Alan Browne - Minolta

D200 and Sony Alpha = same sensor.

Want a nikon sensor but don't need the high frame rate?  Don't want to pay
the Nikon price premium? Want all your lenses to become image stabilised?
Own minolta lenses?

I've seen just as many sony Zealots at work as canon and nikon.  Don't write
them off just yet.

I have.  I watched a pro (who uses an image of a medium format film
camera on their business card and for whom buying a P45 would be well
within the budget) shoot fashion using a Sony "point and shoot" (DSC-R1)
using studio lights - no flash.  My first thought was that the Sony was
pulled out to take some quick "preview" snaps in the style of using a
polaroid back in film days.  However, the entire job was shot with the
Sony, it was the only camera used, the work was published (advert.
supplement in a glossy).
But I don't expect a picture of the Sony on her business card any time soon.

That sounds like brand snobbery.

Even though I agree Canon and Nikon are the two defining companies, I've
seen pros using 35mm or digital versions from Pentax, Minolta, Fuji, and
Olympus. Obviously I've seen a lot of amateurs using them as well.

I expect Sony will sell quite a few cameras to amateurs who like the big
pixel count. And I'm guessing they will be selling the A100 through
chain stores where the average joe shops.

Sony wants a piece of the action, and their name carries a lot of weight
in the market, even if it is mostly fat.

*yawn*

That might be the case for an "amatur/professional" whatever that might be,
but many professionals and a few amateurs use Hasselblad and Linhof and the
like.

Phil used the 50mm Minolta f1.4 (not a Sony model), 50mm f1.8 Nikon,
50mm f1.4 Canon I believe. Though Phil doubts it has any relevance, the
50mm Minolta 1.4 tests out as an extremely high-res lens, particularly
in comparison with the standard Nikon, and might (I think) account for
some of the apparent gain in res of the A100 over the D200. The French
mag Chasseur D'Images found the opposite - a loss in res of about 20 per
cent, not a gain of 5 per cent.

The new Sony 50mm f1.4 is made in China, presumably by the Seagull/KM
facility, and may well not be in any way comparable to the Japanese made
Sakai-works Minolta 50mm f1.4. Or, it may be equal; the Chinese are
getting amazingly good, having acquired the old production equipment and
designs and been trained by the Japanese engineers.

David

DPreview:

"For direct comparisons we always use sharp prime lenses stopped down
(typically to F9). For the 'cropped' sensor cameras we used 50 mm lenses
(EF 50 mm F1.4 for the 20D and the Nikkor 50 mm F1.8 for the D200 and
D2X) and for the full frame EOS 5D the EF 85 mm F1.8"

The only "stand out" difference I've seen in the current 8mp+ dslrs is
that the Canon 5d with a good lens seems to me to offer noticeably
better pixel-level sharpness (and much less noise) - making the rest
from the 20d through to the expensive d2x look near enough to the same
not to really matter from an image quality POV - even arguing about
relative noise levels seems to be splitting hairs and subjective.

However, I've got a stack of Nikon lenses including some dx, and the 5d
is IMO a bit expensive just yet.

Actually Phil's test use RAW files for the resolution, and the camera's
post processing has no effect. A lower power AA filter, or a tighter
default convolution in the raw processor (gathering its information from
a smaller group of surrounding pixels to the target pixel location) will
tend to enhance resolution while also boosting moire or artefacts.

The images as processed by the A100, to JPEG, are pretty awful. I know,
I've used two, I'm getting one today, and I never intend to use a single
JPEG image - I have never used a camera JPEG for anything except an eBay
snap in the last five years, and I only own one pocketable camera which
can't shoot raw. Far from being over sharpened they are over-mushed with
noise reduction and tend to lose resolution compared to a D200 JPEG (or
indeed a Dynax 7D JPEG of best quality).

It is not the same sensor. The CCD layer may be identical, but the
colour filters, the microlenses, the AA lo-pass filter cover glass
assembly, and most importantly the channel feed off the CCD array are
all different. The Nikon uses a 4-channel feed for double the readoff
speed, the Sony uses a standard 2-channel 'one line at a time' readout.
The Sony may have similar colour filter values (we don't know, but it's
usual for each camera maker to specify the exact filter values
differently) and microlenses (quite likely) but has an entirely
different AA filter assembly using indium tin oxide coating which
affects static resistance; the AA filter appears to be weaker, or it
positioned closer to the CCD surface, as happens on the Nikon D50.

'Consumerist leanings' may affect the nature of sharpening in a
processed JPEG, but 'professional leanings' affect how you permit high
frequency, close to extinction, 'beyond Nyquist' detail to escape from
your sensor into a RAW file. Kodak became a champion of letting it all
through, eliminating the AA filter entirely. That was disaster. It's
very hard to post-process out sparkling rainbows and scattered pirate
gems of ruby, sapphire and emerald adorning every fine pattern or
texture in your pic. Canon in the 30D has gone for the opposite extreme;
don't let a single trace of detail close to the limit find its way into
the image, RAW or not. If it might create a false detail artefact, lose it.

Sony and Nikon both seem to be in middle ground but the A100 has been
made with slightly less removal of potentially damaging fine detail.
That suits me fine. I'm looking for the finest possible resolved detail
as a landscape and travel photographer. The D200 was pretty good, the
A100 promises to be better but only - only! - if the lens fitted and the
shooting technique match up to it.

David