DXoMark

http://www.dxomark.com …enses/Camera-Lens-Ratings

Does anyone have comments about the way this site rates Canon Lenses?

Would like to hear your thoughts.

Thanks

Rating is not really interesting, I would even say that each lens can be used with different ways and be perfect for some uses and less than good for other.
They don't test Auto focus speed and accuracy at all, which can be really important (size, weight, colours are other parameters important not taken in account) .

BUT, I would say, that we could say that their measures are much more scientific than others. As far as I could compare with my own experience with my lenses (non scientific) I really trust their sharpness profiles, vignetting and CA fields map.
They only tested a few lenses, but measurements are really interesting ... scoring?? I didn't understand why they try to do it.

Ditto.

If you see anything with DXO in the name, it probably involves lots of number-crunching that has little to do with real photography. The only two sites I visit anymore if I want to get details on a lens are The Digital Picture and Photozone, both have objective, subjective and in-use examples.

I peeked around this site for the first time yesterday and found the 70-200 II review to be off-pudding to me. It did not score very high, and that lens is incredible. I will take their reviews with a big box of kosher salt. I prefer photozone and digital picture. I did not agree with a few of the reviews for many of the lenses I have owned.

It looks like they test at one subject distance only which, to me, means the high level of precision and detail they provide is of limited use. In my experience lenses do not perform the same at all distances. I've often compared lenses and found that the better lens at short to medium distance is not necessarily the better lens at long distance.

I like the DXO site but then I am an engineer and I do want to know the cold hard numbers. I agree with others that there is more to a lens than just it's optical properties but I find their data very useful as it removes emotion from the equation.

However, I would never use just their data in selecting a lens. It's a great data point but that's all it is. You still need to couple their data along with other review sites to get an overall picture on how the lens will peform in the real world.

Personally i have always found SLRGEAR to be accurate in there testing and rely on them & Photozone prior to purchasing or renting.

http://www.slrgear.com​/reviews/index.php

Regards, Ron

There's definitely something wrong when the ultra-sharp EF 70-200 f/2.8L IS II scores lower on resolution than the original 70-200 f/2.8 IS, or even the EF 28-135 IS. It's also suspicious that the excellent EF 100 Macros (L and non-L) scored so low in resolution. Numbers are fine, but garbage in, garbage out. I suspect something is amiss in their testing procedures.

I agree... Many of the lenses tested make little sense.

Thanks for the replies.
Made no sense to me either.

Thanks for the links posted to other sites.

Karl

DxO measures something very rigorously. The problem is, whatever it is that they are measuring has nothing to do with what I'm looking for as far as I can tell.

I've compared their sensor data with cameras I've actually used and all I can say is that their ratings never match my actual experience. The biggest one for me was when I directly compared my BIL's Nikon D90 (a camera that was absolutely top rated at the time) against my 5D and 1D3. My experience with these three cameras did not match DxO's ratings in any way.

Now I look at their lens ratings and it's more of the same. What are they rating? It sure has nothing to do with what I'm looking for.

I have a 85 1.8 and I also have the Big Dog. 70-200 2.8 II and the rankings were ridiculous.

I mean, they have the 85 1.8 as the highest.
I know it's nice but.......

Is there something else going on with those guys or is it just a big mystery?

JeffreyG wrote in post #13178831
Now I look at their lens ratings and it's more of the same. What are they rating? It sure has nothing to do with what I'm looking for.

They definevery clearly their test methodology. Their scoringexplained as well. The data is geared towards engineers and not photographers but it is an accurate data point. As I said earlier, it wouldn't be the only source I would use, but I don't discount it either.

I don't believe that DxO really 'rates' lenses. The purpose of their efforts is too make appropriate 'calibration' of the match between camera bodies and lenses and then in their software adjust for best image quality based on their calbrations.

gjl711 wrote in post #13179114
They definevery clearly their test methodology. Their scoringexplained as well. The data is geared towards engineers and not photographers but it is an accurate data point. As I said earlier, it wouldn't be the only source I would use, but I don't discount it either.

Except I am an engineer. And their ratings do not match my own real experiences.

I'm more familiar with their sensor ratings than their lens ratings. I have owned or used several of the cameras they rated, and I know that if I make equal sized prints from several of these and directly compare the prints, it is very quickly obvious that the DxO ratings on sensors have no bearing on this outcome.

So they are indeed rating something. But it does not match what I am using the cameras for.

My suspicion on the camera ratings is that their values are related to per pixel performance or some such meaningless thing. I can tell you for sure than in print their better rated sensors often are much worse than lower rated sensors in direct comparisons.

I think DxO has a major internal flaw in creating measurements that correlate with what photographers want.