Sharpness issues with 16-35mm lens

[Richard3]

I am using 16-35mm 1:2.8 Canon EF lens. Being wide angle I have noticed the far left and right sides of the images are not sharp. Using auto focus around F11. Does anyone have an explanation. 

[ebiggs1]

It would be helpful to know what camera you are using but in a nut shell here is a reason. Lens diffraction.  This is caused by the aperture blades when you stop down a lens.  Any lens not just the 16-35mm. A perfect lens has no diffraction but we know that is not possible as nothing is perfect.  It is a myth that diffraction only occurs at small f-stops.  It happens at the very first f-stop even if it is f2.  It is just not as noticeable.  Most lenses have a sweet spot where all is in harmony.  This is usually f8 or f5.6.  All the bads are corrected and the diffraction is limited.  All is well with the world.

Another complication is the Bayer array. Green gets more attention than red or blue light. The interpolation of these colors produces the final full image. What happens when the diffraction limit is reached is, there will be the loss of green pixel luminosity. And the resulting loss of sharpness. ( overly simplified   )

If your lens is pretty sharp at wide open f-stop, you could be seeing lens diffraction.  It is also possible you have a decentered lens group.  That will require Canon Service.  I have not got to try that lens yet so I can't say first hand how sharp they are but

I would bet it is darn good.

[Richard3]

Thank you for your thoughts. I will try setting the lens at f8 or f5.6 and compare the results. The Camera I use is 5D Mark 11. Thanks again, Richard

[Waddizzle]

There could be another, simpler explanation...depth of field.  You are using a wider lens.  Are the objects at the edges of your frame the same distance from the lens as the objects in the center of your frame? 

Testing the lens at f/8 would reduce potential issues associated with depth of field.

[ebiggs1]

"Are the objects at the edges of your frame the same distance from the lens as the objects in the center of your frame?"

This would depend on focus point.  Assuming he focused on an object more close to the camera and not a distant mountain, I doubt seriously that is the issue.  Most likely it is diffraction or decentering.  Diffraction he can correct,  Decentering he can not.

Then again it may be all the better the lens is.  Is it version 1 or version 2 ?

[Waddizzle]

---

@ebiggs1 wrote:

"Are the objects at the edges of your frame the same distance from the lens as the objects in the center of your frame?"

This would depend on focus point.  Assuming he focused on an object more close to the camera and not a distant mountain, I doubt seriously that is the issue.  Most likely it is diffraction or decentering.  Diffraction he can correct,  Decentering he can not.

Then again it may be all the better the lens is.  Is it version 1 or version 2 ?

---

Yes, that's exactly what I meant by  "center of frame", the focus point.  I just assumed that the center was the focus point.

[Richard3]

Thanks for advise guys. The lens is series 11. The subjects are generally flat mostly interiors of properties for sale. Focal point generally towards the centre of the image. I think from advise given the issue may have to be resolved by taking to Canon for checking. 

[TCampbell]

It's hard to guess without sample images.

Actually ... even better would be to go outside, find a nice large flat brick wall, and take a photo of it so that the brick wall is literally filling the frame corner to corner... shoot at a low f-stop (so we have a very shallow depth of field) and (the important part) make SURE the camera is "parallel" to the plane of the wall (if the camera is angled then the distance to one corner of the frame will not be the same as the distance to the opposite corner of the frame and that would cause uneven focus even if the lens and camera are working perfectly.)

If you do that and then post the images we can evaluate if it looks like your lens is normal.

[Waddizzle]

---

@TCampbell wrote:

It's hard to guess without sample images.

Actually ... even better would be to go outside, find a nice large flat brick wall, and take a photo of it so that the brick wall is literally filling the frame corner to corner... shoot at a low f-stop (so we have a very shallow depth of field) and (the important part) make SURE the camera is "parallel" to the plane of the wall (if the camera is angled then the distance to one corner of the frame will not be the same as the distance to the opposite corner of the frame and that would cause uneven focus even if the lens and camera are working perfectly.)

If you do that and then post the images we can evaluate if it looks like your lens is normal.

---

Yes, do that.  Use a tripod, and take note of the distance to the wall in the center of the frame and the edges of it.

[ebiggs1]

:... find a nice large flat brick wall, and take a photo of it so that the brick wall ..."

I strongly advise against doing this.  It is almost useless on a ultra wide angle lens.  It won't tell you or indicate much to you.

You have one of two problems. Diffraction or de-centered.  One you can fix the other you can't.  My .02 cents and worth every penny.