Re[2]: [cgmo-webcgm] RE intensity

[Lofton Henderson <lofton@rockynet.com>]

At 12:34 AM 7/14/2004 -0400, Benoit Bezaire wrote:
>Tuesday, July 13, 2004, 8:16:20 PM, Lofton wrote:
>
>[...]
>LH> That said, whether or not to say anything about clamping depends on what's
>LH> the error philosophy -- the equations can only yield out-of-range results
>LH> for illegal parameter values.
>
>Who said intensity had to be limited to [0,1]? :-)

Forrest did, in his original proposal, which I thought you had endorsed as 
the way to go (I don't have the messages here on this system, to check it).

>If it is a relative
>intensity shouldn't I be allowed to go above 100% of the original
>color?

As proposed, the range was limited to [0,1], and given the equations, then 
you could only dim or fade the color.

>I know it doesn't apply well to color that don't have a hue
>ex: (255,0,0)... you end up in negative values as soon as you exceed
>1 (or 100%).  But for a sky blue color (64,128,192) applying an
>intensity of 1.25 (or 125%) would make it a darker blue.
>
>Is there any use for that?  I guess I'm wondering if we should assume
>we want to shade to white, what if the illustration background is
>black?  This is way I introduced clamping into the discussion.

Yes, I had lots of similar questions.

It seem to me that we are proceeding without consensus on what we are 
trying to achieve?  Dim or fade only?  Brighten also?  Maybe we ought to 
decide this first.

-Lofton.


>It's just an idea...
>
>--
>  Benoit   mailto:benoit@itedo.com