Left AF (again) with a test result and news update pending

Started Jun 22, 2012 | Discussions thread
ForumParentFirstPreviousNextNext unread
Flat view
tashley
Veteran MemberPosts: 4,134Gear list
Like?
Left AF (again) with a test result and news update pending
Jun 22, 2012

First let me state that I am not extremely bothered by this: I rarely use any point other than Centre and sometime Right (when in portrait orientation) BUT I do on rare occaisions need the Left point and I do sometimes use multi point tracking. That said, the camera is incredible and I have sufficient experience to know that 'these things happen'. HOWEVER I do think that when they do, manufacturers should "fess and fix".

I have had a D800 and an E for many weeks now and never bothered to test for the dreaded issue until I recently got a 28MM F1.8G BUT I have also tested this with the 85mm F1.8 and the problem seems visible there too.

I sent my D800 to Nikon UK to get fixed around ten days ago, and today received notification that it has been attended to ('To Nikon Standard') and is on its way back to me, so I will test using this setup when I receive it and post the results.

My methodology takes into account the following:

1) There is no point whatsoever in shooting a frame focussed on an appropriate target using, for example, the Left AF Point and then judging sharpness of a subject on the same plane but elsewhere in the frame. That is the test that Tapes ran in his now (in)famous video and it neglects the fact that many lenses do not focus on a plane but rather have, at different places in the Field of View, different planes of focus. This is one of the reasons why one might want to use a point other than the centre one to focus: if your subject is off centre, then even if you focus using the centre point and then recompose perfectly, your subject will be focussed through a different part of the lens and that part of the lens might have a different plane of focus. Boring but true.

2) The thorny issue of an 'appropriate test target' - I have read Nikon guidelines (as well as shot hundreds of thousands of fames over forty years experience with many, many cameras) and I am satisfied that my target, a Spyder Lenscal, is appropriate and was appropriately used. In other words, though the AF area as shown in camera viewfinders is very often larger in fact than it appear in the viewfinder, thereby risking overlapping the target and 'seeing' something behind it, in my test the target greatly overlapped the AF area as displayed in the finder. And though the target has a 'geometrical pattern' it is not a 'regular' one.

3) I have tried the various 'alternative remedy' fixes such as as shooting with the point selector locked etc, and none of them works for me.

So here's the results: the Centre Point works. The Right Point mostly (and to me acceptably) works. The Left Point is very badly back focussed. You can't see that here but an 'in the field' shots shows that it is between a few and many inches depending on subject distance. The methodology is outlined in the image. I post a small version here. The full version here for a 100% pixel view is available here: (you'll need to right-click and select Download Original)

http://tashley1.zenfolio.com/p351773786/h2fbf1c4f#h2fbf1c4f

-- hide signature --
Nikon D800
If you believe there are incorrect tags, please send us this post using our feedback form.
ForumParentFirstPreviousNextNext unread
Flat view
Post (hide subjects)Posted by
(unknown member)
(unknown member)
(unknown member)
(unknown member)
(unknown member)
(unknown member)
(unknown member)
ForumParentFirstPreviousNextNext unread
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark post MMy threads
Color scheme? Blue / Yellow