Interesting article on DxO about 5D III and D800 resolving power

Started Apr 4, 2013 | Discussions thread
Slideshow Bob
Senior MemberPosts: 1,114
Like?
Re: The article premise is a lot of hooey about nothing.
In reply to joger, Apr 6, 2013

joger wrote:

Rick Knepper wrote:

. . . compare systems like the D800E/24-70G to the 5D3/24-70 II since the 24-70 II is such a great lens. In November, I shot the 5D2/24-70 II vs. the D3x/24-70G and much preferred the Canon files (mostly due to the lens). I am also confused by the reference to the P-Mpix score in an artcile about cameras since this score is lens related.

That's what it's all about - you always see a system performance.

Sure it is of academic interest to see which sensor performs better but the end of the day I am only interested in the results.

So on the Nikon you can use Nikon or 3rd party lenses. On Canon all of them with limitations plus Canon lenses of course.

Canon has introduced some astonishing lenses lately and with them together the system performance is virtually on a similar or even better level then the D800(E)

I posted some comparisons and only the 85 f/1.4 outpeformned my 5D II plus lenses.

the difference between the 5D II and III is negligible at lower ISOs - in fact the best performing lens  has a difference of 1 MPixel at DXO's rating (which is doubtable to say the least) But 1 Mpixel more is exactly the specified difference and thus it could make some sense even though the AA filter in the MK III is stronger and other test sites came to other results.

What I would like to say is that only the real world results count and I can only echo the OP - a IQ160 wins in every side by side comparison simply due to it's shier pixel count in combination with the larger sensor. If I shift and stitch to the virtual sensor size of the IQ160 (which is possible with the TS-E 17 and 90 I own) I get stunning results which should be close to the IQ160 back in pixel count and (virtual) sensor size but the most determining factor are the lenses used.

It's all about the lenses and the cameras nearly don't count - cameras are disposals - lenses are keepers. We are not at the end of the sensor development and Nikon did a great job with introducing the D800(E) - yet they have some homework at their lenses and maybe their lens mount too

I can see what you're getting at, but ultimately, I don't think you can dismiss the camera quite so easily. It depends on whether you're looking at sharpness or relative resolution, and also on your source of info.

For the sake of argument...

if you look at the lens reviews on photozone.de, they test FF Canon lenses on a 5DII and Nikon FF lenses on a D3x, so it's 21mp vs. 24mp. The latest 70-200 f/2.8 lenses from Canon and Nikon both show pretty good MTF50 results, but the Canon is clearly a better lens since overall it comes closer to the resolution limits of the camera. However, the relatively small 3mp difference between the cameras means that the Nikon out-resolves the Canon at 200mm pretty much over the entire frame at all reported apertures. So the system performance is slightly higher on the Nikon, but not because of the lens.

I think the article is nonsense. It doesn't take a genius to figure out that bad lenses on a good body won't give you a great result.

I think perhaps a better example of how the cameras perform exists right here on DPR. Load up the  5DIII and D800E into the RAW studio scene comparison widget, and the difference in the fine details is about what you'd expect.

SB

Reply   Reply with quote   Complain
Post (hide subjects)Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark post MMy threads
Color scheme? Blue / Yellow