How to make sense of DXO Mark's varied lens sharpness ratings for different body combos!?

[sid_19911991]

Hi folks,

I have no idea on how the sharpness ratings are awarded to different lens & camera body combos.

I would like to get a general non-technical idea on how to interpret the numbers.

For instance -

The D800E paired with Nikon AF-S NIKKOR 300mm f/2.8G ED VR II scores 33-P Mpix sharpness rating
while a D500 with the same lens scores a measly 14-Mpix.

Do refer to the link below for the details. One can toggle between different lens & body combinations.

https://www.dxomark.com/Lenses/Niko...0mm-F28G-ED-VR-II-mounted-on-Nikon-D500__1061

Does it mean, telephoto primes produce significantly sharper results when paired Full frame bodies as opposed to when used with crop sensor bodies? I am confused.

I would appreciate your opinion about it. Thanks.

 

[StefanSC]

Making sense of DXO ratings for anything is simple:

You go to the window with the rating on their website, find the big X in the corner of the window and you press it. Then you go out shooting.

I've found time and time again that their tests and ratings have no real usefulness in real life shooting that I don't bother with their ratings.

OTOH, DXO Pure Raw is impressive, if slow.

 

[thelordofthelight]

Exactly what Stefan said...dxo, lensscore are all just about bragging rights and have very little significance on the real world performance...

As to your qn on telephoto lens sharpness on Fx vs DX, the opposite is true because a tele lens on a DX body will make use of the creamy central area of the lens where it is at its peak of the optical performance.

 

[sid_19911991]

Oh boy. I wonder who sponsors DXO mark for them to do these tests.

Yes the numbers look puzzling. But surely they must have some relevance.

Surely there must be some consistent logic & measurement standard behind the numbers.

 

[bleirer]

Here is how they do it. I think what you are noticing is the 'perceptual megapixel' part, where things get equated to the same output size. I guess that is one reason to shoot full frame. So the 21 megapixel d500 would be perceived as a 14 megapixel camera with that lens and the 36 megapixel d800 would be perceived as 33 megapixels. I guess you have to compare the same body to compare lenses.

DXOMARK lens testing protocol and scores

For our DxOMark lens reviews, we evaluate the performance of interchangeable lenses for cameras equipped with sensors that can capture images in RAW format. In this article, we explain how we test for different criteria in the DxOMark image quality test lab and how the test results translate...

www.dxomark.com

 

[aolander]

Here is how they do it. I think what you are noticing is the 'perceptual megapixel' part, where things get equated to the same output size. I guess that is one reason to shoot full frame. So the 21 megapixel d500 would be perceived as a 14 megapixel camera with that lens and the 36 megapixel d800 would be perceived as 33 megapixels. I guess you have to compare the same body to compare lenses.

DXOMARK lens testing protocol and scores

For our DxOMark lens reviews, we evaluate the performance of interchangeable lenses for cameras equipped with sensors that can capture images in RAW format. In this article, we explain how we test for different criteria in the DxOMark image quality test lab and how the test results translate...

www.dxomark.com

So does this "perceptual megapixel" part have anything to do with real life results?

 

[sid_19911991]

Here is how they do it. I think what you are noticing is the 'perceptual megapixel' part, where things get equated to the same output size. I guess that is one reason to shoot full frame. So the 21 megapixel d500 would be perceived as a 14 megapixel camera with that lens and the 36 megapixel d800 would be perceived as 33 megapixels. I guess you have to compare the same body to compare lenses.

DXOMARK lens testing protocol and scores

For our DxOMark lens reviews, we evaluate the performance of interchangeable lenses for cameras equipped with sensors that can capture images in RAW format. In this article, we explain how we test for different criteria in the DxOMark image quality test lab and how the test results translate...

www.dxomark.com

Nice article. Although I will need a simpler version of it to understand it better.

The MTF & perceptual megapixel parts are what I am interested in learning about.

I wonder if say a D850 can get more sharpness out of a prime lens like 500 pf or 500 F4 as opposed to a D500 with all other variables being identical or nearly identical.

The DXO Mark ratings say so.

((The analogy I can give is a powerful digital audio player like FIIO X3 (camera body) driving a powerful 300 USD audio technica earphones (lens) way better than a smartphone. If anybody tried this you will know the headphones sound like a 20 USD headphones when paired with a smartphone. Basically the smartphone is not good enough to make best use of the headphones.))

I think Z7 (no AA filter & high MP) with the 58 mm 0.95 f lens has got the highest lens/body combo of all the combinations tested by them.

That sounds right.

Also, Canon 5DSR (49 mp no AA filter) gets the highest rating in the Canon line-up.

That seems right too.

I guess there is some truth to those numbers.