D700/D800 + new 28 1.8G. Hmmmmmmmmmm.

Started May 25, 2012 | Discussions thread
marike6
Senior MemberPosts: 5,070Gear list
Like?
Re: Look again
In reply to Flashlight, May 26, 2012

Flashlight wrote:

marike6 wrote:

Dominique Dierick wrote:

Yes you can. I detected it in the field with my 16-35, very weird behavior.

Where is the de-centering in this brick wall shot at the same f8 aperture?

http://www.flickr.com/photos/russellbarnes/7266785696/sizes/l/in/set-72157629890945240/

It's not there.

Look again and remember this was shot @ f/8!

You can clearly see the right side is softer, although IMO brick walls are a poor target as you do not have much detail.

The post above with the crops of the fence shows nothing since there a way too many variables in such a shot. You are assuming that the fence posts are precisely the same distance from the camera.

Assuming the focus distance in the fence shot was 5 meters you'd have a DOF from 2 meters to infinity. Although such a standard calculation doesn't hold too much water looking at D800 shots @ 100% due to the magnification, a slight difference in distance of the fence posts at such a distance should never have such an impact, even wide open.

You need a Siemens Star Chart to test for de-centering. A brick wall is not an accurate way to test it at all.

Re: DOF, You're suggesting that with DOF 2 meters to infinity all objects that fall under that range will be razor sharp. This is not the case. DOF means objects falling with a given DOF range will be acceptably sharp . However, you'll get a decrease in sharpness in objects that are outside of the focal plane. At normal viewing distances objects appear sharp, but at 100% you'll see varying degrees of sharpness relative to the focal plane.

It's certainly possible that his lens has a de-centering issue, however, the brick wall and random crops do not prove that it does. The camera back could have been angled slightly from the wall or vice versa.

There are way too many variables to draw a conclusion here.

The only way to test for de-centering is with a Siemens Star Chart. The process is described here:

http://www.lensrentals.com/blog/2012/05/testing-for-a-decentered-lens-an-old-technique-gets-a-makeover

Only with flat objects can you detect de-centering. That is why lens test websites like Lenstip and Photozone use test targets that are set up as 100% perpendicular from the camera. Testing is about eliminating variables which in a natural scene like Russ's you cannot do.

I disagree with that. If you pick your target carefully, meaning detailed objects at different distances and a rough surface on the ground, like grass or sidewalk tiles, you can see the plane of focus within the depth of the image.

The reason for standardized studio tests on flat targets is mainly useful when you want to compare different lenses or cameras between shots. When shooting very close the visible differences between sides of a decentered lens are much harder to see than when you shoot at infinity. Look at the treetops in the background in the fence shot and remember it was shot @ f/8 with a DOF extending to infinity.

-- hide signature --

Philip

 marike6's gear list:marike6's gear list
Nikon Coolpix P330 Panasonic Lumix DMC-GH2 Nikon D800 Fujifilm X-E1 Panasonic Lumix G 20mm F1.7 ASPH +7 more
Reply   Reply with quote   Complain
Post (hide subjects)Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark post MMy threads
Color scheme? Blue / Yellow