. Separate them by a minute or 10 metres and you cannot see the difference.

I have just had to explain to someone that an aRGB work flow is pretty pointless when they print out on a textured art paper that doesn't even encompass all of sRGB. That said some colours between aRGB and sRGB were reproduced but then you'd never notice anyway.

Rex - Welcome to CiC. Would you mind completing your profile and enter your name and where you are from. We are a pretty informal bunch here at CiC and tend to work on a first name basis. This functionality can be accessed from the "My Profile" tab at the top of the screen.

There are a few members that are definitely into some of the "non-standard" colour spaces; I expect that they will respond to you on this one.

The Pointer gamut (1980) is an interesting, academic colour space that has been around for a while, but does not seem to have a lot of independent verification. Colour vision will vary from individual and will be age dependent, so in many ways it is an oversimplification of the visible spectrum from a human vision standpoint.

The reason that I suspect it has never really caught on is that the devices we use to view our images; screens and prints cannot reproduce much of the Pointer colour space (100% of human vision). Much of the computer screen can indeed reproduce the sRGB gamut (came out in 1996 and incorporates about 25% of human visual colour range) and wide gamut screens and higher end printers can reproduce a high percentage of the AdobeRGB gamut (came out in 1998; around 55% of human visual colour range).

If you are concerned about not having all the colours that we can see, then the ProPhoto (2011) is a well established colour space that does this. I have not compared it to Pointer, but suspect for the most part Pointer would be a subset of ProPhoto. The main (academic?) criticism I have seen of ProPhoto is that it covers some "artificial" colours that we can't see. If we can't see them, should I really get to concerned about this "flaw" in ProPhoto. I certainly do not.

As for your comments on human vision and the ability to differentiate between sRGB and AdobeRGB, especially at a distance; I can certainly see where you are coming from. On the other hand, I stopped working on a piece that was aimed at posting on the web, when some the vivid colours in the image, which were plainly brilliant in AdobeRGB looked rather muddy in sRGB and did not give me the effect (or quality) I wanted.

[pnodrog]

Strictly speaking, combinations of electromagnetic wavelengths which cannot be perceived by the human eye are not colors. Where I'm from, we say "If yew cain't see it, eeyut ain't no color". However, in the world of photography (as opposed to the science of color) we are able to misuse any word we like and get it published on the 'net - as in the article you linked to.

cheers,

Strictly speaking there are combinations of electromagnetic wavelengths that can be seen but cannot be differentiated by the human eye from some other combination of electromagnetic wavelengths. So provided some component(s) falls within the visible spectrum it will be seen but the colour may not be unique...

[xpatUSA]

Strictly speaking, combinations of electromagnetic wavelengths which cannot be perceived by the human eye are not colors.

Strictly speaking there are combinations of electromagnetic wavelengths that can be seen but cannot be differentiated by the human eye from some other combination of electromagnetic wavelengths.

It looks like we have established that, strictly speaking, some combinations of electromagnetic wavelengths cannot be seen and some can.

A major advance in our understanding of color!

[xpatUSA]

I have an interest in the Pointer's Gamut and am trying to see if it available anywhere as a .icc profile. If anyone can help I'd appreciate a link.

Our eyes are extremely good at comparison and pretty useless at absolute measurement (skin and grass might be the exceptions ). In the world of the amateur photographer the differences in sRGB, aRGB and ProPhoto workflows can only be spotted if the prints are held side by side. Separate them by a minute or 10 metres and you cannot see the difference.

I have just had to explain to someone that an aRGB work flow is pretty pointless when they print out on a textured art paper that doesn't even encompass all of sRGB. That said some colours between aRGB and sRGB were reproduced but then you'd never notice anyway.

Are we sure that a gamut that is shown on an xyY chart can be turned into a .icc profile? Personally, I am struggling with the concept. Are there any links to exactly how that is done? For example, what might a .icc profile for a pure sRGB gamut look like as compared to one for ProPhoto? By "look like", I mean what values the profile tags would have. Do color gamut .cc profiles even exist? Pardon my pedantry . .

A .icc profile, amongst other things, has tags for the primary colors - which are corners of a triangle on the xyY diagram. Printer profiles might have more points to match multiple cartridges (anybody?). But the said Pointers gamut is an irregularly shaped non-geometrical blob, so I still fail to understand how that can be represented in a .icc profile format.