| Introduction | Measurement | Comparisons | Conclusion |
Taking the overall DxOMark score of 19, we can see that the lens is not an especially high performer in terms of overall image quality. However, looking at the scores in detail, we can see where the lens fall down.
Starting with the sharpness score, a rating of 13P-Mpix when attached to an EOS 5D Mark II is not that high when the nominal rating of the camera is 21P-Mpix. However, to put this into context, it is expected that zoom lenses will score lower than prime lenses due to the increased complexity in manufacture and the extra compromises required in lens design. Even the high scoring Canon EF70-200mm f/2.8L II IS USM lens only managed a sharpness score of 18P-Mpix when attached to an EOS-1Ds Mark III, a camera with a very similar nominal resolution.
Although the score itself is not that high, the main issue with the sharpness is the lack of homogeneity cross the frame. At all focal lengths when shooting at the maximum aperture, there is a fall off of sharpness across the frame. This is most noticeable around the 50mm focal length. Closing the aperture down to f/5.6 helps to alleviate the problems, but there is still a fall off. By f/8, the sharpness is close enough to even across the frame to not be a concern. Even so, the 50mm focal length setting is still the one that offers the lowest score.
The transmission score, on the other hand, shows that the maximum aperture of the lens (f4) matches identically with the real life test results. Although a score of 4TStop seems low compared to a prime lens like an EF50mm f/1.2, a score of 4 for a zoom lens is very good showing there is essentially no light lost as it passes through the lens.
Like the transmission score, the distortion score of 0.2% is especially good for a zoom lens, with only the slightest barrel distortion at the wide end of the focal range and an almost completely neutral result at the 70mm end.
The vignetting score is one area that has impacted the overall DxOMark score. Only at 70mm and at aperture of f/8 is there almost no vignetting. At all other settings, there is some corner shading or light loss at the edges of the frame. At 24mm and f/4, the score of -1.3EV is too strong for a lens in this class.
Finally, the chromatic aberration score of 8µm is actually very good. A score of 5µm is considered noticeable, but would only equate to around 1pixel on most cameras – certainly nothing that would be distracting in a final print.
Beyond the tested metrics, both the USM AF motor and the advanced Hybrid Image Stabilizer are two plus points in favor of this lens. The IS system is a hybrid unit, first seen on Canon’s EF100mm f/2.8L IS USM Macro lens. The benefit of this is that whether you are shooting very close up to your subjects in a macro fashion, or whether you subject is more distant, the IS will function effectively to reduce camera shake while hand-holding.
|
|
Added by ag08025 |
February 16
I do not believe
Most likely incorrect tests.I'll explain why DxO does not provide graphical results of the tests but only their personal considerations.So as another to drive the conclusion
Reply |
|
|
Added by ag08025 |
February 08
I do not believe
because your tests are so far from reality?
http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=823&Camera=453&Sample=0&FLI=0&API=0&LensComp=805&CameraComp=453&SampleComp=0&FLIComp=0&APIComp=2 Reply | Read all replies for this comment |
|
|
Added by ageha |
February 10
Re: I do not believeQuote: <div id="linkdxomark">This a comment for <a href="http://www.dxomark.com/index.php/Lenses/Camera-Lens-Database/Canon/Canon-EF-24-70mm-F4L-IS-USM">this page on the website</a></div>because your tests are so far from reality? http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=823&Camera=453&Sample=0&FLI=0&API=0&LensComp=805&CameraComp=453&SampleComp=0&FLIComp=0&APIComp=2 Maybe sample variation? ;) Reply |
|
|
Added by RIVERAKTRE |
April 05
Re: I do not believe
Interesting point but you have to remember you're not really comparing apples to apples here. The link you provided compares a Canon f/4 VS Sigma f/2.8 which isn't a true apple's to apple's comparison. More like a peach and orange.
DXO Lab tests thousands of lenses under Strict controlled environments, I highly doubt they're fluffing up the test results. PLUS "IF" they were...what are they gaining from doing so? Something to think about. Reply |
To provide photographers with a broader perspective about mobiles, lenses and cameras, here are links to articles, reviews, and analyses of photographic equipment produced by DxOMark, renown websites, magazines or blogs.