artur5
Offline
• • •
Upload & Sell: Off
|
 p.40 #17 · Fujifilm X-Pro1 interchangeable lens | |
douglasf13 wrote:
That's the internal question that I'm struggling with. As a current M lens user, this is how I see things, preliminarily:
GXR: great sensor performance with M lenses (even more when the 16mp module comes,) ok add-on tilt EVF, good size
Nex-5N: good sensor performance with M lenses, fantastic add-on tilt EVF, good size
Nex-7: seemingly not so great performance with M lenses as a whole, fantastic built-in EVF, good size
X-Pro1: probably great sensor performance with M lenses, probably ok built-in EVF, a little larger
Douglas, concerning the performance if RF wideangles on the X-Pro-1 I would be pleasantly surprised if they are as good as with the Ricoh GXR. but my bet is they will suffer the same or worse issues than the NEX-7.
http://www.dpreview.com/previews/fujifilmxpro1/page3.asp
If you look at that diagram of the Fuji 18/2 mounted on the camera showing the lens layout and the light path, you'll see a large converging rear element which directs the rays almost perpendicular to the sensor. Thus, we can infer that the sensor doesn't likes acute incident rays at all. Another inference being that the 18/2 is a strong retrofocus design, because a large rear element, almost as large as the sensor, even if it's very close to it, means a very distant exit pupil, like a WA. SLR lens.
We know also that retrofocus lenses, while solving the problem of vigneting, color drift and corner smearing, need a complicated, bulky and costly design if we want to keep the sharpness high and the distortion/CA aberration low.
That Fuji 18/2, while not a pancake, is compact, quite fast at f/2 and not expensive. Retrofocus + WA + fast + cheap + compact = quality not stellar. ( at best.)
Time will tell, but the precedent of latest Fuji lenses for digital Hasselblads with heavy in camera processing, seems to point that their choice was to design affordable and compact lenses for the X-PRO-1 and use in-camera lens correction to make up the 'barely good' performances of their new lenses.
Leicas M8/9 correct the color drift and vigneting for their WA. That means increased noise in the corners but nothing else is compromised. Now, if we need to correct also CA and distortion, the RAW files will be hit noticeably. because the software will have to interpolate, add or subtract pixels, etc. For instance, I have a Canon FD 24/2 -a good lens but has moderate laCA. If I correct this in Photoshop, there's a visible blurring of fine details in the outer part of the image.
I'd like to be proved wrong, but it's possible that the benefit of not using an specific AA filter will be defeated by those in-camera lens corrections.
|
A laser projecting a grid would work too, by detecting parallax, but shining lasers into people faces every time you take a picture might be a bit offensive. 
, so that would mean two cameras 
) this looks to me like the best light system out there. If they'd put IBIS in it I'd call it perfect.