Testing my monitor - Am I interpreting this correctly?

[Damo77]

Ok, I've just got a new monitor (a 19" Eizo, calibrated with an Eye-One Display II), and I wanted to check how its gamut compares with sRGB.

I don't have any fancy profile analysis software, so this is what I did.

I created a document of six swatches of pure R,G,B,C,M & Y - 255-0-0, 255-255-0, etc, in Photoshop.

I assigned my monitor profile to the document, then converted it to sRGB (using Relative Colorimetric, BPC on), and checked the resulting values.

Red: 255,0,0
Green: 0,248,0
Blue: 0,71,255
Cyan: 0,255,255
Magenta: 255,68,255
Yellow: 255,247,0

From this, I deduce:

> My monitor's red and cyan are equal to or more saturated than sRGB red and cyan, and identical in hue (I hope my terminology is accurate enough for this unscientific experiment)

> Monitor green is almost as saturated as sRGB green, and identical in hue

> Monitor yellow is about as saturated as sRGB yellow, but just a teensy bit on the red side

Then I get a little bit uncertain, so this is more of a question than a deduction:

> Monitor blue is at least as saturated as sRGB blue, but much greener

And I've got no idea how to interpret the magenta. The red and blue components would indicate that monitor magenta is beyond sRGB magenta, yet the significant green component would appear to haul that back.

Would anyone care to discuss/contribute/poke fun?

[Damo77]

Gee ... nobody?

[Zoodles]

I was interested in the concept and lurking - afraid I'm still too ignorant of colour management to comment.... ;)

[SolidxSnake]

Try downloading OpenRGB and testing the differences in gamut... I don't know if what you're doing is accurate or not.

[Az2Africa]

Huh???? :confused: I have no clue. How do the images look?

[SolidxSnake]

http://www.logicol.com/index.php?X=ORGB

[Damo77]

LOL! The images look great - you know it ;)

And frankly, very few images I deal with ever challenge the extremes of even sRGB, so I'm not concerned about colours I can't see.

I was just curious, is all.

[Damo77]

Thanks, SolidxSnake, I'll check that out tonight.

[Az2Africa]

BWT, it's a sin to use sRGB on an Eizo. :lol:
Spoils you doesn't it?

[Damo77]

Well, that's the point, isn't it? If my testing is sound, the monitor only just encompasses sRGB. And when I ran the same test with Adobe RGB, it's not even close.

Don't get me wrong - this is the only Eizo I could afford, and I'm rapt with it.

[Az2Africa]

Does that one have the iOne software built in?

[Damo77]

No, it's got Color Navigator.

[Az2Africa]

Then it should have the software for the iOne in Color Navigator. Check the Eizo site for any updates. I also uninstalled my Spyder software (that's what I use) before using CN to calibrate.
Late here. Off to bed.

[gcogger]

Here are a few ideas off the top of my head, for what they're worth! I'm at work, so I can't access Photoshop to try this myself.

Try converting the above RGB vlues into HSL notation - it should be easier to understand the differences. It's difficult to separate saturation and brightness in RGB. Alternatively, just set up the colour values in the info pallette so that it displays RGB and HSL values.

Try taking your image with saturated RGBCMY colours, convert to HSL, and reduce the lightness (while keeping saturation at 100%). Then (via RGB colour) convert to sRGB and see how the saturation values have changed. You need to do this at different values of L in order to test the gamut at different brightness levels.

An easier test is to take the image in the monitor profile, set the soft proof to sRGB and turn the gamut warning on. Then you can easily see what's out of gamut as you change values (while still working in the monitor profile). I'm not sure if you can do this with the image in HSL - probably not? - but you can monitor HSL values in the info pallette.

If you are seeing gamut clipping, try reducing the saturation of those colours (in your monitor profile) to see at what point the gamut clipping disappears. Again, this will be different at different brightness levels - this is a 3 dimensional problem :)

[René Damkot]

Also, when using relative colorimetric, out of gamut colors are clipped.
So if your monitor 255;0;0 is way more red then sRGB red, it would still end up as 255;0;0 in sRGB (assuming it's along the same line of red, so no G or B added).

To see what's happening, create a color spectrum (Click here (http://www.creativepro.com/article/out-of-gamut-realizing-good-intentions-with-rendering-intents)) in ProPhoto RGB. You'll probably see quite a few parts "block up" because they are out of monitor gamut. Now assign AdobeRGB. Should block up a lot less. (on a wide gamut screen; almost no blocking up).

Figuring out things like this "by the numbers" in RGB will drive you up the wall quite quick ;). Here is another read: Click (http://www.creativepro.com/article/out-of-gamut-getting-a-handle-on-color-management?page=0%2C0).

[Damo77]

Also, when using relative colorimetric, out of gamut colors are clipped.
So if your monitor 255;0;0 is way more red then sRGB red, it would still end up as 255;0;0 in sRGB (assuming it's along the same line of red, so no G or B added).

Well, yes, that's why I deliberately chose RelCol. I didn't want the in-gamut colours to be shifted in any way (which happens with Perceptual, in my understanding).

Here is another read: Click (http://www.creativepro.com/article/out-of-gamut-getting-a-handle-on-color-management?page=0%2C0).

Thanks, but that's just a very brief precis of Bruce (and Chris and Fred)'s fine Real World book, which I already own and refer to often.

[Damo77]

An easier test is to take the image in the monitor profile, set the soft proof to sRGB and turn the gamut warning on.

Ok, I took this idea and dabbled with it (but in reverse).

I started with this image:
http://www.bellephotography.net.au/pics/montest.jpg
(Each pure colour blends in three gradients to white (255s), grey (128s) and black (0s))

I assigned sRGB to the image, then set up the softproof to simulate conversion to my monitor profile (again, RelCol, BPC on).

Then I turned on Gamut Warning to see what colours in the sRGB gamut can't be reproduced by my monitor's gamut:
http://www.bellephotography.net.au/pics/moneizo.jpg

So, reds and cyans are fine, yellows and greens are almost fine, but blues and magentas fall short.

(Though this might seem strange to a lot of people, I'm not really interested in how much better than sRGB my new Eizo monitor is in the reds and cyans, so I haven't tried working that out. I just wanted to know where it falls short.)

[René Damkot]

Better then my laptop...
Just for giggles:
http://img.skitch.com/20080713-bf1fyewmk2ur8imfsxre2rrg89.jpg