DXO Labs Camera Comparison System

[Jason Dunn]

<div class='os_post_top_link'><a href='http://www.dxomark.com/index.php/eng/Image-Quality-Database/Compare-cameras/(appareil1)/340|0/(appareil2)/320|0/(appareil3)/295|0/(onglet)/0/(brand)/Panasonic/(brand2)/Nikon/(brand3)/Nikon' target='_blank'>http://www.dxomark.com/index.php/en.../(brand3)/Nikon</a><br /><br /></div><p><img src="http://images.thoughtsmedia.com//dht/auto/1267747464.usr1.png" style="border: 1px solid #d2d2bb;" /></p><p>DXO Labs is known for some of the most scientific camera tests in the world; their software is used by many top magazines and Web sites for analyzing camera and lens performance. I stumbled across their comparison page last week and plugged in my three main camera bodies: a Panasonic GF1, Nikon D5000, and Nikon D300. <a href="http://www.dxomark.com/index.php/eng/Image-Quality-Database/Compare-cameras/(appareil1)/340|0/(appareil2)/320|0/(appareil3)/295|0/(onglet)/0/(brand)/Panasonic/(brand2)/Nikon/(brand3)/Nikon" target="_blank">The results were pretty interesting</a> - I wasn't expecting the D5000 to rate higher than the much more expensive D300 in the areas of low-light performance and the overall sensor score. I don't know enough about the data points they're using to generate these scores...what I do know is that I find my D300 infinitely nicer to shoot with than my D5000. Not that there's anything wrong with the D5000 of course, but the D300 is a much more satisfying overall camera experience. Plug in your DSLR (or Micro Four-Thirds) camera and see what DXO Labs has to say about it!</p>

[touristguy87]

that the reasons that you prefer to shoot your D300 over your D5000 have nothing to do with SNR, DR, or CS vs ISO.

this becomes obvious when you examine the left side of the DxOMark curves vs the right side LOL

The low-light metric makes a few assumptions which are not valid for all users. It would probably be more valuable if it were a) logarithmic, at least in stops and b) made relative to scene exposure. Also I don't like the limited options in terms of viewing the curves based on 100% or in 8MP prints at 300dpi. It ought to include VGA, 1 & 2MP options for web-based viewing.

The camera performs best at the lowest ISO, in most cases. So regardless of scene exposure, given sufficient speed to capture the scene well (in case of moving objects in the scene), photographers will tend to use the lowest ISOs. Unless you can detect the difference between 24bit CS and 20, that won't really matter. What will matter more are the feature-set and usability of the camera coupled with the lens and its performance. The low-light mark really only matters if you're trying to get good handheld shots in low light, and the standards will vary from user to user.

I refer to DxOMark on a regular basis but the most-important part of it for me is the performance near ISO100 and obviously that is of limited importance in the overall scheme of things.