That's kinda the way I do it. Don't always do the math, but I know that if I
sharpen at 100 percent, or "actual pixels", I don't get a good representation.
The sharpening program I use, from Pixel Genius, is, I think, one that Fraser
either developed or helped develop. It has separate modules for capture
sharpening, creative sharpening, and output sharpening. The modules come with
layer masks, which I frequently use to soften sharpening effects in skies or
water, or to brush in sharpening where I'd like to have more of it.
The advice to blow up to full size and crop out is similar to something I heard
in John Paul Caponigro's printing workshop. He always prints a strip of the
image at the final size to make sure there aren't artifacts he couldn't see on
the screen. Usually his strip contains some each of foreground, midground and
sky. But then a lot of his prints are 40x60 and such. I usually just make a
test print on Epson's Ultra Premium Matte Paper (which isn't) at the final size
and inspect it. One good thing about UPPM is that it holds a lot of detail. If
something doesn't show on UPPM, it isn't likely to show on Ultra Smooth Fine
Art, Velvet Fine Art, or some of the other papers I use. If I were doing 40x60
prints, I doubt I'd do that. <g>
--Bob
On Sep 17, 2011, at 1:47 PM, Chuck Norcutt wrote:
> I would try using the sharpening rule I learned from Bruce Fraser's book
> on sharpening. He suggests first, of course, cropping and resizing the
> image to the final pixel dimensions for the print. To prepare for
> sharpening he then displays the image at the ratio between screen ppi
> and print dpi. For example, if the screen has 90 ppi and the print will
> be 300 dpi then display the image at 90/300 = 0.3 or 30%. Then he says
> sharpen until it looks very slightly crunchy to account for some
> softening in the printing process. I find it a very good rule of thumb
> for sharpening and I suspect it would work in your case when trying to
> evaluate the look of a final print after sharpening.
>
> If you can't see the halo when the image is displayed at 30% (or
> whatever your criterion is) I don't think you'll see it in the print
> either. Of course, you can further test that by blowing the image up to
> 20x30 and then cropping out a 4x6 or 5x7 test area that contains the
> halo and print that small image.
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|