Skip navigation
(RichardWagner)
Currently Being Moderated

Composite RGB Curve math

Aug 18, 2007 4:05 PM

In Photoshop's Curves dialog (CS3), you can add control points to the composite RGB curve by command-clicking an area of the image. I've tried to figure out how the location of the added control point is determined from the color that was clicked on, without success.

For example, command-clicking on a solid of RGB (127,127,190) results in a control point at 134,134. I can't figure out the relationship between the source color and the control point.

It's not the average, max, H, S, L, etc.

Anyone know?

Thanks,

--Rich Wagner
453 Views   1 Reply   Latest reply: jacobolus, Jun 24, 2012 8:49 PM
 
Replies
  • jacobolus
    Currently Being Moderated
    1. jacobolus,
    Jun 24, 2012 8:49 PM   in reply to (RichardWagner)
    Report

    It seems to be based on Photoshop’s “luminosity” calculation (Adobe’s invented word for what video people call “luma”) – in other words a weighted sum of the numerical values of the three channels. See http://en.wikipedia.org/wiki/Luma_(video)

     

    This is sort of silly, since the actual effect of the composite channel has nothing to do with “luminosity”, but instead acts on each channel independently, and therefore unless a pixel’s color is a neutral gray, the curve will affect each of its component values independently. The clicked-on pixel might not be at the chosen point on the curve in any of the three components.

     

    Basically, think of it as just a relatively arbitrary point that doesn’t bear much relation to the effect the curve will have on the clicked pixel, unless that pixel happens to be gray.

     
    |
    Mark as:
Actions

More Like This

  • Retrieving data ...

Bookmarked By (0)