Big Three Shootout on luminous landscape, iso performance comparison

[photos of hans koot]

<p>http://www.luminous-landscape.com/reviews/cameras/big-three.shtml</p>

<p>Sony A900 Nikon D3X and Canon 5d MKII for the interested here.</p>

[Matt Laur]

<p>I did get a small chuckle out of his early remarks about the challenges in comparing RAW files from different manufacturers' equipment. Seems only fair to use each maker's own designated software, of course. But he complains that Nikon's Capture NX2 is "pricey." Compared to an $8,000 camera? :-)<br /><br />Still, an interesting read and qualitative results. <em><strong>Spoiler Alert!</strong> </em>At or below ISO1600, he finds the 5DII, the D3X, and the A900 to be indistinguishable in ter ms of noise. The Sony falls behind when you push higher, while the Canon and Nikon remain virtually identical (and very good) right on up to 6400. He likes the Nikon's build and usability much better than the Canon, but likes the Canon's price proposition better than the Nikon's.</p>

[phule]

<p>[[seems only fair to use each maker's own designated software, of course]]</p>

<p>Therein lies all the problems with testing. Some days it seems like Adobe ACR has the advantage and that camera companies don't even understand their own RAW files (I exaggerate, of course), and other days it seems like ACR doesn't know what to with the files from all these newfangled "digital cameras" everyone is talking about. </p>

<p>(Of course, the reality is that the differences are often subtle at best but it sure does get the pixel peepers all hot and bothered.)</p>

<p>As an aside, I could not imagine trying to juggle 3 different camera systems on a trip to the Antarctic. I have trouble enough with just one and a few lenses! :)</p>

[ellis_vener_photography]

<blockquote>

<p>As an aside, I could not imagine trying to juggle 3 different camera systems on a trip to the Antarctic. I have trouble enough with just one and a few lenses! :)</p>

 

</blockquote>

<p>That should be interesting as on the last previous Luminous Landscape trip to Antartica, on a rainy day in the Falklands, several photographers reported lots of pro level Canon D-SLR cameras shutting down (almost all came back to life) and no Nikon D-SLRs shutting down.</p>

[py-photography]

<p>Interesting read...<br>

But it looked like some of the test shots were OOF.... Could be just me, the holiday season can cause blurry vision... lolo</p>

<p> </p>

[bill_tuthill]

<p>Huh? The D700 goes up to ISO 25600 but the D3 only goes to ISO 6400?  I'm confused.<br>

Now if the industry could just make lenses good enough for their camera bodies.  Do I see chromatic aberration in the center of the frame? Or maybe somebody spilled nail polish on the Macbeth chart.</p>

[photos of hans koot]

<p>The canon seems to have the best value for money here, just keep it dry perhaps? D3x shots (low iso) seem a tad out of focus indeed.<br>

I just ordered an epson a2 printer, so if they sent me the camera's I'll perform a check here with real prints, no need to go to the arctics for that....</p>

[Matt Laur]

<p>Bill: the Nikon camera in question is the D3x, not the D3. The D3 has the crazy-high ISO option if you want to use it, but the reviewer was comparing the newer (higher-resolution) body. He was comparing at 6400 because that was a reasonable, practical limit to what any of the bodies at that resolution could quietly produce.<br /><br />Hans: he does mention that wasn't being very careful with the focus and camera movement on some of the shots, which could have something to do (as he even surmises) with a few of those OoF shots.</p>

[photos of hans koot]

<p>Matt, correct, thank you for pointing that out, and this will not affect the noise indeed.</p>

<p>There has just been posted another interesting article on the 50d sharpnes issue.</p>

[sarah_fox]

<p>Very impressive performance from both Nikon and Canon! I think the Canon looks a bit cleaner, but I'm also a Canon photographer. As an aside, the Nikon lens seems to be the least of the three -- somewhat soft (and even a slight double edge atop the Macbeth chart?), with a bit of CA (e.g. around the toy photographer's arms and lens). The Sony showed a bit of CA too, but was still much sharper than the Nikon. I don't know the Nikkor line well enough to know whether this lens is a strong offering. Could have been a defective copy. I'm just nit-picking, of course. ;-)</p>

[bill_tuthill]

The <i>op cit</i> L-L article about the 50D is by Harold M. Merklinger! <a href="http://www.luminous-landscape.com/reviews/cameras/50d.shtml">Definitely worth reading</a>.

<p>

Sarah, your comments in another thread were the most interesting I've ever read about LiveView, thanks.

