Originally posted by Busiko Acho que tem de ler novamente...
Rui, sem ofensa, mas tu está sendo um pouco cabeça dura. A parte sobre exposições longas é só uma descrição de como o sensor CMOS funciona.
I read the paper again, and the two parts where it mentions night photography are identical. As far as I know (not much), CMOS sensors
already have a silicon nitride or silicon dioxide layer beneath the photodiodes to radiate the heat. The part you mention, which I will copy
one more time:
Since the solid-state image pickup element is bonded to the substrate formed of silicon nitride having a favorable thermal conduction property (having a thermal conductivity of 90 W/mK), heat generated in the solid-state image pickup element can be radiated to the outside via the substrate quickly. When the solid-state image pickup device is used in a camera, therefore, exposure to light for a long time becomes possible. For example, it is also possible to take an image of a night view beautifully by conducting exposure to light for a long time.
looks like a description of the way the sensor works, right now, the present ones. Long exposures ARE a possibility in the current sensors, that is not a new feature. Of course improvements are due, I see you're excited at the possibility of longer exposures with less noise/hot pixels, but this document is not about that...
Yet this sensor warping to match the image field is very interesting, thanks for bringing that to light! I hope this becomes available in the near future - that would knock down most of the 4/3s sensor advantages.