Jump to content

Color Theory Question


Recommended Posts

I'm trying to understand the math behind calibration (I'm a geek), so I've been performing

some experiments in Photoshop. I have an Apple Cinema Display calibrated with Eye-One

to a gamma of 2.2, so I'm pretty confident in the fidelity of the display (I also had two of

my images printed by a pro lab and they matched the screen).

 

Anyway, I created a red/black checkerboard and surrounded it by a field of solid red. The

red in the checkerboard was 255 and the solid red was 128. I then used levels to adjust

the gamma until the tones matched. I'm not sure what I expected, but it wasn't what I got.

They matched at a gamma of 1.6 (a resulting value of 165).

 

Can anybody point me to a good (hopefully online) explanation of how this stuff works?

 

Thanks.

 

Andrew

Link to comment
Share on other sites

I was under the impression that Apple's monitors all operated at gamma=1.8, so you may have just more or less independently discovered/rediscovered that fact.<P>

 

The needed gamma can vary with intensity level, and your experiment only tests three points. Try the following image, which you'll see is a more thorough test, and see if you get the result comes out a little differently: http://endlessrapture.info/pics/GAMMA2.gif .

Link to comment
Share on other sites

Chris, Mac's traditionally used a gamma of 1.8, but (at least under OSX) you can set them

anywhere you want. I'm still not sure how what you wrote would explain needing a

gamma of 1.6. Thanks for the gif; what gamma is it supposed to represent?

 

Craig, at least with mine, I get equivalent shadow detail (about every other level up to

about 20) with 1.8 or 2.2.

 

Ellis, thanks for the hint. I had no idea that the intensity perception in different color

channels would be so different (though I guess had I studied the RGB -> HSV conversion

equations more closely, it may have been obvious). Using a white/black checkerboard and

128 grey gives a required gamma of 2.3. Still not quite what I expect, but interesting.

 

Andrew

Link to comment
Share on other sites

There's no difference in a display for a Mac or a PC, the all have a native Tone Response

Curve that's specified using a term known as Gamma. Most displays have a native gamma

in the Neighborhood of 2.2. The Mac OS assumes a gamma of 1.8 which is silly but that's

what it expects outside of non color managed applications.

 

When you make a document to test all this fun stuff, the working space has a gamma

(which doesn't have to match the gamma of the display). All kinds of things have their own

gamma (images, displays, printers, etc). At the very least, you should assign the display

profile to the image you're playing with so the numbers get sent to the screen and not

though the rest of the CMS.

 

I'm not sure what this test image is supposed to do for you...

 

Andrew Rodney

http://digitaldog.net/

Link to comment
Share on other sites

> ...but (at least under OSX) you can set them anywhere you want.

 

Ok, I guess what I meant was that the *nominal* Apple value is 1.8. My PC's video card doesn't spit out an actual number - one just moves a slider (actually there are three, but you can lock them together) until you get it right.

 

 

> I'm still not sure how what you wrote would explain needing a gamma

> of 1.6. Thanks for the gif; what gamma is it supposed to represent?

 

The tif doesn't represent any particular gamma. It just does your checkerboard type experiment using more than only three points in the 0-255 range. Investigate the pixel values in each of the nine boxes and you'll see what I mean.

 

If you can perfectly null out all but one or two of the boxes then you've got a value which works well over the entire intensity range. If each color needs a substantially different setting to get a null then something's really flakey (and the grey squares are not likely to null very well).

 

I'm not sure if I'd be too concerned over your value of 1.6 being that much different from the nominal value. Each system is bound to be a little bit different considering the number of components involved, and a tenth or two doesn't seem like much to me. If the part of the chain which you can control requires a little bit less gamma that just means the rest of the system has a bit more than usual.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...