QUOTE (BernardLanguillier @ Oct 20 2009, 10:58 AM)

Good news. It is fascinating to witness how technology manages every time to outdo the predictions of some of our forum physics experts.

Well, not in terms of pixels and resolution, which was I believe the subject of the last discussion. As far as low light sensitivity is concerned, this is mostly a QE efficiency issue (raw, and effective) and there is still a lot of room for improvement. DR is a mix of many factors with QE and read noise being major factors. Relatively cheap amateur astronomy CCD cameras still beat the DSLRs by a wide margin, at the cost of ease of use of course. Also, while there are incremental improvements on all the component of the chain (micro lenses geometry, transparency, read noise, etc...) there is also a huge amount of in-camera processing that has essentially moved a lot of the post processing work to the pre-processing stage. The same in-camera processing that give (or gave until now) Nikon cameras the advantage in low light makes them less than optimal for astrophotography, where maximum actual SNR is sought through optimal post-processing. Regardless, this if of course great news for photographers.
One interesting bit of data I couldn't find is the well capacity in e- of these sensors. If it is in the 16000 e- range, no one is ever going to get more that 14 bits of DR out of them.