> Sorry, but that's not true. I know the difference between Bayer
> interpolation and gamma-encoding. And gamma-encoding is not "simply
> applying a logrithmic function" to the data before quantizing. It's a
> mite bit more involved than that, although it presents a first order
> approximation.
>
> Now that you explain what you meant more clearly, I understand
> exactly what you're saying. But it is meaningless when it comes to
> RAW workflow for someone using tools like Photoshop, and nearly all
> other production RAW conversion tools, on Mac OS X or Windows. The
> theoretical advantages of dividing Bayer interpolation from gamma-
> encoding, keeping the data in a linear representation, etc, is of no
> real significance when it comes to producing pictorial photographs
> when you have tools that do it well combined with good color
> management and [EMAIL PROTECTED] quantization space.
>
If you haven't already, you might want to take a read of this
guy's rantings on linear vs. gamma errors. The color management part of
it is different from "gamma errors."
http://www.aim-dtp.net/aim/evaluation/gie/index.htm
I'm pretty sure that even with Winders/MacOSX, one can establish a
workflow to edit in linear colorspace.
Pedantry aside, "real significance" is not the same as
significance. The latter is objective and quantifiable. The former
requires an individual's interpretation.... a *very* movable target
depending on the individual. I don't trust myself to make those calls, so
I rely on what's objective.
-Cory
--
*************************************************************************
* Cory Papenfuss, Ph.D., PPSEL-IA *
* Electrical Engineering *
* Virginia Polytechnic Institute and State University *
*************************************************************************
--
PDML Pentax-Discuss Mail List
PDML@pdml.net
http://pdml.net/mailman/listinfo/pdml_pdml.net
