| || |
Re: Official: post your DNA photos here
All things being equal, noise generally indicates a worse sensor - usually because it is a smaller sensor or a sensor otherwise making use of smaller photodiodes. Optics might have nothing to do with it at all, you can get noise with no light coming into the sensor at all. Once there is actually light coming into the sensor better optics reduce the noise, usually, because more light will result in a better signal to noise (of the dark current type) ratio.
Originally Posted by JHBThree
Depending on what you mean by "better sensor" though, you can be correct if "better sensor" means higher sensor density. An 8 megapixel and 5 megapixel sensor of the same physical size, again all else being equal, will have less noise with the 5 megapixel sensor because of the larger diodes on the 5 megapixel sensor (and thereby once again a better SNR given an equal amount if incident light per unit area). Usually though all things are not equal, noise reduction is applied on nearly everything so unless you are looking at the raw file - prior to debayering even - you're not likely to actually be seeing the true noise of the sensor itself. Also you can mitigate the higher noise by multisampling, a 20 megapixel sensor of the same size as a 5 megapixel sensor will probably show similar noise as the 5 megapixel sensor if that 20 megapixel sensor is multi-sampled down to 5 megapixels, which is how I suspect Nokia gets such good noise performance out of their real PureView sensors.