x% of NTSC color gamut

I am looking at 3 different monitors just to understand how this color gamut thing works.
One 23 inch HP monitor says that it has %68 NTSC color gamut, then I see another HP monitor that is not LED a bit older same 23 inch screen and a bit more expensive although every other spec is the same, but it has %72 NTSC color gamut, then I look at my HP 2710m and it has %92 NTSC color Gamut. Now as far as I know my 2710m is a TN screen so this color gamut is not that related to the VA or IPS situation I think. I also assume higher the percentage the better the monitor. Can somebody clear this up for me?

Andyk5 said:

I am looking at 3 different monitors just to understand how this color gamut thing works.
One 23 inch HP monitor says that it has %68 NTSC color gamut, then I see another HP monitor that is not LED a bit older same 23 inch screen and a bit more expensive although every other spec is the same, but it has %72 NTSC color gamut, then I look at my HP 2710m and it has %92 NTSC color Gamut. Now as far as I know my 2710m is a TN screen so this color gamut is not that related to the VA or IPS situation I think. I also assume higher the percentage the better the monitor. Can somebody clear this up for me?

Click to expand...

72% NTSC color is basically 16.7 million colours or sRGB. This is good. Anything higher means wide gamut.

So it depends what you use your monitor for, if you are a casual user, sRGB would be best if you are trying to get the best color calibration out of your unit, as basically almost everything done with a computer uses the sRGB colour space.

If you are a professional who from start to finish has programs (and usually a printer) which can go beyond that then you require a higher gamut monitor, however, unless if there is a dedicated srgb mode, it will cause inaccurate colouration with srgb sources.

Basically (and very simplified) colour is measuerd Red (0-255), Green (0-255), Blue (0-255), so an sRGB source that tells your monitor to display (230. 200, 150) on a sRGB source it will look fine, but because (230, 200,150) in sRGB is not the same in a widegamut format, a monitor that only does wide gamut, will display a similar colour but it will be a different shade or intensity.

On the same topic, I have a similar question.

How do screens emulate other color modes? If a screen was a wide gamut monitor, how would it emulate that the chromaticity of sRGB is lower? If it just limits the maximum amount of red, green and blue, it would correspond to a drop in brightness, am I right?
All this stuff with 6,8,10 bit panels in my head only means the adjustability of the luminance, while the chromaticity is a feature of the color filters and backlight.

I'm imagining the chromaticity to be equivalent to varying the amount of water color dissolved in water, i.e. to vary the paleness of the resulting color when light shines through it. The water color would correspond to the color filters applied to the LCD.

So how does this REALLY work?

The last sentence you wrote, Daemos, is what I'd consider to regard the luminance/brightness, i.e. to what amount the liquid crystal is twisted. I often see people write similar stuff, but I just don't get how this translates to chromaticity/gamut.

tk-don said:

So how does this REALLY work?

The last sentence you wrote, Daemos, is what I'd consider to regard the luminance/brightness, i.e. to what amount the liquid crystal is twisted. I often see people write similar stuff, but I just don't get how this translates to chromaticity/gamut.

Click to expand...

Some monitors have built in dedicated sRGB modes which have built in eletronics (in what I would akin to a translator) to display the proper colour (and they can be each calibrated to such)

I'm not sure if I'd call it luminance/brightness difference, as it is actually a different colour being presented.

here are some links I posted in another thread:
http://www.colorforums.com/viewtopic.php?t=852&sid=3ee5c0e1c631931bde22e0c61ef7f651
http://www.jseaman.com/articles/srgb.html

The first link explains how the colours can get translated to another.

So where does this rank my 2710m with its "listed" %92, is this good, I am not by my monitor atm so i do not know if it has a dedicated srgb mode, however there was a big visible difference between this monito and other similarly priced 27inch monitors as I was picking it and I am not sure if that is due to the wide color gamut, personal preference or that shiny, bright look HP monitors have.

