Canon 135L

[Socrates]

This a comment for this page on the website

"... the Sigma 85mm f/1.4 EX DG HSM is noticeably sharper." (than the 135L)

This just shows how meaningless your new metric is. Your own measurements do not show any noticeable difference. There is almost none here: http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=108&Camera=453&Sample=0&FLI=0&API=0&LensComp=756&CameraComp=453&SampleComp=0&FLIComp=0&APIComp=2

Also, comparing 85 and 135mm can be tricky. For the same background blur, you need f/1.4 at 85mm vs. f/2 at 135mm (same physical aperture). The 135L is then really sharper, much less PF/LoCA.

BTW, nobody buys the 135L for sharpness - not that it is not sharp. It has gorgeous bokeh. Your "second hand" recommendation is really stupid.

[Nerval]

Sharpness has nothing ado with depth of field....

And even wide open, or at least at f2 central sharpness on the Sigma is quite impressive.

So yes the sigma is sharper than the 135L.

It is funny though how people complain about the irrelevance of a sharpness test, while they are actually complaining that their lens is better...

Both are telephoto portrait lens, so that's why DxO compare them.

Sharpness is not the most relevant factor for portraits, that's a given, but DxO does not pretend to test usability or practicality of a lens.

DxOMark tests resolving power, vignetting, light transmission and so on.

A kind of pure image quality test, be it relevant or not to the application.

If you want field tests and reviews, there are plenty of sources such as SLRGear, DPReview, photozone.de... and so on.

DxOMark is just a metric...

[Socrates]

You missed my point. DXO does not just report data, it makes a conclusion about the value of the lens overall. Then bokeh, f-stop vs. FL, etc., are primary factors.