Scott Stoness Offline Upload & Sell: On
|
p.1 #14 · p.1 #14 · Dxomark way out of date on Lens - No 5dsr, No a7rII | |
gdanmitchell wrote:
I understand the GIGO concept, but in the case of DXO I think the issue is something else. The input data are what they are and the output conclusions are not exactly wrong. The problem is more that people don't understand what the data from DXO mean, yet they jump to conclusions that the ratings are telling them things that the ratings can't say.
Dan
I agree with Dan - Dxo data is useful and leads to questions that are not easily answered:
Example - if you compare Sigma 35/1.4 on d800 vs 5diii it yields 23/6 (mpx/ca) vs 17/6 respectively. Similarly zeiss 25/2 yields 22/5 on d800 and 15/5 on 1ds3. This suggests, there is some significant benefit to more mpx (not surprising) . However, surprisingly, the d800 does not even get close to 36mpx, which suggest the resolution is lens limited. Which on its own suggests in the wide range lens the benefits of an uwa are not great because the lens are the limiting factor.
However if you select the d800e, it achieves 30/6 - which in turn suggests that the resolution is not lens limited, but limited by the aa filter in the d800. Most dxo scores on d800e and d810 are 1/3 higher because they do not have the aa filter. And that leads to the question, is it real resolution or just false resolution because these tests are done with lines as opposed to more complicated test patterns ( that monito should like to know because he chose the 5ds vs me choosing the 5dsr )
And even as much interesting, the a7r with the best available lens in this range (sony 28/2) only achieve 24/18 (mpx/ca) which is a surprising outcome because the a7r does not have a aa filter and you would expect it to be able to achieve close to 30. Which leads to the question of whether the sony lens are just not as good in this range or is there some goofy choice that sony made that limited resolution in the wide range. Which questions whether all the people who moved to sony because of greater resolution in the time when canon did not have a high resolution camera were mislead by the available data. And then dxomark shows that in the 50mm range the a7r does perform to is 36mpx achieving 29/9 with the sony 55/1.8. So not the outstanding question is whether the a7rii fixes the problems of underperforming of a7r at wide mm or it remains.
This can only be answered by more data from a7rii and 5dsr from Dxo.
But ultimately, I agree with Monito, too much seeking of data ignores the other stuff like - familiarly with canon, battery life, the fact that I have lots of canon lens, and that mpx is only one measure of a good lens.
But I find it interesting.
|