Andyk5 said:

So where does this rank my 2710m with its "listed" %92, is this good, I am not by my monitor atm so i do not know if it has a dedicated srgb mode, however there was a big visible difference between this monito and other similarly priced 27inch monitors as I was picking it and I am not sure if that is due to the wide color gamut, personal preference or that shiny, bright look HP monitors have.

Click to expand...

I'm not sure what you mean by "rank" but, even if properly calibrated, if it does not have an sRGB mode that can be calibrated, the colours will not be accurate, unless if you have software on your computer that can translate sRGB colour space to it's wide gamut colour space your monitor natively accepts.

If it does have an sRGB mode, it still will need to be calibrated to give accurate colours, there is no way around that unless if it's already calibrated.

I'd second what Daemos said... if the monitor gamut matches with the working space you're working in, it's fine, but even then you need to make sure that the colors are caillbrated to this working space.

Daemos said:

Some monitors have built in dedicated sRGB modes which have built in eletronics (in what I would akin to a translator) to display the proper colour (and they can be each calibrated to such)

I'm not sure if I'd call it luminance/brightness difference, as it is actually a different colour being presented.

here are some links I posted in another thread:
http://www.colorforums.com/viewtopic.php?t=852&sid=3ee5c0e1c631931bde22e0c61ef7f651
http://www.jseaman.com/articles/srgb.html

The first link explains how the colours can get translated to another.

Click to expand...

Thanks for the links, it is really appreciated... This was only regarding the modification of gamut in software, but it made me think.
I played a bit with photoshop by changing the saturation in HSL-defined colors.
When selecting an RGB as 255,0,0, the more the saturation (chromaticity, gamut) is lowered, the more green and blue rises equally. The further away from a true red it becomes, the more unequal G and B becomes as saturation is decreased, but they still rise as saturation is decreased.
Comparing this to the "water color dissolved in a glass of water" analogy - less chromaticity would just mean more wavelengths in the blue and green region passes through from the white light shining through it (basically).

So, now, regarding the "native" gamut of displays, this made me realize:
The color filter must match extremely well with the backlight. The spectrum of the backlight must be known precisely, because the color filters must be created to fit it exactly. If we take the red sub-pixel as an example, the red color filter must pass a very exact amount of other colors from backlight to limit the color gamut. The same principle applies to the other two sub-pixels.

When emulating a color space, the other sub-pixels are trying to change the saturation by emitting light to compensate, just as the principle in demonstrated in photoshop, so some kind of translation matrix is required. This would also mean that contrast is going to be lowered by some amount.

All this this might also explain some parts of why some panels have lower contrast than others.

Awesome

tk-don said:

So, now, regarding the "native" gamut of displays, this made me realize:
The color filter must match extremely well with the backlight. The spectrum of the backlight must be known precisely, because the color filters must be created to fit it exactly. If we take the red sub-pixel as an example, the red color filter must pass a very exact amount of other colors from backlight to limit the color gamut. The same principle applies to the other two sub-pixels.

Click to expand...

You are correct on the backlight, the backlight must beable to reproduce the wide gamut as well.
More reading:
See: http://www.tftcentral.co.uk/speccontent.htm#gamut
http://www.tftcentral.co.uk/featurescontent.htm#gamut

tftcentral overall is a great website to learn about monitors.

What wavelengths are used play important role on how humans perceive colors and especially those with better color sensitivity. sRGB standard sets where should be chromatic coordinates but it doesn't really tell what wavelengths and what combination of them should be used to get there...

Having good enough spectrometer one could check what exact wavelengths given monitor actually generate and I saw such measurements. Trinitron CRT used very different light wavelengths that LCD did... The same now LED LCD have sRGB too but colors feel quite different than older sRGB CCFL...

Daemos said:

72% NTSC color is basically 16.7 million colours or sRGB. This is good. Anything higher means wide gamut.

Click to expand...

This is wrong. Screen color counts have absolutely nothing to do a with display’s color gamut, though manufacturers will attempt to tie them together.

