[Rover]

And they're trying to exonerate the lens by stating that Klaus might have got a bad sample, though AFAIR bad samples actually manifest with tilted focus plane (and thus sharpness), not a uniform edge/corner sharpness falloff at one focal length?

It's EPZ's results that I actually find a little less believable (not for this lens, but in general). Let's wait for Lenstip to take a stab at this lens.