>The 12.2 stops quoted by DxO is measured at base iso??? but >the cambridge colour site states: > >Quote > >Technical Note: In some digital cameras, there is an extended >low ISO setting which produces less noise, but also decreases >dynamic range. > >Unquote > >Why would DxO test at base iso if this is true? The chart at >DXO doesn't state it is tested at base iso and I think that >testing at base iso wouldn't give a true test result. >
This is quite a complex subject that unless you are really interested can be difficult to comprehend(I know it makes my eyes glaze over an turns my brain to mush when I start to read about this stuff!!! ) My limited understanding is that it mainly comes down to full well capacity and read noise. As to the low ISO figures, if you check Dxomark’s measured ISO against the manufacture reported ISO you will discover that the measured ISO of the low setting is the same as the base ISO figure. For instance the D700 has a base ISO of 200 which Dxomark measures at 162, and the measured ISO of the Lo ISO equivalent of the manufacturer reported 100 is also measured at 162 (the Lo setting is achieved through data manipulation not sensor gain). And when you look at the dynamic range chart they do not include the Lo setting.
There also appears to be no one standard when doing these evaluations and depending on your tolerance to noise it can alter the numbers. I believe Dxomark gives technical engineering limitations where as others give what they determine as practical every day use numbers(less tolerance to noise in the shadows).
For a good read on sensor performance you may wish to check out Dr Clark’s article.