[photos of hans koot]

<p>Sarah, to me it also looks like the Canon has the edge. How much this will appear in normal prints I am not sure. As pointed out before by Matt, focussing was not the strongest point in the test :-). From this test we cannot make up if the nikon lens is not ok. In the background you can see the viewpoint has been different for all three. For noise comparison it will make no difference.<br>

Bill, it indeed is very interesting.</p>

[sarah_fox]

<p>"Sarah, your comments in another thread were the most interesting I've ever read about LiveView, thanks."</p>

<p>[chuckling] You're quite welcome, but I think it must have been someone else who said something enlightening or profound about LiveView. ;-)</p>

<p>Merklinger's always a good read. Thanks for posting the link!</p>

<p>Happy New Year!</p>

[ilkka]

<p>"Huh? The D700 goes up to ISO 25600 but the D3 only goes to ISO 6400? I'm confused."</p>

<p>D3x has twice the pixel count. More pixels mean smaller pixels and smaller pixels mean more noise. I suppose Nikon has enough sense not to put a nonusable ISO speed in their top of the line camera.</p>

[bill_tuthill]

<p>Thanks, I have difficulty noticing the X after Nikon product numbers.<br>

It is hard for me to believe the Canon 24-70/2.8 is as good as the Nikon 24-70/2.8, if properly focused. Traditionally Nikon has engineered better wide-angle lenses than Canon, although Canon (like Sigma) did better with telephoto lenses. Although I know they are not supposed to be compared across DSLR mounts, to heck with it. Here are the MTF numbers from photozone.de:</p>

<dl>

<table border="1">

<tbody>

<tr>

<td> </td>

<td> </td>

<td>wide<br /> open</td>

<td>stopped<br /> down</td>

</tr>

<tr>

<td>Canon 24-70/2.8</td>

<td>24mm</td>

<td>1840</td>

<td>2040</td>

</tr>

<tr>

<td> </td>

<td>40mm</td>

<td>1885</td>

<td>2050</td>

</tr>

<tr>

<td> </td>

<td>70mm</td>

<td>1800</td>

<td>1975</td>

</tr>

<tr>

<td>Nikon 24-70/2.8</td>

<td>24mm</td>

<td>2225</td>

<td>2320</td>

</tr>

<tr>

<td> </td>

<td>40mm</td>

<td>2270</td>

<td>2280</td>

</tr>

<tr>

<td> </td>

<td>24mm</td>

<td>2080</td>

<td>2210</td>

</tr>

</tbody>

</table>

</dl>

[sarah_fox]

<p>Interesting comparisons. I checked out the data on slrgear.com, and indeed the Nikkor is sharper. However its CA is high -- REALLY high. In fact at 35mm the max CA level goes off their scale, which tops out at approx 4 pixels on the D3 (15/100 of a % of frame height). So it seems to be another apples and oranges thing -- sharpness or CA. No clear winners. BTW, I've heard Nikon's AI filters are a bit more aggressive. That might account for some of the sharpness differences.</p>

[photos of hans koot]

<p>ca can be adjusted in Raw, at least quite much, but when you have to deal with many shots it gives a lot of work.</p>

[photos of hans koot]

<p>In this article, a comparison of the 5d mKII and the A900 a part is about the sharpness issue. Makes some more clear I think. </p>

<p>http://www.luminous-landscape.com/reviews/cameras/a900-5dmkii.shtml</p>

[ncarrasco]

<p>Interesting discussion! For the novice in photography, what does CA mean?<br>

Thank you.</p>

<p>Nicolas</p>

[bill_tuthill]

<p>CA = chromatic aberration.<br>

According to photozone.de testing, the Canon 24-70/2.8 has much (much!) worse CA at 24mm, but the Nikon is worse at 70mm, especially wide open.<br>

http://www.photozone.de/canon-eos/184-canon-ef-24-70mm-f28-usm-l-test-report--review?start=1<br>

http://www.photozone.de/nikon--nikkor-aps-c-lens-tests/236-nikkor-af-s-24-70mm-f28g-ed-review--test-report?start=1</p>

[aslavihervuori]

<p>In Photozone, Canon lenses are tested with 8mp body, Nikons with 10mp body. Not big difference perhaps, but still clearly visible in number of lenses. Even 17-50 Tamron gives different results with Nik/Can bodies (Just like 18-55IS gives different readings between 8mp and 15mp bodies) a s l a</p>