Small is Beautiful! Yes, But Also for Pixels of Digital Still Cameras?

In the rat race forever higher pixel counts for the digital still market, it is clear that if the chip size has to remain constant, then “more pixels” inherently means “smaller pixels”. The main driving force to keep the silicon area of the sensor as small as possible is the cost price of the sensor chip. Extra silicon costs quite a bit of money. A larger chip size also requires a larger and more costly lens, extra PCB area, and also a larger volume for the camera body. So, smaller pixels are the key words towards higher resolution sensors on the same silicon area, or towards a smaller chip size with the same resolution! Unfortunately someone has to pay the bill for these improvements, and in this case it is the overall performance of the pixels. Irrespective of the sensor technology used, be it interline-transfer CCD (IL), frame-transfer CCD (FT), full-frame CCD (FF) or a CMOS imager with passive (PPS) or active pixels (APS), the performance of smaller pixels will deteriorate. Parameters such as light sensitivity, quantum efficiency, saturation level, dynamic range, signal­ to-noise ratio and pixel non-uniformity will suffer from the trend towards ever smaller pixels. Very often a pixel of a solid-state imager with its electron packet is compared with a bucket containing water. This analogy helps very much not only in understanding the working principle of the pixels, but also in this discussion to explain the influence of smaller pixels on imaging performance.