At the risk of having my head explode again......
On a different thread the point was made that the marketing guys won the debate when moving from 12 bit depth to 14.
Does the new sensor provide better information i.e. less noise in the last couple of bits and therefore better IQ?
Does the increased ISO sensitivity translate into greater dynamic range at any given ISO?
Mike
If you see me with a wrench, call 911
I think they claim two things. Canon's propaganda says the new sensor technology gathers more of the light. It also says the new processors are 17x faster than the old models, and they are now able to do more with the files in camera. So it is probably a combination, some improvement with the sensors and because of faster processors they are able upgrade the firm ware significantly. If you remember the early days of the PC it was always a race for new software to use up memory and speed developments of processors. This fueled the technology boom in the early 90's because the latest and greatest was always obsolete within a year. To use new software you were continually upgrading. I think to a lesser extent were probably seeing some of that here. More processing power equals better firmware equals better processed pictures.Originally Posted by Busted Knuckles
They gave us in camera HDR, so sure it increased dynamic range if you use that function. The 1D X is supposed to have an increased dynamic range.
We wont know until we can get our hands on raw files from production cameras. But given that Canon has made no improvement in this area in the last seven years, I'm not holding my breath.
Not necessarily, but it's possible. It depends on whether they improve only the pixel, or if they also improve the downstream electronics (e.g. ADC).
The increased ISO sensitivity suggests one of three possibilities to me.
The first is that the level of noise at the sensor, or elsewhere in the rest of the hardware involved in reading out the pixel charges, has been reduced which is what I'm hoping for myself. Solving noise issues is a better option than improving filters to deal with it.
The second is that the conversion to JPEG in the camera has been improved with the extra processing power. So far, the only sample photos I've seen to demonstrate the increased ISO ability have been JPEGs. Call me a skeptic if you wish.
The third is that there has been a filter improvement in Canon's RAW conversion which is implemented in the JPEG processing of the new bodies. In time, this will be released through updates to DPP and all Canon cameras will suddenly improve by two stops before the marketing department knows what hit them. It is a good dream anyways.
It will likely turn out to be some combination of the above.
The dynamic range of the camera is a limitation of the pixels. Once a certain number of photons have been received into a pixel and converted to charge, the pixel will be full and unable to convert more photons. This is the upper limit of the range. On the bottom end, a pixel that receives no photons should have zero charge, at least in theory. In practice there is a small charge on this pixel. This is a limitation of the device and it is not a constant for every pixel across the entire sensor. It also varies with temperature which is why a black level calibration can be conducted on items such as high-speed cameras and other scientific equipment. This is where the fine pixel level noise in shadows originates. It is the low level electronic noise that gets amplified alongside of the information that you want. The ratio of signal to noise is abbreviated to SNR in the electronics and signal processing worlds.
Hope this helps.
Posting Permissions
Reply With Quote