Cousin Ricky <ric### [at] yahoocom> wrote:
> On 2021-04-01 8:55 AM (-4), Kenneth wrote:

> > ...So it looks like the parser
> > *is* making some kind of mathematical distinction, whatever that is.
>
> It's not mathematical; it's procedural.  Once the srgb keyword does its
> job, it goes home for the day.  So in your color TEMP_COLOR*50000, there
> is no srgb calculation.
>
> Try thinking of TEMP_COLOR as having been permanently converted to
> linear rgb; once the srgb keyword conversion is applied, it has no
> memory of having ever been non-linear.

Ah!! My "light bulb of understanding" has finally turned on. That's fascinating,
and extremely helpful. Thanks.

Before reading your post, I did a visual 'difference' test in Photoshop (wildly
exaggerated) of two renders, one using
       color TEMP_COLOR*50000  (with my pre-#declared srgb color)
and the 2nd using the #debug results of that construct...
       rgb <10702.1, 3661.9, 22399.4>

The results are absolutely identical. Maybe that was to be expected, given that
the 2nd is derived from the 1st(?) But I wanted to be certain that the colors
and brightness values agreed, and that it wasn't just 'wishful thinking'.

Post a reply to this message