That's the gist of it, as I understand it.
The other thing I learned that surprised me was how lots of noise
reduction, as the first step, then sharpening and other processing, of
film scans, especially color negs., could improve them. I didn't use the
NR techniques he did, though. NeatImage is better.
Moose
Chuck Norcutt wrote:
> I've read most of it but have yet to read every bit of it and know I'll
> need to re-read it at least once. However, there are two things you can
> do easily with one being sort of obvious and the other being derived
> from my reading of the book.
>
> 1) If you're sharpening for the display then just sharpen to what looks
> good to you at the actual size you intend to display.
> 2) (and this is my very rough cut summation from the book) If you're
> sharpening for a print then first, pre-size the image to the resolution
> you'll use for the print (say, 240-300 ppi) then display the image in
> your editor at roughly 25-30% and then sharpen until the screen image at
> 25-30% looks just slightly "crunchy" or a bit over-sharpened. The point
> of displaying at 25-30% is to account for the significant differences in
> resolution between screen and print (say 75ppi vs 300dpi or 1:4). Then
> sharpening the screen image until it's very slightly over-sharpened is
> to account for the softening effect that will come about from laying
> dots on paper.
>
> This is the simplistic rule I now use for sharpening and it seems to
> work pretty well. But YMMV.
==============================================
List usage info: http://www.zuikoholic.com
List nannies: olympusadmin@xxxxxxxxxx
==============================================
|