« Last post by CreeDorofl on February 15, 2017, 07:43:16 pm »
Like seemingly everyone else on dxomark, you don't get it.
The overall score isn't a "lens score", it's a "lens+sensor combination" score.
That's why the score changes every time you choose a different camera from the "mounted on" dropdown.
So why is the Canon 5DSR version rated lower? Because you're paying for 50 megapixels, but despite excellent glass you're only getting 40 megapixels worth of sharpness / detail. And that loss happens with every Canon sensor - Attach it to a 24 megapixel 760D, and you only get 18 megapixels of detail. Put it on a 7DmkII, and you only see 14 megapixels of detail on a 20 megapixel sensor.
But if you get this lens for virtually any nikon camera, you get the absolute maximum level of detail that the camera is capable of. Use a d800 and you get 36 megapixels. Use a d500 and you get 19 out of 20. Use a D4 and you get the full 20 megapixels from its sensor.
Basically, the lens is being held back by Canon's sensors, which lag behind Sony and Nikon. That's not Sigma's fault, but the point is to pass info to the consumer so they can make an educated decision. If I have an older D800E and a brand new 5DmkIV with virtually identical megapixel count, I need to know that I'm losing about 7 megapixels of detail if I choose the canon mount of this lens.