Forum Twitter Facebook Google + RSS

Author Topic: Lenses scores calculation  (Read 3583 times)

Offline Vadim.Leb

  • Newbie
  • *
  • Posts: 18
Lenses scores calculation
« on: July 11, 2013, 12:35:15 pm »
This a comment for this page on the website
I have several questions about dxomark score for lenses. As I understand dxomark score calculates taking into account Sharpness in P-MPix, CA, Vignetting, Transmission, and in the same time vignetting and CA taken into account when you calculate sharpness. Does it mean that bad|good CA and vignetting affects final score twice?

Other question about sharpness of Nikon vs Canon lenses. In most cases Nikon lenses less sharp than Canon and according to tests of lenses made by Sigma\Tamron this caused not by lenses but because of some difference in sensor (may be AA filter). It means that canon lenses always tested on "sharper sensor" and nikon lenses on not so "sharp sensor". So intresting what results will show lenses if test them on body of other vendor (for example nikon`s 70-200 on MarkIII and canon`s 70-200 on D800).

Offline Emilie_DxOMark

  • Hero Member
  • *****
  • Posts: 530
Re: Lenses scores calculation
« Reply #1 on: July 23, 2013, 08:35:24 am »

Thanks for your question, perhaps we will publish a detailed documentation for DxOMark score and Sharpness score… There is a lot of misunderstanding between our different scores. You can check our score documentation:

Shortly, here are some key points about our scores:
- DxOMark score depends on sharpness, distortion, vignetting, TStop, sensor color depth, sensor dynamic range. This score is computed for a mid-light using condition (150 Lux and 1/60s)
- DxOMark score is not a metric computed with the optical score. They are independent even if they use the same data.
- DxOMark score takes measurement of sharpness (acutance) but not the sharpness score.
- Sharpness score is computed only with sharpness measurement and do not take into account other aspects.

Your proposition is interesting, we will see if we have time to do it (more likely we could test a prime).

Best regards,

The DxOMark team