Chuck Norcutt wrote:
> You've posted this page several times before and I have to say that I
> don't see what you see. At least not anything that would get me to
> spend money for FM's stuff.
>
I agree that it's hard to compare while scrolling the screen. Try this
version (but not in IE) where you can switch between overlaid images
<http://www.moosemystic.net/Gallery/tech/Process/UpSample/Uprez.htm>.
It's arranged so that you can easily move your mouse quickly between
between the last version and the bicubic versions. If you don't see a
big difference in detail and sharpness with no increase in noise and no
annoying artifacts, I'll eat that man's clams. :-)
FM stair interpolation doesn't blow bicubic away, but is definitely
better at full pixel. Again the boxes are arranged so you can quickly
slide between it and the best bicubic version*. and will work on any
sort of image, not just those from a specific camera. What was most
interesting to me is that the results are the reverse of Butzi's results
with the test chart. I also like the fact that I can just do it all in
one step, rather than manually doing several to many steps with bicubic.
I could make my own actions(s), but Fred's already done it for less $
than my time to do all the experimenting is worth to me. And it works.
To my eyes, the camera specific Resize Pro simply blows bicubic away at
full pixel on screen. How big the difference on a print would be I don't
know. So far, I go on the naive assumption that more detail and
sharpness on screen will at least not be worse on paper, and may be
better when the difference is so great.
It makes no difference to me whether folks buy FM's software, although I
find his price/performance ratio very good. I'm simply interested in
tools that:
1. Accomplish what I want effectively.
2. Do it without a great deal of time and effort on my part.
3. Are priced reasonably for what they do.
Moose
* Notice that Butzi uses bicubic smoother because it "is what Adobe
recommends for resampling images to be larger:" and doesn't try sharper,
which works slightly better with my test image. Maybe he's not very curious?
==============================================
List usage info: http://www.zuikoholic.com
List nannies: olympusadmin@xxxxxxxxxx
==============================================
|