My experience mirrors yours, Paul. I can easily see the difference
between an FA35/2 and the FA20-35/4, and both are very good
performers. I can also see the difference with a less capable
performer like the F100-300 or A28/2.8.
How can manufacturers sell bodies that are spec'ed better than we
need? Simple: add features, add higher quality expectations, and
market them. That's all. I've seen some amazing work produced from a
Canon D30 with 'only' 3Mpixel resolution. I've been complimented by
knowledgeable critics on work from Minox 8x11 negatives scanned at
2820 ppi that means only *1 Mpixel* resolution.
6Mpixel (5Mpixel typically after cropping and sizing) does well for
99% of the photography I see.
8-10 Mpixel, for the size range of prints I normally make, is way
more than enough but gives additional room for cropping and
adjustment. It's value increases as you move to shorter, wider field
of view lenses. It would be nice, it's not essential.
Without using a tripod, more degradation in image quality is due to
camera movement than lack of pixels. For instance:
http://homepage.mac.com/ramarren/IoM-4/PoA-lighthouse-0730usFR.jpg
That's taken with a Panasonic FZ10 on a tripod... 4Mpixel itty-bitty-
sensor camera with an ultrazoom lens. Unretouched, just straightened
and cropped. Prints to a beautiful 11x14, framed out to 16x20 or 20x24.
The vast majority of photographers never exploit the full
capabilities of their equipment.
Godfrey
On Oct 27, 2005, at 3:33 AM, Paul Stenquist wrote:
I can see the difference in my *ist D prints from the FA 35/2 vs.
the DA16-45/4. The latter show excellent resolution in 12 x 18
size, the former are superb. Wearing my most powerful reading
glasses and examining the prints at a distance that is far closer
than that from which they would normally be displayed, the
difference is discernible. Based on this very unscientific
experiment, I would have to say that the D can take advantage of
better lenses. I'm sure you can come up with some mathematics that
belies that, but the physical evidence says otherwise.