Number of colours just represents scale of colours in that triangle in CIE diagram.

see this and you get the answers: Bigger Isn't Better, NTSC? Never! , Adobe RGB or Not to Be?, Bit-Depth Misconceptions
http://www.maximumtech.com/display-...itor-hdtv-companies-cook-their-specs?page=0,3

.
.. Thought I'd pass along some of my thoughts on this from my main monitor's thread since it relates...

elvn said:

http://images.anandtech.com/graphs/apple27inchcinemadisplay_092610174635/24902.png

As I've said before, colorspace testing is usually done in dark rooms with the testing device right up against the panel. I doubt that the highest ratings are perceived as such to the human eye at viewing distance, with the panel flooded with light. That is one reason the high end nec's come with a three sided monitor hood - to reduce image degradation caused by light pollution. Furthermore, the over-aggressive AG coatings on other mfg's IPS displays have been reported to screw up the way whites look, which is a pretty common background color when doing color specific work. They also compromise fine details like text, so may compromise fine details in authored art and photos. So as you can see, the higher rated colorspaces may not be so much to the human eye in an improper/light-polluted environment and having a thick crystal/greasy glaze on top of the screen.

The acd is very lush and vibrant, imo the glossy screen enhances it. Unless you are holding up color swatches to your screen or work in magazine publishing I doubt you'll see any difference with the ACD's 83.1 % adobeRGB 1998 in real human eye viewing.

Click to expand...

..

Ashok0 said:

All LG IPS displays (minus the Apple LED Cinema Display 27" and Hazro HZ27A/C) have an anti-glare layer that creates a "sparkle" effect on the panel. It looks like there is sand or glitter all over the screen. Whites in particular are heavily effected. Samsung's upcoming PLS display is matte and is confirmed to NOT have this effect, so this issue has nothing to do with matte versus glossy. Some people are more sensitive to it than others depending on their vision. Here is an animated GIF showing the "sparkle" effect:

Click to expand...

sRGB standard sets where should be chromatic coordinates but it doesn't really tell what wavelengths and what combination of them should be used to get there...

Click to expand...

It is true that all spectral information is lost when describing a color sample with colorimetric values. They represent the evaluation of the color stimulus through an appropriate function. For the ICC workflow the CIE 2 degree standard observer is used.

It would be a disaster if we would have to emulate every possible color stimulus. Fortunately the human eye evaluates a color stimulus altogether. Apart from limitations of measurement equipment (e.g. filter characteristic for colorimeters, sample intervall for spectrophotometers) we can of course experience limitations of colorimetry based on a fixed observer function too. There is no perfect match with our very own sensitivity - so measurements can suffer from effects of observer metamerism. Regarding this the sharp spectra of screens with WCG-CCFL and RGB-LED backlights is quite disadvantageous and can lead to visible deviations between measurement and actual perception. You can see this very well in a dual screen setup of a CCFL and WCG-CCFL screen where a calibration to colorimetric identical whitepoints will result in an unsatisfactory matching. Solution is a calibration to a visual but not colorimertic identical whitepoint (or an evaluation function that matches your own sensitivity better). The same is true for normlight matching - using D50 normlight I have calibrated my screen to 5800K which gives a match for my perception.

How do screens emulate other color modes? If a screen was a wide gamut monitor, how would it emulate that the chromaticity of sRGB is lower?

Click to expand...

The flexible color space emulation of the higher end screens uses a known standard condition to accomplish the required transformations. You can compare this with an ICC workflow based on matrix profiles where 3x3 matrices are used for forward and backward transformations (from and to the PCS). NEC even implements a gamut mapping for simulation of color spaces beyond the native color space. These emulations work very well when operated in a correct way (the user should for example carry out a chromatic adaption when deviating from the reference whitepoint of his target values). There will be some deviations with time (as I said, the transformations are based on a known standard condition) but this is true also for non WCG "sRGB" screens. For some newer CG modells Eizo offers the possibility of a "gamut update" that remeasures the initial state.

Best regards

Denis