Lens test - Is this normal

[SteveL54]

I ran the same test at 5-7 feet. This lens shouldn't have any
problems at this distance. The images simply do not seem tact sharp
like some of my other lenses.

Agree. At that distance this lens should be quite sharp -- even wide open. Although it should be sharpest around f/4 in the center.

How did you verify focus at these longer distances? The focus test chart indicated good focus accuracy at very close range. At five to seven feet the AF might be off causing the softness when wide open.

When I ran sharpness tests with my D70 I used a focusing rail to tweak the distance between bracketed shots once focus was locked. With the D300 it's much simpler. Manual focus with liveview in tripod mode at 100%.

Steve

 

[PixelDave]

I am using a split prism focus screen. Focus is definately on.

I ran the same test at 5-7 feet. This lens shouldn't have any
problems at this distance. The images simply do not seem tact sharp
like some of my other lenses.

Agree. At that distance this lens should be quite sharp -- even wide
open. Although it should be sharpest around f/4 in the center.

How did you verify focus at these longer distances? The focus test
chart indicated good focus accuracy at very close range. At five to
seven feet the AF might be off causing the softness when wide open.

When I ran sharpness tests with my D70 I used a focusing rail to
tweak the distance between bracketed shots once focus was locked.
With the D300 it's much simpler. Manual focus with liveview in
tripod mode at 100%.

Steve

 

[SteveL54]

I am using a split prism focus screen. Focus is definately on.

I'm just eliminating possibilities here, but things aren't looking that good for the lens.

Split prism focus screens can be off. But I assume you would have noticed this with your other fast lenses.

Split prism focusing is not necessarily more precise than other methods. In order to work for f/5.6 lenses they don't take full advantage of an f/2.8 lens baseline. Additionally, split prism focusing can be thrown off by the spherical aberration in a lens.

I'd still do some slight focus bracketing. Or split prism focus on an element within a target that's slanting away. Then compare the sharpest point on the final image with the element you focused on.

Steve

 

[tarkan]

Your focus points look spot-on. The lens is just soft wide open as mentioned by others. Try doing the test again but stopping down to f4 and see how that effects edge sharpness and contrast.

Tarkan
--
------------
My blog and gallery
http://www.tarkan.info

 

[PixelDave]

I guess my lens was simply crappy.

I receive a new version of the lens it appears significantly better. Center sharpness is virtually the same at f/2.8 and f/8.0.

New comparison:



Old comparison:

 

[NRJ]

looks good to me, but both Nikon and Canon indicated that the 45
degree test shot is not an accurate way to test a lens.

Please specify a link to both company's statement.

Steve

here is what nikon says.


The "AF Fine Tune" feature allows users to make focus adjustments to compensate for minor differences in their particular lens and camera combination. In general it is not recommended to adjust the AF Fine Tune unless you are having a specific focus issue, as unneeded adjustments can decrease the overall focus quality of the camera.

There are many ways to test and set the AF Fine Tune but one effective way is to reproduce a "real life" situation where a problem is suspected. For instance, if you feel that the focus is incorrect when shooting close-up subjects, you should reproduce that type of shot.

Place a subject with depth (a ruler works well) into the scene and focus on a pre-defined spot. Choose "AF-S" focus mode, make sure that the focus point is on a part of the subject that can be easily focused upon; see the camera manual for tips on autofocus point selection. Take a photo and review the results on the computer.

If some part of the photo other than the selected focus point is sharper than the focus point use the AF Fine Tune adjustment. Adding a "+" correction will move the focused point away from the camera and would be used in a "front focus" situation. A "-" adjustment moves the focus closer to the camera and would be used if the test photo showed the focus behind the selected point.

The "±20" is an arbitrary scale and does not represent any actual distance. Trial and error will be needed to determine the best setting.

Up to 20 different lens or lens converter adjustments may be saved. Zoom lenses should be tested at the middle of their focal length range unless a specific error is seen at one particular setting or the lens is always used at particular zoom setting,

If the AF Fine Tune does not increase the sharpness it could indicate a more serious problem, and the lens and camera body should be sent to Nikon Service for evaluation.

 

[r.kleineisel]

The general suggestion for lens testing is that target should be flat
and the camera should be perpendicular to the target surface.

this is not usefule if you want to see the depth of field behaviour and front- backfocus issues.

--
Regards, Ralf

 

[SteveL54]

looks good to me, but both Nikon and Canon indicated that the 45
degree test shot is not an accurate way to test a lens.

Please specify a link to both company's statement.

Steve

here is what nikon says.


The "AF Fine Tune" feature allows users......

That quote doesn't mention the type of test that the OP used is invalid.

 

[SteveL54]

I receive a new version of the lens it appears significantly better.
Center sharpness is virtually the same at f/2.8 and f/8.0.

Thanks for the update. That's some improvement.

Steve