Strange resolution numbers at f/2.8 (especially away from the center)

[simont]

This a comment for this page on the website

My previous message was a response to Xzeihoranth, but that's not too apparent from the formatting.

I also have one further question of my own regarding this lens test. I noticed that the measured resolution values at f/2.8 are really low, much lower than at both f/2 and f/4. Could this be a measurement error? If not, what could explain such behaviour and is it common?

Kind regards,
Simon

[cccx]

Hi Simont, I agree it is strange, I would also be interested in knowing why this is.

However, it appears to occur on all the camera bodies they tested this lens with.

I was thinking it could be caused by any of the following reasons:

• -a very 'different' testing methodology by DxO (perhaps taking into account more vertical resolution than any other lens testing site)?
• -individual lens fault?
• -lens design
• -a measurement error by DxO (unlikely if it occurs on all camera bodies they tested)?

Note that the SLRGear test did not show any such behaviour:
http://www.slrgear.com/reviews/showproduct.php/product/1443/cat/14

The Lenstip review linked to by Xzei showed f2.8 to be one of the highest resolution apertures:
http://www.lenstip.com/316.4-Lens_review-Olympus_M.Zuiko_Digital_45_mm_f_1.8_Image_resolution.html

So there are some very different test results for this lens.