dynamic range 1dII, 5D, 1dmkII, 20D

Krotki test specjalisty:
http://www.naturescapes.net/phpBB2/v...ic.php?t=51068

Cytat:

Zamieszczone przez iczek

Krotki test specjalisty:
http://www.naturescapes.net/phpBB2/v...ic.php?t=51068

tak, tylko podaj login i pass

ctrl+c, ctrl+v

Cytat:

Zamieszczone przez E.J. Peiker, Senior Technical Editor

The next in my round of tests of the current stable of Canon DSLR's was that of dynamic range. Often one sees dynamic range tested on a 18% gray card and dynamic range is expressed as the range between rendering a totally white 255,255,255 image and a totally black 0, 0, 0 image. However there is literature that this is a faulty technique and I concur with that because such a test does not test the ability to resolve different tonal values. I have therefore constructed my own dynamic range target. My target includes bands of 120,120,120 and 136,136,136 on the otherwise 18% gray card (128,128,128). I then take shots at 1/3 stop intervals and determine where the camera is no longer able to resolve the difference on the bright and dark side. I feel that this gives us a much more real world photography dynamic range since we actually want to resolve highlight and shadow detail. Here are the results:

20D
-3.0 to +3.33 for a range of 6.33

1D MkIIn
-2.67 to +3.67 for a range of 6.33

5D
-3.33 to +4.33 for a range of 7.67

1Ds Mk II
-3.0 to +3.67 for a range of 6.67

The data from this test reveals that the EOS 5D is hands down the dynamic range champ with the EOS 1Ds Mk II coming in second with a full stop less. Despite having smaller pixel sites than the 1D Mark IIn, the 1Ds II slightly outperforms it. Surprisingly the tiny pixel 20D and the large pixel 1D Mark IIn are equal at 6.33 stops. This data does not track with pixel size which many believe. Clearly there are other factors at play.
EOS 5D.

This test was conducted at ISO 200 and could vary somewhat at different ISOs.

BTW you can add between 2 and three stops of range to this if you just do the solid white to solid black test but realize you will not record any bright or dark detail when in those regimes. That's why I think my method above is a more real world usability test.

Ciekawe jak wypadłby w tym teście 350D.