On 4/03/2012 4:43 p.m., Rich wrote:
(Floyd L. Davidson) wrote in
:

wrote:
Then why is there more "visible" noise in the D4 images on Dpreview?

http://www.pbase.com/andersonrm/image/141858950

You don't seem to be able to tell what is "more" and
what is not.

I personally can't tell from those image if there is
more noise or not. All I can tell is that the images
are *clearly* not comparable by visual inspection. They
are not taken with exactly the same light, they are not
normalized to the same pixel dimensions

That is a B.S. tactic. "Normalized." Take a 500,000 pixel area from BOTH
cameras, the D3s is CLEANER. If you can't see it in the images, you are
blind.
No it isn't.

If you scale the D3s image to 4928 pixels on the long side, it's very
clear that there's more image noise in the D3s sample.
here (converted from original raw files):
http://i39.tinypic.com/2v26yxu.jpg
This also reveals another problem with DPReview, there is someone there
who is "on the ****" (colloquialism = "likes to have a drink,
understatement"), and wasn't very careful with hiding their tracks.
This might explain their sloppy test technique:
The D3s sample was taken with the 85mm f1.4 d @ f8
The D4 sample was taken with the new 85mm f1.4 "g" @ f11
The camera settings were different (D3s image even has high ISO NR
toggled on and other differences), but result from raw would depend on
what software they used for raw conversion. It seems that some
sharpening was applied to the D3s image (not by me in scaling).
I could go further into this, and do another raw conversion, removing
all "camera settings", but I can't be bothered wasting any more time.
I've seen enough to show that the comparison shots between camera models
at DPReview aren't very useful.