Unless the sensor has a DR > 72dB, the 2 extra bits aren't really going to have much of an impact. If the DR is less than 72dB, then all the extra bits are doing is specifying the noise more precisely.
I'd be interested to hear about these cases where the extra bits contributed something when the signal itself did not have the dynamic range. Was the imaging chain exactly the same pre-ADC? Could the effect you were seeing be explainable by dithering (analogous to the effect in the audio domain), rather than captured extra "detail" in the extra bits? If this is the case, then you could simulate the effect with 12bit/pixel captures and good noise shaping.
Sony probably won't announce this week, possibly beginning of September. It's also likely to be the Advanced Amateur model, not the flagship.
The IMX021 is an APS-C sized sensor, should work perfectly well with the DX lenses AFAIK.
Why do you say this? Why should Nikon (hypothetically) bringing a 36mm x 24mm sensor camera to the table drive Nikon users to Canon? There'll still be APS-C sized sensor cameras (D80-D200 class) for people with an attachment to DX lenses I would suppose.
Fair competition is good. I hope We see some innovative cameras based upon the IMX021 to spur Canon on, and vice versa.
If Sony is doing the 12.4 mp sensor on the advanced amateur, I can't wait to see what they have up their sleeves for the flagship. Poor Canon.
For a single pixel, true. However, images are collections of pixels and many gradual tone gradients span very many pixels. With each pixel quantized to 14 bits, the average across a cluster of pixels can settle to a value consistent with 14 bits. Plus, it's a bit easier to manipulate the data during postprocessing (esp. steps that stretch the tonal range) without adding gaps.
I wonder what Pentax will do with it if they decide to use it. It's all starting to get very interesting.
--
Note that we don't know the maximum frame rate of Canon's 40D sensor, only the frame rate of the entire system (which could be bottlenecked by the strength of the mirror and shutter mechanisms, data bandwidth, raw-> jpeg processing speed in the DIGIC III, etc.). If Sony builds a DSLR with this new sensor, the beefed up mechanicals and high bandwidth wouldn't be cheap.
Exactly. It should be clearer with some real type numbers and quantization noise.
Has the Pentax shown any benefit from a 22bit ADC? None that I see. This seems like marketing more than anything else, or a move toward consistency. Perhaps the big pixels on the 1Dmk3 might squeeze out enough S/N that it needs 14bit, but I seriously doubt the 40D does.
